Volume 25 Issue 6 by Western Journal of Emergency Medicine

Volume 25, Number 6, November 2024

Journal of

Behavioral Health

858 Recent Interventions for Acute Suicidality Delivered in the Emergency Department: A Scoping Review

Alex P. Hood, Lauren M. Tibbits, Juan I. Laporta, Jennifer Carrillo, Lacee R. Adams, Stacey Young-McCaughan, Alan L. Peterson, Robert A. De Lorenzo

869 Buprenorphine-Naloxone for Opioid Use Disorder: Reduction in Mortality and Increased Remission

Krishna K. Paul, Christian G. Frey, Stanley Troung, Laura vita Q. Paglicawan, Kathryn A. Cunningham, T. Preston Hill, Lauren G. Bothwell, Georgiy Golovko, Yeoshina Pillay, Dietrich Jehle

875 Opioid Treatment Is Associated with Recurrent Healthcare Visits, Increased Side Effects, and Pain

Caroline E. Freiermuth, Jenny A. Foster, Pratik Manandhar, Evangeline Arulraja, Alaattin Erkanli, Charles V. Pollack, Stephanie A. Eucker

883 Impact of COVID-19 Pandemic on Emergency Department Visits for Opioid Use Disorder Across University of California Health Centers

Matthew Heshmatipour, Ding Quan Ng, Emily Yi-Wen Truong, Jianwei Zheng, Alexandre Chan, Yun Wang

Clinical Practice

890 An Assessment of the Presence of Clostridium tetani in the Soil and on Other Surfaces

Michael Shalaby, Alessandro Catenazzi, Melissa F. Smith, Robert A. Farrow II, David Farcy, Oren Mechanic, Tony Zitek

Critical Care

894 External Validation of the RESCUE-IHCA Score as a Predictor for In-Hospital Cardiac Arrest Patients Receiving Extracorporeal Cardiopulmonary Resuscitation

Yi-Ju Ho, Pei-I Su, Chien-Yu Chi, Min-Shan Tsai, Yih-Sharng Chen, Chien-Hua Huang

Contents continued on page iii

Western Journal of Emergency Medicine:

Integrating Emergency Care with Population Health

Andrew W. Phillips, MD, Associate Editor DHR Health-Edinburg, Texas

Edward Michelson, MD, Associate Editor Texas Tech University- El Paso, Texas

Dan Mayer, MD, Associate Editor Retired from Albany Medical College- Niskayuna, New York

Wendy Macias-Konstantopoulos, MD, MPH, Associate Editor Massachusetts General Hospital- Boston, Massachusetts

Gayle Galletta, MD, Associate Editor University of Massachusetts Medical SchoolWorcester, Massachusetts

Yanina Purim-Shem-Tov, MD, MS, Associate Editor Rush University Medical Center-Chicago, Illinois

Section Editors

Behavioral Emergencies

Leslie Zun, MD, MBA

Rosalind Franklin University of Medicine and Science

Marc L. Martel, MD

Hennepin County Medical Center

Cardiac Care

Fred A. Severyn, MD University of Colorado School of Medicine

Sam S. Torbati, MD

Cedars-Sinai Medical Center

Clinical Practice

Cortlyn W. Brown, MD Carolinas Medical Center

Casey Clements, MD, PhD Mayo Clinic

Patrick Meloy, MD

Emory University

Nicholas Pettit, DO, PhD Indiana University

David Thompson, MD University of California, San Francisco

Kenneth S. Whitlow, DO Kaweah Delta Medical Center

Critical Care

Christopher “Kit” Tainter, MD University of California, San Diego

Gabriel Wardi, MD University of California, San Diego

Joseph Shiber, MD University of Florida-College of Medicine

Matt Prekker MD, MPH Hennepin County Medical Center

David Page, MD University of Alabama

Erik Melnychuk, MD

Geisinger Health

Quincy Tran, MD, PhD University of Maryland

Disaster Medicine

John Broach, MD, MPH, MBA, FACEP University of Massachusetts Medical School

UMass Memorial Medical Center

Christopher Kang, MD Madigan Army Medical Center

Education

Danya Khoujah, MBBS University of Maryland School of Medicine University of Colorado

Michael Epter, DO

Mark I. Langdorf, MD, MHPE, Editor-in-Chief University of California, Irvine School of MedicineIrvine, California

University of California, Irvine School of MedicineIrvine, California

Michael Gottlieb, MD, Associate Editor Rush Medical Center-Chicago, Illinois

Niels K. Rathlev, MD, Associate Editor Tufts University School of Medicine-Boston, Massachusetts

Rick A. McPheeters, DO, Associate Editor

R. Gentry Wilkerson, MD, Associate Editor University of Maryland

Maricopa Medical Center

ED Administration, Quality, Safety

Tehreem Rehman, MD, MPH, MBA Mount Sinai Hospital

David C. Lee, MD Northshore University Hospital

Gary Johnson, MD Upstate Medical University

Brian J. Yun, MD, MBA, MPH Harvard Medical School

Laura Walker, MD Mayo Clinic

León D. Sánchez, MD, MPH

Beth Israel Deaconess Medical Center

William Fernandez, MD, MPH University of Texas Health-San Antonio

Robert Derlet, MD

Founding Editor, California Journal of Emergency Medicine University of California, Davis

Emergency Medical Services

Daniel Joseph, MD

Yale University

Joshua B. Gaither, MD University of Arizona, Tuscon

Julian Mapp

University of Texas, San Antonio

Shira A. Schlesinger, MD, MPH Harbor-UCLA Medical Center

Geriatrics

Cameron Gettel, MD Yale School of Medicine

Stephen Meldon, MD

Cleveland Clinic

Luna Ragsdale, MD, MPH Duke University

Health Equity

Emily C. Manchanda, MD, MPH

Boston University School of Medicine

Faith Quenzer

Temecula Valley Hospital San Ysidro Health Center

Payal Modi, MD MScPH University of Massachusetts Medical

Infectious Disease

Elissa Schechter-Perkins, MD, MPH

Boston University School of Medicine

Ioannis Koutroulis, MD, MBA, PhD

George Washington University School of Medicine and Health Sciences

Stephen Liang, MD, MPHS

Shadi Lahham, MD, MS, Deputy Editor Kaiser Permanente- Irvine, California

Susan R. Wilcox, MD, Associate Editor Massachusetts General Hospital- Boston, Massachusetts

Elizabeth Burner, MD, MPH, Associate Editor University of Southern California- Los Angeles, California

Patrick Joseph Maher, MD, MS, Associate Editor Ichan School of Medicine at Mount Sinai- New York, New York

Donna Mendez, MD, EdD, Associate Editor University of Texas-Houston/McGovern Medical School- Houston Texas

Danya Khoujah, MBBS, Associate Editor University of Maryland School of Medicine- Baltimore, Maryland

Washington University School of Medicine

Victor Cisneros, MD, MPH Eisenhower Medical Center

Injury Prevention

Mark Faul, PhD, MA Centers for Disease Control and Prevention

Wirachin Hoonpongsimanont, MD, MSBATS

Eisenhower Medical Center

International Medicine

Heather A.. Brown, MD, MPH Prisma Health Richland

Taylor Burkholder, MD, MPH Keck School of Medicine of USC

Christopher Greene, MD, MPH University of Alabama

Chris Mills, MD, MPH Santa Clara Valley Medical Center

Shada Rouhani, MD

Brigham and Women’s Hospital

Legal Medicine

Indiana University School of Medicine

Statistics and Methodology

Shu B. Chan MD, MS Resurrection Medical Center

Stormy M. Morales Monks, PhD, MPH

Texas Tech Health Science University

Soheil Saadat, MD, MPH, PhD University of California, Irvine

James A. Meltzer, MD, MS Albert Einstein College of Medicine

Musculoskeletal

Juan F. Acosta DO, MS

Rick Lucarelli, MD Medical City Dallas Hospital

William D. Whetstone, MD University of California, San Francisco

Neurosciences

Antonio Siniscalchi, MD Annunziata Hospital, Cosenza, Italy

Pediatric Emergency Medicine

Paul Walsh, MD, MSc University of California, Davis

Muhammad Waseem, MD Lincoln Medical & Mental Health Center

Cristina M. Zeretzke-Bien, MD University of Florida

Public Health

Henry Ford Hospital

John Ashurst, DO Lehigh Valley Health Network

Tony Zitek, MD Kendall Regional Medical Center

Trevor Mills, MD, MPH Northern California VA Health Care

Erik S. Anderson, MD Alameda Health System-Highland Hospital

Technology in Emergency Medicine

Nikhil Goyal, MD Henry Ford Hospital

Phillips Perera, MD Stanford University Medical Center

Trauma

Pierre Borczuk, MD

Massachusetts General Hospital/Havard Medical School

Toxicology

Brandon Wills, DO, MS Virginia Commonwealth University

University of California, Irvine

Ultrasound J. Matthew Fields, MD

Shane Summers, MD Brooke Army Medical Center

Robert R. Ehrman Wayne State University

Ryan C. Gibbons, MD Temple Health

Journal of the California Chapter of the American College of Emergency Physicians, the America College of Osteopathic Emergency Physicians, and the California Chapter of the American Academy of Emergency Medicine

American Academy of Emergency Medicine

Western Journal of Emergency Medicine:

Western Journal of Emergency :

Integrating Emergency with Population Health

Integrating Emergency Care with Population Health

Indexed in MEDLINE, PubMed, and Clarivate Web of Science, Science Citation Index Expanded

Amin A. Kazzi, MD, MAAEM

Gayle Galleta, MD

The American University of Beirut, Lebanon

Beirut,

The American University of Beirut, Beirut, Lebanon

Brent King, MD, MMM University of Texas, Houston

Brent King, MD, MMM University Texas, Houston

Christopher E. San Miguel, Ohio State University Wexner Medical Center

Christopher E. San Miguel, MD Ohio State University Wexner Medical Center

Daniel J. Dire, MD

Daniel J. Dire, MD University of Texas Health Sciences Center San Antonio

Daniel J. Dire, MD University Texas Health Sciences Center San Antonio

Douglas Ander, Emory University

Emory University

Douglas Ander, MD Emory University

Edward Michelson, MD Texas Tech University

Edward Michelson, Texas Tech University

Edward Michelson, MD Texas Tech University

Edward Panacek, MD, MPH South

Edward MD, MPH University South Alabama

Edward Panacek, MD, MPH University of South Alabama

Francesco

“Maggiore della Carità,” Novara, Italy

Francesco Della Corte, MD Azienda Ospedaliera Universitaria “Maggiore della Novara, Italy

Francesco Della Corte, MD Azienda Ospedaliera Universitaria “Maggiore della Carità,” Novara, Italy

Sørlandet Sykehus HF, Akershus Universitetssykehus, Lorenskog,

Hjalti Björnsson, MD

Editorial Board

Board

Editorial Board

Sørlandet Sykehus HF, Akershus Universitetssykehus, Lorenskog, Norway

Hjalti Björnsson, MD Icelandic Society of Emergency Medicine

Hjalti MD Icelandic Society of Emergency Medicine

Jaqueline Le, MD Desert Regional Medical Center

Jaqueline Le, MD Desert Medical Center

Regional

Jeffrey Love, MD

Jeffrey Love, MD The George Washington University School of Medicine and Health Sciences

Jeffrey Love, The George Washington University School of Medicine and Health Sciences

Katsuhiro Kanemaru, MD University of Hospital, Miyazaki, Japan

Katsuhiro Kanemaru, MD University of Miyazaki Hospital, Miyazaki, Japan

Kenneth V. Iserson, MD, MBA University of Arizona, Tucson

Leslie Zun, MD, MBA Chicago Medical School

Linda S. Murphy, MLIS University of California, Irvine School of Medicine

The George Washington University School of Medicine and Health Sciences Arizona, Chicago Medical School Librarian

Linda S. Murphy, MLIS University of California, Irvine School of Medicine Librarian

Advisory Board

Elena Lopez-Gusman, JD

California ACEP

Elena Lopez-Gusman, JD California ACEP

California ACEP

Mark I. Langdorf, MD, MHPE, MAAEM, FACEP

Langdorf, MAAEM, FACEP

Tufts University School of Medicine

Niels K. Rathlev, Tufts University School of Medicine

Niels K. Rathlev, MD Tufts University School of Medicine

Scott Zeller, MD

Scott Zeller, MD University of California, Riverside

Pablo Aguilera Fuenzalida, MD Pontificia Universidad Catolica de Bell,

Pablo Aguilera Fuenzalida, MD Pontificia Universidad Catolica de Chile, Región Metropolitana, Chile

Pablo Aguilera Fuenzalida, Pontificia Universidad Catolica de Chile, Región Metropolitana, Chile

Peter A. Bell, DO, MBA Baptist Health Sciences University

Peter A. Bell, DO, MBA Baptist Health University

Peter Sokolove, MD University of California, Francisco

Peter Sokolove, MD University of California, San Francisco

University of California, San Francisco

Rachel A. Lindor, MD, JD

Rachel A. Lindor, MD, JD Mayo Clinic

Robert Suter, DO, MHA

Robert Suter, DO, UT Southwestern Medical Center

Robert Suter, DO, MHA UT Southwestern Medical Center

Robert W. Derlet, MD University of California, Davis

University of California, Davis

Rosidah Ibrahim, MD Hospital Serdang, Selangor, Malaysia

Rosidah Ibrahim, MD Hospital Serdang, Malaysia

Rosidah Ibrahim, MD Hospital Serdang, Selangor, Malaysia

Scott Rudkin, MD, MBA

Scott Rudkin, MD, University of California, Irvine

Scott Rudkin, MD, MBA University of California, Irvine

Steven H. Lim, MD Changi General Hospital, Simei, Singapore

Singapore

Steven H. Lim, Changi General Hospital, Simei, Singapore

Terry Mulligan, DO, MPH, FIFEM ACEP Ambassador to the Netherlands Society of Emergency Physicians

Terry Mulligan, DO, MPH, FIFEM ACEP Ambassador to the Netherlands

Terry Mulligan, DO, MPH, FIFEM ACEP to the Netherlands Society of Emergency Physicians

Wirachin Hoonpongsimanont, MD, MSBATS

Siriraj Hospital, Mahidol University, Bangkok, Thailand

Editorial Staff

Staff Advisory Board

Isabelle BS Executive Editorial Director

Isabelle Nepomuceno, BS Executive Editorial Director

American of Emergency Physicians

American College of Emergency

American College of Emergency Physicians

Jennifer Kanapicki Comer, MD FAAEM

California Chapter Division of AAEM Stanford University School of Medicine

DeAnna McNett, CAE

Kimberly Ang, MBA

DeAnna McNett, CAE

American College of Osteopathic Emergency Physicians

UC Irvine Health School of Medicine

Randall J. Young, MD, MMM, FACEP California ACEP

Kimberly Ang, MBA

UC Irvine Health School of Medicine

American College of Emergency Physicians Kaiser Permanente

J. American College of Emergency Physicians Kaiser Permanente

Randall J. Young, MD, MMM, FACEP California ACEP American College of Emergency Physicians Kaiser Permanente

UC Irvine Health School of Medicine

Mark I. Langdorf, MD, MHPE, MAAEM, FACEP

UC Irvine Health School of Medicine

Robert Suter, DO, MHA American College of Osteopathic

Robert Suter, DO, MHA American College of Osteopathic Emergency Physicians UT Southwestern Medical Center

Shahram Lotfipour, MD, MPH FAAEM, FACEP

UC Irvine Health School of Medicine

Shahram Lotfipour, MD, MPH FAAEM, FACEP UC Irvine Health School of Medicine

Jorge Fernandez, MD, FACEP

UC San Diego Health School of Medicine

Visha Bajaria, BS WestJEM Editorial Director

Emily Kane, MA WestJEM Editorial Director

Stephanie Burmeister, MLIS JEM Liaison

Visha Bajaria, BS WestJEM Editorial Director WestJEM

Stephanie Burmeister, MLIS WestJEM Staff Liaison

Cassandra Saucedo, MS Executive Publishing Director

Nicole Valenzi, BA WestJEM Publishing Director

Cassandra Saucedo, MS WestJEM Publishing Director

Nicole Valenzi, WestJEM Publishing Director

June Casey, BA Copy Editor

Official Journal of the California Chapter of the American College of Emergency Physicians, the America College of Osteopathic Emergency Physicians, and the California Chapter of the American Academy of Emergency Medicine

Available in MEDLINE, PubMed, PubMed Central, Europe PubMed Central, PubMed Central Canada, CINAHL, SCOPUS, Google Scholar, eScholarship, Melvyl, DOAJ, EBSCO, EMBASE, Medscape, HINARI, and MDLinx Emergency Med. Members of OASPA. Editorial and Publishing Office: WestJEM/Depatment of Emergency Medicine, UC Irvine Health, 3800 W. Chapman Ave. Suite 3200, Orange, CA 92868, USA Office: 1-714-456-6389; Email: Editor@westjem.org

Editorial and Publishing Office: JEM/Depatment of Emergency Medicine, UC Irvine Health, 3800 W. Chapman Ave. Suite 3200, Orange, CA 92868, USA

Western Journal of Emergency Medicine:

Integrating Emergency Care with Population Health

Indexed in MEDLINE, PubMed, and Clarivate Web of Science, Science Citation Index Expanded

JOURNAL FOCUS

Emergency medicine is a specialty which closely reflects societal challenges and consequences of public policy decisions. The emergency department specifically deals with social injustice, health and economic disparities, violence, substance abuse, and disaster preparedness and response. This journal focuses on how emergency care affects the health of the community and population, and conversely, how these societal challenges affect the composition of the patient population who seek care in the emergency department. The development of better systems to provide emergency care, including technology solutions, is critical to enhancing population health.

Education

903 Teaching the New Ways: Improving Resident Documentation for the New 2023 Coding Requirements

Nathan Zapolsky, Annemarie Cardell, Riddhi Desai, Stacey Frisch, Nicholas Jobeun, Daniel Novak, Michael Silver, Arlene S. Chung

907 Telesimulation Use in Emergency Medicine Residency Programs: National Survey of Residency Simulation Leaders

Max Berger, Jack Buckanavage, Jaime Jordan, Steven Lai, Linda Regan

913 Palliative Care Boot Camp Offers Skill Building for Emergency Medicine Residents

Julie Cooper, Jenna Fredette

Emergency Department Operations

917 Improving Patient Understanding of Emergency Department Discharge Instructions

Sarah Russell, Nancy Jacobson, Ashley Pavlic

921 Why Do Patients Opt for the Emergency Department over Other Care Choices? A Multi-Hospital Analysis

Charles W. Stube, Alexander S. Ljungberg, Jason A. Borton, Kunal Chadha, Kyle J. Kelleran, E. Brooke Lerner

929 Emergency Department Patient Satisfaction Scores Are Lower for Patients Who Arrive During the Night Shift

Tony Zitek, Luke Weber, Tatiana Nuñez, Luis Puron, Adam Roitman, Claudia Corbea, Dana Sherman, Michael Shalaby, Frayda Kresch, David A. Farcy

938 Comparison of Emergency Department Disposition Times in Adult Level I and Level II Trauma Centers

Sierra Lane, Jeffry Nahmias, Michael Lekawa, John Christian Fox, Carrie Chandwani, Shahram Lotfipour, Areg Grigorian

946 Weighing In

Iyesatta M. Emeli, Patrick G. Meloy

Emergency Medical Services

949 Perceptions and Use of Automated Hospital Outcome Data by EMS Providers: A Pilot Study

Michael Kaduce, Antonio Fernandez, Scott Bourn, Dustin Calhoun, Jefferson Williams, Mallory DeLuca, Heidi Abraham, Kevin Uhl, Brian Bregenzer, Baxter Larmon, Remle P. Crowe, Alison Treichel, J. Brent Myers

958 Validation of the Turkish Version of the Professional Fulfillment Index

Merve Eksioglu, Ayca Koca, Burcu Azapoglu Kaymak, Tuba Cimilli Ozturk, Atilla Halil Elhan

Policies for peer review, author instructions, conflicts of interest and human and animal subjects protections can be found online at www.westjem.com.

No. 6: November 2024

Western Journal of Emergency Medicine:

Integrating Emergency Care with Population Health

Indexed in MEDLINE, PubMed, and Clarivate Web of Science, Science Citation Index Expanded

Table of Contents continued

Endemic Infection

966 Use of Parenteral Antibiotics in Emergency Departments: Practice Patterns and Class Concordance

Megan Elli, Timothy Molinarolo, Aidan Mullan, Laura Walker

975 Diagnostic and Prognostic Value of SCUBE-1 in COVID-19 Patients

Vildan Ozer, Ozgen Gonenc Cekic, Ozlem Bulbul, Davut Aydın, Eser Bulut, Firdevs Aksoy, Mehtap Pehlivanlar Kucuk, Suleyman Caner Karahan, Ebru Emel Sozen, Esra Ozkaya, Polat Kosucu, Yunus Karaca, Suleyman Turedi

985 Feasibility of Emergency Department-Initiated HIV Pre-Exposure Prophylaxis

Ezra Bisom-Rapp, Kishan Patel, Katrin Jaradeh, Tuna C. Hayirli, Christopher R. Peabody

Health Equity

993 Emergency Department Food Insecurity Screening, Food Voucher Distribution and Utilization: A Prospective Cohort Study

Alexander J. Ulintz, Seema S. Patel, Katherine Anderson, Kevin Walters, Tyler J. Stepsis, Michael S. Lyons, Peter S. Pang

Health Policy Perspectives

1000 The California Managed Care Organization Tax and Medi-Cal Patients in the Emergency Department

Lauren Murphy, Gita Golonzka, Ellen Shank, Jorge Fernandez

Neurology

1003 A Review of the Clinical Presentation, Causes, and Diagnostic Evaluation of Increased Intracranial Pressure in the Emergency Department

Cristiana Olaru, Sam Langberg, Nicole Streiff McCoin

Pediatrics

1011 Barriers to Adoption of a Child-Abuse Clinical Decision Support System in Emergency Departments

Alanna C. Peterson, Donald M. Yealy, Emily Heineman, Rachel P. Berger

1020 Reframing Child Protection in Emergency Medicine

Joseph P. Shapiro, Genevieve Preer, Caroline J. Kistin

Letters to the Editor

1025 Comments on “Bicarbonate and Serum Lab Markers as Predictors of Mortality in the Trauma Patient”

Patrick McGinnis, Samantha Camp, Minahil Cheema, Shriya Jaddu, Quincy Tran, Jessica Downing

Western Journal of Emergency Medicine:

Integrating Emergency Care with Population Health

Indexed in MEDLINE, PubMed, and Clarivate Web of Science, Science Citation Index Expanded

This open access publication would not be possible without the generous and continual financial support of our society sponsors, department and chapter subscribers.

Professional Society Sponsors

American College of Osteopathic Emergency Physicians

California American College of Emergency Physicians

Academic Department of Emergency Medicine Subscriber Albany Medical College Albany, NY

Allegheny Health Network Pittsburgh, PA

American University of Beirut Beirut, Lebanon

AMITA Health Resurrection Medical Center Chicago, IL

Arrowhead Regional Medical Center Colton, CA

Baylor College of Medicine Houston, TX

Baystate Medical Center Springfield, MA

Bellevue Hospital Center New York, NY

Beth Israel Deaconess Medical Center Boston, MA

Boston Medical Center Boston, MA

Brigham and Women’s Hospital Boston, MA

Brown University Providence, RI

Carl R. Darnall Army Medical Center Fort Hood, TX

Cleveland Clinic Cleveland, OH

Columbia University Vagelos New York, NY

State Chapter Subscriber

Arizona Chapter Division of the American Academy of Emergency Medicine

California Chapter Division of the American Academy of Emergency Medicine

Florida Chapter Division of the American Academy of Emergency Medicine

International Society Partners

Conemaugh Memorial Medical Center Johnstown, PA

Crozer-Chester Medical Center Upland, PA

Desert Regional Medical Center Palm Springs, CA

Detroit Medical Center/ Wayne State University Detroit, MI

Eastern Virginia Medical School Norfolk, VA

Einstein Healthcare Network Philadelphia, PA

Eisenhower Medical Center Rancho Mirage, CA

Emory University Atlanta, GA

Franciscan Health Carmel, IN

Geisinger Medical Center Danville, PA

Grand State Medical Center Allendale, MI

Healthpartners Institute/ Regions Hospital Minneapolis, MN

Hennepin County Medical Center Minneapolis, MN

Henry Ford Medical Center Detroit, MI

Henry Ford Wyandotte Hospital Wyandotte, MI

Emergency Medicine Association of Turkey Lebanese Academy of Emergency Medicine Mediterranean Academy of Emergency Medicine

California Chapter Division of American Academy of Emergency Medicine

INTEGRIS Health Oklahoma City, OK

Kaiser Permenante Medical Center San Diego, CA

Kaweah Delta Health Care District Visalia, CA

Kennedy University Hospitals Turnersville, NJ

Kent Hospital Warwick, RI

Kern Medical Bakersfield, CA

Lakeland HealthCare St. Joseph, MI

Lehigh Valley Hospital and Health Network Allentown, PA

Loma Linda University Medical Center Loma Linda, CA

Louisiana State University Health Sciences Center New Orleans, LA

Louisiana State University Shreveport Shereveport, LA

Madigan Army Medical Center Tacoma, WA

Maimonides Medical Center Brooklyn, NY

Maine Medical Center Portland, ME

Massachusetts General Hospital/Brigham and Women’s Hospital/ Harvard Medical Boston, MA

Great Lakes Chapter Division of the American Academy of Emergency Medicine

Tennessee Chapter Division of the American Academy of Emergency Medicine

Norwegian Society for Emergency Medicine Sociedad Argentina de Emergencias

Mayo Clinic Jacksonville, FL

Mayo Clinic College of Medicine Rochester, MN

Mercy Health - Hackley Campus Muskegon, MI

Merit Health Wesley Hattiesburg, MS

Midwestern University Glendale, AZ

Mount Sinai School of Medicine New York, NY

New York University Langone Health New York, NY

North Shore University Hospital Manhasset, NY

Northwestern Medical Group Chicago, IL

NYC Health and Hospitals/ Jacobi New York, NY

Ohio State University Medical Center Columbus, OH

Ohio Valley Medical Center Wheeling, WV

Oregon Health and Science University Portland, OR

Penn State Milton S. Hershey Medical Center Hershey, PA

Uniformed Services Chapter Division of the American Academy of Emergency Medicine

Virginia Chapter Division of the American Academy of Emergency Medicine

for Emergency Medicine

To become a WestJEM departmental sponsor, waive article processing fee, receive electronic copies for all faculty and residents, and free CME and faculty/fellow position advertisement space, please go to http://westjem.com/subscribe or contact:

Stephanie Burmeister

WestJEM Staff Liaison

Phone: 1-800-884-2236

Email: sales@westjem.org

25, No. 6: November 2024

Sociedad Chileno Medicina Urgencia Thai Association

Western Journal of Emergency Medicine:

Integrating Emergency Care with Population Health

Indexed in MEDLINE, PubMed, and Clarivate Web of Science, Science Citation Index Expanded

This open access publication would not be possible without the generous and continual financial support of our society sponsors, department and chapter subscribers.

Professional Society Sponsors

American College of Osteopathic Emergency Physicians

California American College of Emergency Physicians

Academic Department of Emergency Medicine Subscriber Prisma Health/ University of South Carolina SOM Greenville Greenville, SC

Regions Hospital Emergency Medicine Residency Program St. Paul, MN

Rhode Island Hospital Providence, RI

Robert Wood Johnson University Hospital New Brunswick, NJ

Rush University Medical Center Chicago, IL

St. Luke’s University Health Network Bethlehem, PA

Spectrum Health Lakeland St. Joseph, MI

Stanford Stanford, CA

SUNY Upstate Medical University Syracuse, NY

Temple University Philadelphia, PA

Texas Tech University Health Sciences Center El Paso, TX

The MetroHealth System/ Case Western Reserve University Cleveland, OH

UMass Chan Medical School Worcester, MA

University at Buffalo Program Buffalo, NY

State Chapter Subscriber

Arizona Chapter Division of the American Academy of Emergency Medicine

California Chapter Division of the American Academy of Emergency Medicine

Florida Chapter Division of the American Academy of Emergency Medicine

International Society Partners

University of Alabama Medical Center Northport, AL

University of Alabama, Birmingham Birmingham, AL

University of Arizona College of Medicine-Tucson Tucson, AZ

University of California, Davis Medical Center Sacramento, CA

University of California, Irvine Orange, CA

University of California, Los Angeles Los Angeles, CA

University of California, San Diego La Jolla, CA

University of California, San Francisco San Francisco, CA

UCSF Fresno Center Fresno, CA

University of Chicago Chicago, IL

University of Cincinnati Medical Center/ College of Medicine Cincinnati, OH

University of Colorado Denver Denver, CO

University of Florida Gainesville, FL

University of Florida, Jacksonville Jacksonville, FL

Emergency Medicine Association of Turkey Lebanese Academy of Emergency Medicine Mediterranean Academy of Emergency Medicine

California Chapter Division of American Academy of Emergency Medicine

University of Illinois at Chicago Chicago, IL

University of Iowa Iowa City, IA

University of Louisville Louisville, KY

University of Maryland Baltimore, MD

University of Massachusetts Amherst, MA

University of Michigan Ann Arbor, MI

University of Missouri, Columbia Columbia, MO

University of North Dakota School of Medicine and Health Sciences Grand Forks, ND

University of Nebraska Medical Center Omaha, NE

University of Nevada, Las Vegas Las Vegas, NV

University of Southern Alabama Mobile, AL

University of Southern California Los Angeles, CA

University of Tennessee, Memphis Memphis, TN

University of Texas, Houston Houston, TX

University of Washington Seattle, WA

Great Lakes Chapter Division of the American Academy of Emergency Medicine

Tennessee Chapter Division of the American Academy of Emergency Medicine

Norwegian Society for Emergency Medicine Sociedad Argentina de Emergencias

University of WashingtonHarborview Medical Center Seattle, WA

University of Wisconsin Hospitals and Clinics Madison, WI

UT Southwestern Dallas, TX

Valleywise Health Medical Center Phoenix, AZ

Virginia Commonwealth University Medical Center Richmond, VA

Wake Forest University Winston-Salem, NC

Wake Technical Community College Raleigh, NC

Wayne State Detroit, MI

Wright State University Dayton, OH

Yale School of Medicine New Haven, CT

Uniformed Services Chapter Division of the American Academy of Emergency Medicine

Virginia Chapter Division of the American Academy of Emergency Medicine

for Emergency Medicine

Stephanie Burmeister

WestJEM Staff Liaison

Phone: 1-800-884-2236

Email: sales@westjem.org

Sociedad Chileno Medicina Urgencia Thai Association

REVIEW

RecentInterventionsforAcuteSuicidalityDeliveredinthe EmergencyDepartment:AScopingReview

AlexP.Hood,BA*†

LaurenM.Tibbits,MD*

JuanI.Laporta,BS*

JenniferCarrillo,BS*

LaceeR.Adams,BS*

StaceyYoung-McCaughan,RN,PhD‡

AlanL.Peterson,PhD,ABPP‡§

RobertA.DeLorenzo,MD,MSM,MSCI*

SectionEditor:BradBobrin,MD

*UniversityofTexasHealthScienceCenteratSanAntonio, DepartmentofEmergencyMedicine,SanAntonio,Texas

† BaylorUniversity,DepartmentofPsychologyandNeuroscience, Waco,Texas

‡ UniversityofTexasHealthScienceCenteratSanAntonio,Department ofPsychiatryandBehavioralSciences,SanAntonio,Texas

§ UniversityofTexasatSanAntonio,DepartmentofPsychology, SanAntonio,Texas

Submissionhistory:SubmittedDecember21,2023;RevisionreceivedAugust12,2024;AcceptedAugust16,2024

ElectronicallypublishedOctober9,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18640

Introduction: SuicidalityisagrowingproblemintheUS,andtheemergencydepartment(ED)isoften thefrontlineforthemanagementandeffectivetreatmentofacutelysuicidalpatients.Thereisadearthof interventionsthatemergencyphysiciansmayusetomanageandeffectivelytreatacutelysuicidal patients.TotheextentthatrecentlydescribedinterventionsareavailableforEDpersonnel,noreview hasbeenconductedtoidentifythem.Thisscopingreviewisintendedto ﬁllthisgapbysystematically reviewingtheliteraturetoidentifyrecentlydescribedinterventionsthatcanbeadministeredintheEDto reducesymptomsandstabilizepatients.

Methods: WeconductedasearchofPubMed,SCOPUS,andCINAHLinJanuary2024toidentify paperspublishedbetween2013–2023fororiginalresearchtrialingrecentinterventionsfortheeffective treatmentofsuicidalityintheED.Weassessed16full-textarticlesforeligibility,andninemetinclusion criteria.Includedstudieswereevaluatedforfeaturesandcharacteristics,the ﬁtoftheinterventiontothe EDenvironment,andinterventionalefﬁcacy.

Results: Fourstudiesassessedtheefficacyofasingledoseoftheanesthetic/analgesicagentketamine. Threestudiesassessedtheef ficacyofabriefpsychosocialinterventiondeliveredintheED,twoofwhich pairedthisinterventionwiththeprovisionoffollow-upcare(postcardcontactandreferralassistance/ casemanagement,respectively).Theremainingtwostudiestrialedabrief,motivationalinterviewingbasedintervention.Includedstudieshadstrongexperimentaldesigns(randomizedcontrolledtrials)but smallsamplesizes(average57).Amongtheinterventionsrepresentedacrosstheseninestudies,a singledoseofketamineandthebriefpsychosocialinterventionCrisisResponsePlanning(CRP)show promiseasED-appropriateinterventionsforsuicidality.KetamineandCRPdemonstratedthestrongest fittotheEDenvironmentandmostrobustefficacy findings.

Conclusion: Thisreviewidentifiedonedrug(ketamine)andfouruniquepsychological/behavioral interventionsthathavebeenusedtotreatacutesuicidalityintheED.Thereiscurrentlyinsufficient evidencetosuggestthattheseinterventionswillproveefficaciousandwell-suitedtobedeliveredinthe EDenvironment.FuturestudiesshouldcontinuetotesttheseinterventionsintheEDsettingtodetermine theirfeasibilityandefficacy.[WestJEmergMed.2024;25(6)858–868.]

INTRODUCTION

Overthepasttwodecades,thesuiciderateintheUS generalpopulationincreasedbyover33%.1 Uptohalfof suicidedecedentsvisitanemergencydepartment(ED)during theyearbeforetheirdeath,andapproximately25%visitin themonthimmediatelyprior.2,3 Theriskfordeathbysuicide amongEDpatientspresentingwithsuicidalthoughtsand behaviorsremainshighforatleastoneyearafter discharge.4,5 TheEDisoftenthe ﬁrstmedicalaccesspointfor thosewithanacutedeteriorationintheirmentalhealth; approximately10%ofallEDvisitsareformentalhealth concerns. 5–7 Newandinnovativeapproachesareneededto stemthetideofsuicidesandtohelpmitigatethecrisisof psychiatricboardinginEDs.8,9

Emergencydepartmentpersonnelhaveincreasingly voicedconcernsoverabrokensystemofmentalhealthcare thathasexacerbatedconditionsforEDpatientswith psychiatricemergencies.10 Suchserioussystemdeﬁciencies maycontributetotheperceptionofsuicidalEDpatientswho describeEDpersonnelaslackingempathy,andbeing brusque,irritable,andevenhostile.11 Exacerbatingthe problemisthatthenumberofstate-fundedinpatient psychiatricbedshasdroppedsubstantially,from340beds per100,000peoplein1995tounder12bedsper100,000by 2016.8,9 Conversely,thenumberofEDvisitsforpsychiatric complaintshasrisenby50%.8 Thishasledtoasituation wheremanypatientswhorequireinpatientmentalhealth caremustwaitintheEDuntilapsychiatricbedbecomes available.Thisdelayintransferringpatientstoaninpatient unitleadsto “psychiatricEDboarding.”12

Thestate-of-the-artinterventionsavailabletoemergency physiciansareorientedtowardsafelydischargingpatients homeandconnectingthemtodeﬁnitivementalhealth services.13,14 Briefinterventionsorreferralfollowedby dischargehomearecommonforpatientspresentingwith non-life-threateningsuicidalthoughtsandbehaviors, whereaspatientspresentingwithmoderatetosevererisk behaviorsforsuicideareusuallykeptintheEDuntiltransfer toaninpatientpsychiatricfacilityispossible.15 Thissplitting ofpatientsintocategoriesofriskseverity16 meansthatthe higherapatient’sriskforsuicide,thefewerinterventionsare availabletoaddressthepatient’sparticularneeds.Notably, nopharmacologicagenthasbeenapprovedbytheUS FoodandDrugAdministrationtotreatsuicidalityinthe ED;mostmedicationsadministeredtosuicidalED patientstypicallytargetonlyagitation,notthesuicidal symptomsthemselves.16,17

Fromapsychiatricperspective,mostavailable interventionstargetsuicidalthoughtsandbehaviorsoverthe longtermasopposedtotheshort-ormediumterm17 andare thereforeill-suitedtotheacutecareenvironment. Psychopharmacologicagentssuchasantidepressants, lithium,andantipsychoticsgenerallyrequireacourseof weeksormonthstotakeeffect,14 andbeginningacourseof

antidepressanttreatmentcanparadoxicallyincrease suicidalityinsomepopulations.18 Similartimescalesare requiredforempiricallysupportedpsychotherapiessuchas cognitivebehavioraltherapyandothers,17,19 andeventhe mostabbreviatedstandardinterventionscantakeupto sixweeks.20

Whiletheimportanceofscreeningforsuicidalityiswell understood,21 thereisgrowingneedforevidence-based, rapidlyacting,effectivetreatmentoptions.15 Manyexisting toolssuitedtotheEDenvironmentthattargetsuicidality lacksupportingevidenceor,worse,are counterproductive.22,23 Onesuchinterventionisthesafety contractorno-suicideagreement.Whileatonetimethegold standardforEDanti-suicidalinterventions,thesafety contracthasbeenshowntoproduceworseoutcomesthanno interventionatall.21,24 Totheextentthatmorerecent interventionsfortheeffectivetreatmentofacutesuicidality haveemerged,therehasbeennoreviewcreatedspeciﬁcallyto identifyanddescribepotentialinterventions.

AnanalysisbyInagakiandcolleagues25 identifiedbroad classesofinterventionstopreventrepeatsuicideattemptsin patientsadmittedtoanEDbutdidnotinvestigatewhich interventionswouldbebestsuitedtotheEDenvironment. Changandcolleaguesprovidedareviewofmajordepressive disorderandsuicidalityintheEDbutdidnotofferan analysisofrecentlydescribedinterventions.21 Ina2021 review,Mannandcolleagues26 surveyedthelandscapefor evidence-basedtherapiesforsuicidalityingeneral,butthey didnotfocusspecificallyontheED.Whileotherrecent reviewshaveassessedtheavailabilityofclinician-oriented educationalinterventions,23 orinterventionsformental decompensationingeneral,27 nonehavethoroughlyassessed theliteratureforrecentlydescribedtoolsthatcliniciansmay usetotreatacutesuicidalityinthecontextoftheED. Lengvenyteandcolleagues28 publishedasystematicreview ontheimmediateandshort-termefficacyofsuicide-targeting interventionsbutdidnotfocusonrecentinterventionsused intheED.Weundertookthisreviewto fillthegapand exploretheliteraturetoidentifyanddescriberecent,patientcenteredinterventionsfortheeffectivetreatmentofacute suicidalityintheED.

Inthisreviewwefocusedonrecentlydescribed interventionsthatcanbeadministeredinthecontextofa patient’sstayintheED,namely,brieftherapiesand pharmacologicagentsthat fitwiththestandardmedical modeloftreatment.State-of-the-artpractice(ie,generally acceptedcare),definedasinterventionsforacutesuicidality, aredescribedin Rosen ’ sEmergencyMedicine:Concepts andClinicalPractice 29 or KaplanandSadock ’ s ComprehensiveTextbookofPsychiatry . 30 Theseinclude screening,jointsafetyplanning,patienteducation,lethal meanscounseling,follow-upcontacts,andtheinvolvement offriendsandfamily.29,30 Interventionslistedinor moderatelymodifiedfromthosedescribedinthesetextbooks

wereconsideredstate-of-the-artandexcludedfromthe search.Theprimaryquestionofthisreviewwasasfollows: Whatrecentlydescribedinterventionsareavailableforusein reducingsuicidalityandstabilizingpatientsduringa psychiatriccrisisintheED?

METHODS

WesearchedPubMed,SCOPUS,andCINAHLon January15,2024.Thisreviewwasconductedinaccordance withbest-practicerecommendationsofthePreferred ReportingItemsforSystematicReviewsandMeta-Analysis (PRISMA)extensionforscopingreviews.31 Inclusioncriteria includedthefollowing:1)studyparticipantpatientswere presentingtotheEDwithsuicidalideation;2)thestudy assessedtheefﬁcacyofoneormorepatient-centered intervention(s)aimedatreducingsuicidalthoughtsand behaviors;3)theinterventionbeingtestedwasadministered topatientsintheED;4)theinterventionwasadministeredby emergencyphysiciansorpersonnel;5)thestudywas availableinEnglish;and6)thestudyhadbeenpublishedin thelast10years.

DeﬁnitionofSuicidalityandRecentInterventions

Weadoptedthesuicidalitynomenclatureproposedby Silvermanandcolleagues.32 Studiesimplementingthebroad termsuicide-relatedthoughtsandbehaviors(SRTB),orany sub-categorythereof,wereconsideredeligibleforinclusion. Foraresourceonresearch-validatedscalesforthe measurementofsuicidalitywereliedonthelistcompiledby Ghasemi,Shaghaghi,andAllahverdipour.33 Wesoughtto identifyrecentlydescribed,effectivetreatmentsforthe preventionofsuicidalbehaviorthatareoutsidethestate-ofthe-art(currentstandards).Tothisend,wedeﬁnedrecent interventionsasbeingpatient-centered,deliveredintheED, describedwithinthepast10years,andnotalreadypartof recognizedstate-of-the-artpractice.

FeaturesofEligibleStudies

Weassessedstudiesforcharacteristicfeaturesoncethey wereincludedintheanalysis,andweevaluatedthe comparativestrengthsandweaknessesofstudydesign, samplesize,etc.Studiesconsideredidentiﬁedaspeciﬁc, recentinterventionforacutesuicidalityincontextoftheED intheprevious10years,sinceearlierinterventions wereconsideredmorelikelytobeconsistentwith state-of-the-artpractice.

Searchstrategy

Weusedathree-stepsearchstrategyinconsultationwitha libraryscientist.Inthe ﬁrstphase,weconducteda preliminarysearchofPubMedtoensurerelevantresultswere retrievedfromoursearchterms.Inthesecondphase,the searchtermswereappliedtoPubMed,SCOPUS,and CINAHL.See AppendixA forthesearchtermsused.Inthe

thirdphase,wescannedtheresultsfromthesearchconducted inphasetwoforreferencesincludedinstudybibliographies thatcouldhaveprovidedadditionalarticles.Thedatabase searchstrategyissummarizedin AppendixA.Weconducted additionalsearchesofGoogletoidentifygrayliterature orpublicationsnotdiscoveredviatheabove-described searchprocess.

Studyselection

Oncesearchtermsandkeywordswerenarroweddown,we removedduplicatesfromthelistofarticles.Four independentreviewersscreenedthetitlesandabstractsfor relevanceofallremainingstudies.Articlesdeterminedtobe relevantatthisstagewereretrievedinfull-textformand screenedforrelevancebytwoindependentreviewers.Apreselectedarbitersettledinconsistenciesbetweenreviewers.We recordedanddocumentedreasonsforexclusionforany article.Avisualizationofthisprocessisincludedina PRISMA ﬂowdiagram34 in Figure1,whichalsoprovidesaa summaryofresultsinstandardPRISMAformat.

Dataextraction

Data ﬁeldscollectedfromincludedstudiesare summarizedinthedataextractiontoolgivenin AppendixB. TheprimaryauthorA.P.H.extracteddatausingthetool, andthedatawascheckedforaccuracyandcompletenessby anindependentreviewer.

RESULTS

Afterduplicateswereremoved,weanalyzed1,197studies. Thereareafewreasonsforthislargenumberofresults.In keepingwithbest-practiceguidelines,andtoavoidthe

iden ﬁed through database search (N = 1213)

iden ﬁed in gray literature (n = 0)

a er duplicates removed (n = 1197)

iden ﬁed via References sec ons (n = 0)

Titles/abstractsscreened (n = 1197) Records excluded (n = 1181)

Full-text ar cles assessed for eligibility (n = 16) Full-text ar cles excluded, with reasons (n = 7)

Studies included (n = 9)

Figure. PRISMA ﬂowdiagramofstudyselection.

improperexclusionofanyrelevantarticles,weusedbroad searchtermstoreturnthemaximumnumberofpotentially relevantarticles.Additionally,noMeSHtermsspeciﬁcally aimedatexcludingscreeningandriskassessmentwereused. Thetitle/abstractreviewphasewas,therefore,acriticalstage fortheisolationofrelevantarticles,andweremoved1,181 fromfurtheranalysis.

Inthenextphaseofeligibilityscreening,twoindependent reviewersretrievedandevaluatedthefulltextsof16articles, ofwhichsevenwereexcluded.Disagreementsweresettledby anemergencyphysicianwithrelevantexpertiseinacutecare interventionsforsuicidalityR.A.D.Oftheexcludedarticles, twostudiesinvolvedinterventionsthatweretailoredtothe uniqueculturalpracticesofspeciﬁcindigenousgroupsand werethereforedeemednotgeneralizabletoallsuicidal patientspresentingtoEDs.35,36 Twoadditionalarticleswere excludedastheyusedasafetyplanningintervention operationallydeﬁnedaspartofstate-of-the-artpractice.One articlewasexcludedbecausethestudyinterventiondidnot occurintheEDsetting.37 Finally,weexcludedtwo secondaryanalysesofarticlesthathadalreadybeen included.38,39 Oftheninearticlesincludedintheanalysis,we extracteddatausingthetoolgivenin AppendixA.Anoverviewofrelevantdatafromeachstudy ispresentedinthe Table

Fourincludedarticlesassessedtheuseofasingledoseof thepharmacologicinterventionketamine,aN-methyl-Daspartateantagonistcommonlyusedasasedative,analgesic, andanesthetic.40–43 Threestudiesassessedtheuseofan intravenous(IV)infusion,40,42,43 andoneassessedthe efﬁcacyandtolerabilityofanintranasaladministration.41 Twooftheselectedarticlesassessedtheefﬁcacyof interventionscenteredonmotivationalinterviewing(MI) embeddedinaninterventionalframeworkwithprovisions forfollow-upcareorreferralassistance.44,45 BothTeen OptionsforChange(TOC)45 andSuicidalTeensAccessing TreatmentAfteranEmergencyDepartmentVisit(STATED)44 targetedadolescentsamples.

Thethreeremainingincludedarticlesstudiedvarious interventionscenteredonacutepsychotherapyand/or behavioralmanagementinthepost-acuteperiod.Byfarthe largestsampleamongincludedarticleswasastudyof assertivecasemanagementforthosepresentingtotheED afterasuicideattempt.46 Whilethelengthy(18+ months) interventionunderstudyinthisarticlelargelytookplace followingdischargefromtheED,theintervention proceduresbeganwhilepatientswereintheEDandwere deliveredbypsychiatristsorothermedicalpersonnel.46

Anotherstudythatmetourcriteriainvestigatedtheefficacy ofthenovel,manualizedProblemSolvingand ComprehensiveContactIntervention(PS-CCI),whichusesa collaborativelycompletedworksheetaimedatenhancing self-efficacyandcognitive flexibilityinsuicidalEDpatients pairedwithfollow-upcare.47 Finally,astudyoftheCrisis

ResponsePlan(CRP)interventionconductedbyBryanand colleaguesinamilitaryEDmetinclusioncriteria.22 TheCRP pairsabriefhistoricalinterviewwithacollaborative identiﬁcationanddocumentationofcopingstrategiesand resourcesavailabletopatients.22

StudyFeaturesandCharacteristics

Severalmeasuresofstudyfeaturesandcharacteristics weregatheredintheprocessofdataextraction.Weused PRISMAguidelinestohelpdeﬁneelementsofquality31 includingsamplesize,studydesign,follow-uptimeframe, andmeasuresused.

Sample

Thesamplesizesofthemajorityofstudiesmeeting inclusioncriteriaweresmall.Excludingtheoutlierof Kawanishietal46 withtheirrobustsampleof914,theaverage samplesizeforincludedstudieswas57,withthesmallest sample(10)collectedbyBurgerandcolleagues.40

Design

Sevenofninestudiesconductedarandomizedcontrolled trial(RCT),onewasaquasi-experimentaltwo-arm prospectivelongitudinalstudy,47 andonewasanonexperimentalpilotstudydesignedtoevaluatethefeasibility andefﬁcacyofasinglelowdoseofketaminedeliveredviaIV bolus.43 Double-blindingandrandomassignmentwere consistentlypracticedamongtheRCTsassessingtheefﬁcacy ofketamine,andparticipantsinthecontrolconditions receivedaninactiveplaceboinjection/atomizationofnormal saline.40–42 Kawanishiandcolleagues,46 Kingetal,45 and Grupp-Phelanetal44 usedsingleblindingandacomparator conditionofenhancedusualcare(EUC).Intheirthree-arm RCT,Bryanandcolleagues22 comparedtwoversionsofCRP (standardandenhanced)tothecontrolconditionofa contractforsafety,andparticipantswereblindedtogroup assignment.Althoughthecontractforsafetywaspreviously astandardintervention,ithasmanynoteworthy shortcomings22 makingitaless-than-idealcomparison conditiontoCRP.24

Follow-upmeasuresandtimeframe

Sevenoftheninestudiesincludedinthisreviewused standard,well-subscribed,psychometricallyvalidated measuresofsuicidalitytoassessoutcomevariablesof interest,aswellasevaluationsofrepeatedhospitalizations andhealthcareutilization.Themostcommonscalesused weretheColumbiaSuicideSeverityRatingScale,theBeck ScaleforSuicidalIdeation,andtheMontgomery-Asberg DepressionRatingScale.However,twostudiesevaluated onlypost-dischargesuicideattempts,suicidedeaths,and psychiatrichospitalizationrecidivismwithoutmakinguseof psychometricmeasures.46,47 Thefollow-uptimeframes

Table. Summaryofindividualstudies.

PS-CCIisfeasibleas aninterventionfor suicideintheED

3mosDidnothavethe sampleforstatistical signi fi cance,butthey appearedto fi ndthat thisinterventionwas feasibleasmeasured byacceptability, engagement, fi delity, subjectretention,and lackofAEs.

Structured worksheet manualized

AuthorsYearCountryObjective(s)InterventionMethodsN

Intervention somewhateffective shorttermandamong women,butoverall assertivecase managementdidnot outperformcontrol condition

Lowerrecurrent suicidalbehaviorat1, 3,and6mos;butat 18mostherewasno signi ﬁ cantdifference inratesofattempted orcompletedsuicides

BriefMIwasnot usefulinthiscontext

443mos “ Clinician ” not otherwise de ﬁ ned

Longitudinal design

2014USATodeterminethe feasibilityand acceptabilityofa novel,manualized problem-solvingand comprehensive contactintervention (PS –CCI)aimedat improvingtreatment engagementof suicidalindividuals. Manualized problem-solving and comprehensive contact intervention

Dana Alonzo

91418mos to 5years Psychiatrist, nurses,clinical psychologists, etc Case management manual, phone 18 mos-5 yrs

Two-armed RCT

Assertivecase management

2014JapanTodetermineif assertivecase managementcan reducerepeated suicideattemptsin peoplewithmental healthproblemswho hadattempted suicideandwere admittedtoEDs

Kawashini etal

None6mosMIwasnotbetter thanenhancedusual care

1592mos total Licensed socialworker

Two-armed RCT

2019USAToexaminewhether motivational interviewing(MI) increaseslinkageof adolescentsto outpatientmental healthservicesfor depressionand suiciderisk MIpluslinking tofollowup services

Grupp- Phelan etal

CRPishighly effectiveforuseinthe acutecaresettingfor acutelysuicidal patients

Indexcard6mosStandardCRPand reasonsforliving CRPwerebothmore effectivethana contractforsafetyin reducingsuicide attemptsandinpatient hospitalization.No differencebetween twoversionsofCRP

971hrCliniciannot de ﬁ ned

CRPThree-armed RCT

Bryanetal2017USAToevaluatethe effectivenessofcrisis responseplanning (CRP)forthe preventionofsuicide attempts

IVketamineappears tobefeasibleand effectiveintheshort term (

Signi ﬁ cantdifferences insuicidalityscores 90 –180minsafterIV ketamineinfusion. Safetyandtolerability wereexcellent.

Two weeks

IVbag,drug, 10mlsyringe, vitalsign monitoring equipment

185minsBoard-certi ﬁ ed

EDphysician

KetamineIVPlacebo- controlled doubleblind RCT

2020USAToassessthesafety, feasibility,tolerability, andef ﬁ cacyofa singlelow-dose intravenous(IV) ketamineinreducing suicidalideation

Domany and McCullums Smith

Table. Continued.

Ketamineshows promiseasanacute caretreatment modality,butfurther researchisneededto validatetheeffectsof thisverysmalltrial.

Groupsvaried signi ﬁ cantlyinthe short-termduring hospitalization; however, discrepanciesin scoresontherating scalesdiminishedat thetwo-weekfollow- uppoint.

Two weeks

AuthorsYearCountryObjective(s)InterventionMethodsN

Thisintervention appearstobe promisingbutwill needtobetestedina largersample

2mosLargeeffectforthe lowingofdepression scores,moderate effectfor hopelessness,and insigni ﬁ canteffects forsuicidalideation andalcoholuse (thoughthestudy wasn ’ tquitelarge enoughtostatistically verifythelatter ﬁ ndings

Theauthorsnotethat ketaminedidnot showitselfasa promisingtoolto mitigateacute suicidalityintheED context,asthe reductionsin suicidalityscores werenotclinically meaningful.However, theydonotethat moreresearchis neededandthatthis resultisfarfrom de ﬁ nitive(giventhe samplesizeandlack ofcontrolgroup).

Theketamineinfusion wasassociatedwitha bringingdownthe MADRSandSSI scoresbya statisticallysigni ﬁ cant margin,butthe authorsnotethat,for theirpurposes,the reductioninscores wasnotclinically meaningful(e.g., belowthethreshold setforthepurposes ofthestudyofa MADRSscoreof ≤ 4). Thelargestreduction inself-reported suicidalityoccurred within40minsof ketamine administration.

( Continuedonnextpage )

105minsTrainednurseIVbag,vital sign monitoring equipment

KetamineIVDouble-blind RCT

Burgeretal2016USATocompareIV ketaminetoplacebo amongacutely suicidalpatientsina militarysetting

Handwritten followup note

Trained mentalhealth professional withminimum of40hrsof specialized training

4945mins ofMI,5 daysof follow up

RCTwith enhanced TAUas control condition

TeenOptions forChange (TOC)

Kingetal2015USAThisrandomizedtrial examinedthe effectivenessofTeen OptionsforChange (TOC),an interventionfor adolescentsseeking medicalemergency serviceswhoscreen positiveforsuicide

IVbag,vital sign- monitoring equipment 10 days

495minsNurseor attending doctor

Uncontrolled pre-test/post- test

KetamineIV bolus

2014IranThisstudywas conductedto examinetheeffects ofasingle intravenousbolusof ketamineonpatients withsuicidalideation intheED.

Kashani etal

Table. Continued.

Whilethesampleis comparativelysmall andtheoverallresults arefarfrom conclusive,thisstudy importantlyechoes previous ﬁ ndings demonstratingthata lowdoseofketamine doesappearto transientlyreduce suicidalideationin thoseathighacute riskforsuicide.In addition,itprovides promisingevidence thatasingle, ﬁ xed doseofketaminemay havethepotentialto reducehospital lengthsofstay,and thatintranasalisa feasiblerouteof administrationforthe acutecaresetting.

Participantswhowere administered ketaminehadnotonly amuchhigherrateof remission(roughly 80%ofthesample), butalsotypically shorterlengthsof hospitalization. Ketaminewas generallysafe;minor sideeffectswere transientandresolved within1 –2hourspost- administration. However,thelinear mixedstatistical modeldidnotshow signi ﬁ canceon randomgroupbytime interaction.

4 weeks

PsychiatristIntranasal atomizer

AuthorsYearCountryObjective(s)InterventionMethodsN

3040mins to1hr

Double-blind RCT

Ketamine intranasal

2022USAEvaluationofsingle, ﬁ xed-dosed intranasalketamine foracutesuicidal ideationintheED

AEs ,adverseevents; ED ,emergencydepartment; IV ,intravenous; MADRS ,MontgomeryAsbergdepressionratingscale; RCT ,randomized,controlledtrial; SSI ,BeckScalefor SuicidalIdeation; TAU ,treatmentasusual.

Domany and McCullum Smith

variedsigniﬁcantlydependingontheinterventionunder studyandaddedtotheheterogeneityofthesample.

FitofInterventiontoEmergencyDepartmentEnvironment

Sinceweintendedtoanalyzerecentlydescribedtools availabletoemergencyphysiciansforuseintheacutecare setting,attentionwaspaidtotheusabilityofeach interventionintheEDsetting.Wedeﬁnedusabilityaseaseof useand ﬁttotheEDenvironment,andthesewereevaluated alongthedimensionsoftotaltimerequiredtoadminister,the requiredtrainingorcredentialsoftheadministering practitioner,andthetoolsandmaterialsrequiredtodeliver theintervention.

Timetoadminister

Byfarthebriefestinterventionalmodalityamongthe articlesreviewedwasasingledoseofketaminedeliveredIV, which,atthemodestdosesused,averaged5–10 minutes.40,42,43 Whenadministeredintranasally,theinterval requiredtocompletetheintervention,althoughbrief(40–60 minutes),wassomewhatlonger.41 Notincludedinthe interventiondurationwasthemonitoringtimerequiredfor ketamineadministration,whichdependingonlocalprotocol canexceedseveralhours.EquivalentlybriefisCRP,which requires30–60minutestoadminister,makingitwell-suited tothedemandsofthefast-pacedEDenvironment.22 The studiesassessingMI-basedinterventionsforadolescentshad briefED-deliveredcomponents,requiringapproximately45 minutestodeliver.44,45 Thetimerequiredtoadministerthe ED-basedproblem-solvingcomponentoftheProblemSolvingandComprehensiveContactIntervention(PS-CCI) interventionwasnotspeciﬁed.47 Finally,Kawanishiand colleaguesdidnotspecifythetimecourseoftheED-based portionofassertivecasemanagement,buttheydidnotethe interventioninvolvedregularfollow-upappointments outsidetheEDoverthecourseof18months.46

Trainingrequiredtoadminister

Dueinparttovariabilityinhospitalpracticesindifferent regionsandcountries,thecredentialsofthehealthcare professionalsadministeringketaminevariedslightlyacross thefourstudiesthatinvestigateditsuse.40–43 Intravenous administrationwasconductedbyeitheranurseora physician,40,42,43 whereasthestudyusingintranasal ketaminerequiredsigniﬁcantinputfromapharmacist.41 BoththePS-CCI47 andCRP22 statedonlythatthe interventionwasdeliveredbya “clinician,” nototherwise speciﬁed.TheSTAT-EDdescribedbyGrupp-Phelanetal44 andTOCstudiedbyKingetal45 wereadministeredbya socialworkerandtrainedmentalhealthprofessional, respectively,withthelatterspecifyingthatinterventionalists wererequiredtohaveaminimumof40hoursof specializedtraining.Finally,theassertivecasemanagement interventiondescribedbyKawanishiandcolleagues46 was

conductedbycasemanagersatvariouslevelsoftraining, includingnurses,emergencyphysicians,psychiatrists,and clinicalpsychologists.

Toolsandmaterials

Formostpsychosocialinterventionsunderstudyinthe presentreview,fewspecializedmaterialswererequiredfor administration.Speciﬁcally,theSTAT-EDintervention,44 CRP,22 andTOC45 requirebasicofﬁceequipmentsuchas copypaperandnotecards.ThePS-CCIinterventionrequires theavailabilityofastructuredworksheet,47 andtheassertive casemanagementinterventionrequiresastandardized manual,46 makingtheirresourcedemandsminimal.Aswith allnovelpharmacologicinterventions,thestudiesassessinga singledoseofketaminerequiredtheavailabilityof equipmenttomonitorvitalsigns.40–43 ThoseassessingIV ketaminerequiredIVbags,pumps,lines,andhanging apparatuses,whichareusuallyavailableinED environments,40,42,43 whilethestudyofintranasal ketaminerequiredaspecializedatomizerpreparedbya pharmacyteam.41

EfﬁcacyFindings

Theinterpretationof findingsforarticlesdescribedinthe presentreviewshouldbemoderatedbylimitationsregarding samplesize,methodologicaldiscrepancies,andevidentiary quality.Twopromisinginterventionsweidentifiedarethe variousadministrationroutesofasingle,lowdoseof ketamine,40–43 andasinglemeetingtodevelopaCRP.22 For asingledoseofketamine,threearticlesreportedpositive findingsontheshort-termreductionofself-reported suicidalityanddepression,40–42 andonereported inconclusiveresults.43 Bryanandcolleagues22 foundthat participantsrandomizedtoeitherCRPcondition(standard orenhanced)showedsignificantreductionsinacutesuicidal ideation,fewersuicideattempts,andlowerratesofinpatient hospitalizationpost-dischargethanthoseinthe comparatorgroup.

Twootherinterventionsthatwereevaluated,PS-CCIand MI,showpromise,butthereisinsufficientevidenceto supporttheirefficacy.ThePS-CCItrial47 wasnotstatistically poweredtodetermineefficacy,buttheauthorsnotethatthe interventionisfeasibletoadministerintheEDsettinggiven itshightolerability.TheTOCstudy45 andtheSTAT-ED study44 trialedsimilarMI-basedtreatmentsincomparable adolescentsamplesbutreturnedconflictingresults.The studyofassertivecasemanagementbyKawanishiand colleagues46 hadalargesamplesizebutdemonstratedno significantdifferencebetweengroupsoverthecourseof thestudy.

DISCUSSION

Thepreliminaryresultsfromthefourketaminestudies includedinthisreviewecho ﬁndingsoftheuseofketamine

forsuicidalideationinoutpatientsettings.48 Therearea numberofadvantagestothisinterventionalmodality.43 First,ketamine,whenadministeredIVover5–10minutes,is byfarthebriefestinterventionnotconsideringthepostinfusionmonitoringtime.Intravenousketamineiswell suitedtothefast-pacedenvironmentoftheED.The intranasaladministrationrouteisalmostasbrief.

Intramuscular(IM)ketamineisanotheroptionbutrelatively unstudied;however,itmaybefamiliartoemergency clinicians.IfconfirmedinfullypoweredRCTs,sucharapidactinginterventionmaygiveemergencyphysicians additionaloptionsfortheplacementorevendischargeof patientswhopresentwithacutelyelevatedsuicideriskand couldserveasabridgetodefinitivementalhealthcarethat circumventstheneedforalengthydetentionintheED. Furthermore,adoseofgenericketamineisrelatively inexpensive49 andregularlystockedinmostEDs.The intranasalformofketamine,esketamine,incontrast,is moreexpensiveandlesswidelystocked.Additional researchontheefficacyofketamineforacutelysuicidalED patientsiswarranted.

ThisreviewfoundevidencethatCRPshowspromiseasan interventionwellsuitedtocombatacutesuicidalityintheED environment.Whilethereislimitedevidenceinsupportofthe efficacyofCRPintheED,thisinterventionhasseveral featuresthatmakeitwellsuitedtothespecificdemandson emergencymedicalpersonnel.First,similartoasingledose ofketamine,CRPisaninterventionalmodalitythatisbrief inadministrationandappearstorapidlydiminishacute suicidalityandimprovepatientmood.38 Additionally,CRP ismaximallyportabletoavarietyofenvironments,requires comparativelylittlespecializedtraining,toolsormaterialsto administer,and,asapsychosocialintervention,isnot contraindicatedforusewithanyconcomitantmedications. Despiteitsadvantages,theliteraturetodateontheuseof CRPintheEDcontextislimitedtoonestudy.22 Whilethe evidenceforinterventionssuchasthePS-CCI47 andTOC45 is mixed,ED-deliveredinterventionstargetingconstructsof cognitive flexibilityandadaptiveproblem-solvingappearto bearecipewithsomepromise(similartoCRP).

FutureDirections

LIMITATIONS

Althoughthepresentstudyhasmanynotablestrengths, someshortcomingsshouldbedelineated.First,wefocused oninterventionswithevidencesupportingtheabilitytobe performedinthechallengingEDenvironment.Itispossible thatrecentinterventionsunderstudyinotherclinical environmentsmayholdpromiseforadaptationtotheED setting.Second,asisthecasewithanyreview,itispossible thatcertaininterventionsextantintheliteraturewere erroneouslyexcludedfromouranalysisgiventhelimitations ofourMeSHsearchterms.Finally,tolimitouranalysisto onlythemostrecentinterventionswithanevidentiarybasis inthecurrentliterature,weassessedonlyarticlespublishedin theprevious10years.Itispossiblethattherearepromising, ED-basedinterventionsdescribedmorethan10yearsago thathavereceivednofurtherstudyintheinterveningtimeor havebeenstudiedexclusivelyoutsidetheEDcontextsince theirinitialdescription.

CONCLUSION

Therecentlydescribedinterventionsidentifiedfor emergencyphysicianstotreatacutesuicidalityarelimitedto onedrug(ketamine)andfouruniquepsychological/ behavioralinterventions.Twoofthe fiveinterventional modalitieshavepreliminaryevidenceandmayholdpromise inmitigatingacutesuicideriskintheED:asingle,lowdose ofketamineandcrisisresponseplanning.However,thereis insufficientevidencetosupporttheirwidespreadadoption. Futureresearchshouldextendthepreliminary findings summarizedinthisreview.

AddressforCorrespondence:AlexP.Hood,BA,UniversityofTexas HealthScienceCenteratSanAntonio,DepartmentofEmergency Medicine,703FloydCurlDrive,MC7736,SanAntonio,TX78229. Email: alex_hood1@baylor.edu

Thisstudyidentiﬁedtwopromisinginterventionssuitedto theEDenvironment:CRPandketamine.Theevidentiary basisfortheseinterventions,particularlyinbroad-based populationsofemergentlysuicidalEDpatients,isnotfully developed.Furtherstudyisrequiredtoascertaintheextentto whichtheseinterventionsserveaseffectivetreatmentsacross presentingpsychiatricsymptoms,especiallygiventhehigh incidenceofSRTBamongpatientswithseriousmental illnessoracuteintoxicationwithasubstance,whichmay complicateeffectivetreatment.50 Giventhecrisisofboarding inEDs,additionalfundingandstudyingeneralshouldbea nationalpriority.Futurestudiesshouldalsoinvestigate ketaminedeliveredviaalternativeroutesofadministration suchasorallyandIM.WhileCRPhasdemonstrated preliminaryefﬁcacy, 24 futureresearchshouldcompareCRP tovalidatedcurrentstandardpracticeinterventionsto properlyevaluateitseffectivenessagainsttreatmentasusual orEUC.Futurestudiesshouldalsovalidateuseofthe interventionoutsidethemilitarycontextwithparticipantsof variousbackgrounds,abilitylevels,andages.Giventhat briefMI-andCBT-basedinterventionsshowpromise,future studiesmayconsidercontinuingtohoneinterventionsthat approximatelyadheretothismodel.

relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconﬂictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.StoneDM,SimonTR,FowlerKA,etal.Vitalsigns:trendsinstate suiciderates-UnitedStates,1999–2016andcircumstances contributingtosuicide-27states,2015. MorbMortalWklyRep. 2018;67(22):617–24.

2.AhmedaniBK,SimonGE,StewartC,etal.Healthcarecontactsinthe yearbeforesuicidedeath. JGenInternMed. 2014;29(6):870–7.

3.GairinI,HouseA,OwensD.Attendanceattheaccidentandemergency departmentintheyearbeforesuicide:retrospectivestudy. BrJPsychiatry. 2003;183:28–33.

4.CrandallC,Fullerton-GleasonL,AgueroR,etal.Subsequentsuicide mortalityamongemergencydepartmentpatientsseenforsuicidal behavior. AcadEmergMed. 2006;13(4):435–42.

5.WeissAJ,BarrettML,HeslinKC,etal.(2006).Trendsinemergency departmentvisitsinvolvingmentalandsubstanceusedisorders, 2006–2013.In HealthcareCostandUtilizationProjectstatisticalbriefs Rockville,MD:AgencyforHealthcareResearchandQuality(US). Availableat: https://www.ncbi.nlm.nih.gov/books/NBK52651/ AccessedAugust6,2024.

6.HakenewerthAM.Emergencydepartmentvisitsbypatientswithmental healthdisorders–NorthCarolina,2008–2010. MorbMortalWklyRep. 2013;62(23):469–72.

7.OwensC,HansfordL,SharkeyS,etal.Needsandfearsofyoung peoplepresentingataccidentandemergencydepartmentfollowingan actofself-harm:secondaryanalysisofqualitativedata. BrJPsychiatry. 2016;208(3):286–91.

8.TreatmentAdvocacyCenter.Going,going,gone:trendsand consequencesofeliminatingstatepsychiatricbeds.2016.Availableat: https://www.treatmentadvocacycenter.org/reports_publications/ going-going-gone-trends-and-consequences-of-eliminatingstate-psychiatric-beds/.AccessedSeptember4,2023.

9.GoldJ.Adearthofhospitalbedsforpatientsinpsychiatriccrisis.2016. Availableat: https://kffhealthnews.org/news/a-dearth-of-hospitalbeds-for-patients-in-psychiatric-crisis/.AccessedAugust6,2024.

10.ZunL.Careofpsychiatricpatients:thechallengetoemergency physicians. WestJEmergMed. 2016;17(2):173–6.

11.FitzpatrickSJandRiverJ.Beyondthemedicalmodel:futuredirections forsuicideinterventionservices. IntJHealthServ. 2018;48(1):189–203.

12.NicksBAandMantheyDM.Theimpactofpsychiatricpatientboardingin emergencydepartments. EmergMedInt. 2012;2012:360308.

13.RihmerZPandMaurizioP.Mooddisorders:suicidalbehavior.In: SadockBJ(Ed.), KaplanandSadock’sComprehensiveTextbookof Psychiatry (3825),10thed.Philadelphia,PA:WoltersKluwer Health,2017:3825–51.

14.WassermanD.Suicide:overviewandepidemiology.In:SadockBJ(Ed.), KaplanandSadock’sComprehensiveTextbookofPsychiatry (5954). 10thed.Philadelphia,PA:WoltersKluwerHealth,2017:5954–72.

15.CapocciaLM.Caringforadultpatientswithsuiciderisk:aconsensusbasedguideforemergencydepartments.2015.Availableat: https://sprc.org/wp-content/uploads/2022/11/EDGuide_full.pdf AccessedAugust6,2024.

16.BetzJ.M.Suicide.In: Rosen’sEmergencyMedicine:Conceptsand ClinicalPractice (pagerange).9thed.Philidelphia,PA:Elsevier, Inc,2018:1366–73.

17.SudakH. Suicidetreatment.In:SadockB.J.(Ed.), KaplanandSadock’s ComprehensiveTextbookofPsychiatry (5973).10thed.Philadelphia, PA:WoltersKluwer,2017:5973–85.

18.ReevesRRandLadnerME.Antidepressant-inducedsuicidality: anupdate. CNSNeurosciTher. 2010;16(4):227–34.

19.RuddMD,BryanCJ,WertenbergerEG,etal.Briefcognitive-behavioral therapyeffectsonpost-treatmentsuicideattemptsinamilitary sample:resultsofarandomizedclinicaltrialwith2-yearfollow-up. AmJPsychiatry. 2015;172(5):441–9.

20.DavidsonKM,BrownTM,JamesV,etal.Manual-assistedcognitive therapyforself-harminpersonalitydisorderandsubstancemisuse: afeasibilitytrial. PsychiatrBull(2014). 2014;38(3):108–11.

21.ChangBP,TezanosK,GratchI,etal.Depressedandsuicidalpatientsin theemergencydepartment:anevidence-basedapproach. EmergMed Pract. 2019;21(5):1–24.

22.BryanCJ,MintzJ,ClemansTA,etal.Effectofcrisisresponseplanning vs.contractsforsafetyonsuicideriskinU.S.Armysoldiers: arandomizedclinicaltrial. JAffectDisord. 2017;212:64–72.

23.ShinHD,CassidyC,WeeksLE,etal.Interventionstochangeclinicians’ behaviorinrelationtosuicidepreventioncareintheemergencydepartment: ascopingreviewprotocol. JBIEvidSynth. 2021;19(8):2014–23.

24.RuddMD,MandrusiakM,JoinerTE.Thecaseagainstno-suicide contracts:thecommitmenttotreatmentstatementasapractice alternative. JClinPsychol. 2006;62(2):243–51.

25.InagakiM,KawashimaY,KawanishiC,etal.Interventionstoprevent repeatsuicidalbehaviorinpatientsadmittedtoanemergency departmentforasuicideattempt:ameta-analysis. JAffectDisord. 2015;175:66–78.

26.MannJJ,MichelCA,AuerbachRP.Improvingsuicideprevention throughevidence-basedstrategies:asystematicreview. AmJPsychiatry. 2021;178(7):611–24.

27.JohnstonAN,SpencerM,WallisM,etal.Reviewarticle:Interventions forpeoplepresentingtoemergencydepartmentswithamental healthproblem:asystematicscopingreview. EmergMedAustralas. 2019;31(5):715–29.

28.LengvenyteA,OliéE,StrumilaR,etal.Immediateandshort-term efﬁcacyofsuicide-targetedinterventionsinsuicidalindividuals:a systematicreview. WorldJBiolPsychiatry. 2021;22(9):670–85.

29.MarxJA,HockbergerRS,WallsRM,etal.In: Rosen’sEmergency Medicine:ConceptsandClinicalPractice.9thed.Philadelphia,PA: Mosby/Elsevier,2018.

30.SadockBJS,AlcottV,RuizP. KaplanandSadock’sComprehensive TextbookofPsychiatry.10thed.Philadephia,PA:WoltersKluwer,2017.

31.TriccoAC,LillieE,ZarinW,etal.PRISMAextensionforscoping reviews(PRISMA-ScR):checklistandexplanation. AnnInternMed. 2018;169(7):467–73.

32.SilvermanMM,BermanAL,SanddalND,etal.RebuildingtheTowerof Babel:arevisednomenclatureforthestudyofsuicideandsuicidal behaviors.Part2:suicide-relatedideations,communications,and behaviors. SuicideLifeThreatBehav. 2007;37(3):264–77.

33.GhasemiP,ShaghaghiA,AllahverdipourH.Measurementscalesof suicidalideationandattitudes:asystematicreviewarticle. HealthPromotPerspect. 2015;5(3):156–68.

34.MoherD,LiberatiA,TetzlaffJ,etal.Preferredreportingitemsfor systematicreviewsandmeta-analyses:thePRISMAstatement. PLoSMed. 2009;6(7):e1000097.

35.CwikMF,TingeyL,LeeA,etal.Developmentandpilotingofabrief interventionforsuicidalAmericanIndianadolescents. AmIndianAlsk NativeMentHealthRes. 2016;23(1):105–24.

36.HatcherS,CoupeN,WikiriwhiK,etal.TeIraTangata:aZelen randomisedcontrolledtrialofaculturallyinformedtreatmentcompared totreatmentasusualinMaoriwhopresenttohospitalafterself-harm. SocPsychiatryPsychiatrEpidemiol. 2016;51(6):885–94.

37.MillerIW,CamargoCAJr.,AriasSA,etal.Suicidepreventioninan emergencydepartmentpopulation:theED-SAFEstudy. JAMAPsychiatry. 2017;74(6):563–70.

38.BryanCJ,MintzJ,ClemansTA,etal.Effectofcrisisresponseplanning onpatientmoodandcliniciandecisionmaking:aclinicaltrialwith suicidalU.S.soldiers. PsychiatrServ. 2018;69(1):108–11.

39.FurunoT,NakagawaM,HinoK,etal.Effectivenessofassertivecase managementonrepeatself-harminpatientsadmittedforsuicide attempt: ﬁndingsfromACTION-Jstudy. JAffectDisord. 2018;225:460–5.

40.BurgerJ,CapobiancoM,LovernR,etal.Adouble-blinded,randomized, placebo-controlledsub-dissociativedoseketaminepilotstudyinthe treatmentofacutedepressionandsuicidalityinamilitaryemergency departmentsetting. MilMed. 2016;181(10):1195–9.

41.DomanyYandMcCullumsmithCB.Single, ﬁxed-doseintranasal ketamineforalleviationofacutesuicidalideation.Anemergency department,trans-diagnosticapproach:arandomized,double-blind,

placebo-controlled,proof-of-concepttrial. ArchSuicideRes. 2022;26(3):1250–65.

42.DomanyY,SheltonRC,McCullumSmithCB.Ketamineforacute suicidalideation.Anemergencydepartmentintervention:A randomized,double-blind,placebo-controlled,proof-of-concepttrial. DepressAnxiety. 2020;37(3):224–33.

43.KashaniP,YouseﬁanS,AminiA,etal.Theeffectofintravenous ketamineinsuicidalideationofemergencydepartmentpatients. Emerg(Tehran). 2014;2(1):36–9.

44.Grupp-PhelanJ,StevensJ,BoydS,etal.Effectofamotivational interviewing-basedinterventiononinitiationofmentalhealthtreatment andmentalhealthafteranemergencydepartmentvisitamong suicidaladolescents:arandomizedclinicaltrial. JAMANetwOpen. 2019;2(12):e1917941.

45.KingCA,GipsonPY,HorwitzAG,etal.Teenoptionsforchange:an interventionforyoungemergencypatientswhoscreenpositivefor suiciderisk. PsychiatrServ. 2015;66(1):97–100.

46.KawanishiC,ArugaT,IshizukaN,etal.Assertivecasemanagement versusenhancedusualcareforpeoplewithmentalhealthproblemswho hadattemptedsuicideandwereadmittedtohospitalemergency departmentsinJapan(ACTION-J):amulticentre,randomisedcontrolled trial. LancetPsychiatry. 2014;1(3):193–201.

47.AlonzoD.Suicidalindividualsandmentalhealthtreatment: anovelapproachtoengagement. CommunityMentHealthJ. 2016;52(5):527–33.

48.XiongJ,LipsitzO,Chen-LiD,etal.Theacuteantisuicidaleffectsof single-doseintravenousketamineandintranasalesketaminein individualswithmajordepressionandbipolardisorders:asystematic reviewandmeta-analysis. JPsychiatrRes. 2021;134:57–68.

49.BrendleM,RobisonR,MaloneDC.Cost-effectivenessofesketamine nasalspraycomparedtointravenousketamineforpatientswith treatment-resistantdepressionintheUSutilizingclinicaltrialefﬁcacy andreal-worldeffectivenessestimates. JAffectDisord. 2022;319:388–96.

50.RabascoA,AriasS,BenzMB,etal.Longitudinalriskofsuicide outcomesinpeoplewithseverementalillnessfollowinganemergency departmentvisitandtheeffectsofsuicidepreventiontreatment. JAffectDisord. 2024;347:477–85.

ORIGINAL RESEARCH

Buprenorphine-NaloxoneforOpioidUseDisorder: ReductioninMortalityandIncreasedRemission

KrishnaK.Paul,BS*

ChristianG.Frey,BS*

StanleyTroung,BA*

LauravitaQ.Paglicawan,BS*

KathrynA.Cunningham,PhD†

T.PrestonHill,MD*

LaurenG.Bothwell*

GeorgiyGolovko,PhD†

YeoshinaPillay,MD‡

DietrichJehle,MD*

SectionEditor:BradBobrin,MD

*DepartmentofEmergencyMedicine,UniversityofTexasMedicalBranch, Galveston,Texas

† DepartmentofPharmacologyandToxicology,UniversityofTexasMedical Branch,Galveston,Texas

‡ DepartmentofEmergencyMedicine,MichiganStateUniversity, EastLansing,Michigan

Submissionhistory:SubmittedDecember15,2023;RevisionreceivedJuly5,2024;AcceptedJuly11,2024

ElectronicallypublishedSeptember6,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18569

Introduction: Asfentanylhasbecomemorereadilyavailable,opioid-relatedmorbidityandmortalityin theUnitedStateshasincreaseddramatically.Preliminarystudiessuggestthathigh-afﬁnity,partialmuopioidreceptoragonistssuchasthecombinationproductbuprenorphine-naloxonemayreducemortality fromoverdoseandpromoteremission.Withtheescalatingprevalenceofopioidusedisorder(OUD),itis essentialtoevaluatetheeffectivenessofopioidagonistslikebuprenorphine-naloxone.Thisstudy examinesmortalityandremissionratesforOUDpatientsprescribedbuprenorphine-naloxoneto determinetheef ﬁcacyofthistreatmenttowardtheseoutcomes.

Methods: WecarriedoutaretrospectiveanalysisusingtheUSCollaborativeNetworkdatabasein TriNetX,examiningde-identiﬁedmedicalrecordsfromnearly92millionpatientsacross56healthcare organizations.ThestudyspannedtheyearsfromJanuary1,2017–May13,2022.Cohort1included OUDpatientswhobeganbuprenorphine-naloxonetreatmentwithinone-yearpost-diagnosis,while Cohort2,thecontrolgroup,consistedofOUDpatientswhowerenotadministeredbuprenorphine.The studymeasuredmortalityandremissionrateswithinayearoftheindexevent,incorporatingpropensity scorematchingforage,gender,andrace/ethnicity.

Results: Priortopropensitymatching,weidentifiedatotalof221,967patientswithOUD.Following exclusions,61,656patientstreatedwithbuprenorphine-naloxoneshowed34%fewerdeathswithinone yearofdiagnosiscomparedto159,061patientswhodidnotreceivebuprenorphine(2.6%vs4.0%; relativerisk[RR]0.661;95%confidenceinterval[CI]0.627–0.698; P < 0.001).Theremissionratewas approximately1.9timeshigherinthebuprenorphine-naloxonegroupcomparedtothecontrolgroup (18.8%vs10.1%;RR1.862;95%CI1.812–1.914; P < 0.001).Afterpropensitymatching,theeffecton mortalitydecreasedbutremainedstatisticallysignificant(2.6%vs3.0%;RR0.868;95%CI0.813–0.927; P < 0.001)andtheremissionrateremainedconsistent(18.8%vs10.4%;RR1.812;95%CI 1.750–1.876; P < 0.001).Numberneededtotreatforbenefitwas249fordeathand12forremission.

Conclusion: Buprenorphine-naloxonewasassociatedwithsigniﬁcantlyreducedmortalityand increasedremissionratesforpatientswithopioidusedisorderandshouldbeusedasaprimary treatment.Therecognitionandimplementationoftreatmentoptionslikebuprenorphine-naloxoneisvital inalleviatingtheimpactofOUD.[WestJEmergMed.2024;25(6)869–874.]

INTRODUCTION

Background

Theincidenceandprevalenceofopioidusedisorder (OUD)isincreasingbothintheUnitedStatesandglobally. Therecentproliferationoftheillicitdrugfentanylhasonly intensiﬁedthepotentialforopioidabuseandraisedthe mortalityrate.1 Opioid-relateddeathshavesurgednearly four-foldsince1999,andmortalityrateshavecontinuedtheir upwardtrajectorywiththeadventofsyntheticopioids.2–4 Fentanylhasnotablybecometheprimarydriverofdrugrelatedoverdoses,withanalmost7.5-foldincreasein overdose-relateddeathsfrom2015to2021.2 However,there isarangeofUSFoodandDrugAdministration-approved medicationsavailableforOUDthatcanreduceoverdoserelatedmortalityandpromoteremission.

Partialopioidagonistmedications,suchasbuprenorphine, whichcanbefoundasamono-productformorcombinedwith naloxone(suchasbrandnameSuboxone),havedemonstrated efﬁcacyinreducingtheriskofoverdose-relateddeath comparedtopureopioidantagonistslikenaltrexone.5–8 Buprenorphineisapartialopioidagonistthatfunctionsby bindingtothemu-opioidreceptorwithparticularlyhigh afﬁnitywhenthereceptorisempty,therebyblockingthe bindingofotheropioidswithabusepotentialwhilealso alleviatingwithdrawalsymptomsandcravings.9 However, whenthereceptorisoccupied,buprenorphinedislodgesthe opioidfromthereceptor,thusprecipitatingwithdrawal.

Buprenorphine-naloxoneisdistinctiveinthatwhengiven asacombination,itcanbeusedtomitigatepossible inappropriateusageofbuprenorphinealone.When administeredsublingually,thenaloxonecomponenthaslittle tonoeffectduetohigh ﬁrst-passhepaticmetabolism. However,ifbuprenorphine-naloxoneweretobeused intravenouslyorintranasally,thenaloxonecanprecipitate withdrawalaswellasdiminishanyperceivedeuphoria.9 This uniquecharacteristicofbuprenorphine-naloxonegreatly enhancesitspotentialtoreduceopioid-relatedoverdoses.A studyconductedovera22-yearperiod,publishedin2020, revealedthattherelativeriskofoverdose-relateddeathwas upto3.2timeshigherintheabsenceofopioidagonist therapywithbuprenorphine.8 Moreover,priorresearch demonstratespotentialindecreasingfutureillicitopioiduse followinginitiationoftreatmentwithbuprenorphinenaloxone.1 A2021studyfoundthatbuprenorphinenaloxonetherapywasassociatedwithsigniﬁcantlylower oddsoffentanylexposureovertimecomparedtomethadone orslow-releasemorphinetreatment.1 Thispositions buprenorphine-naloxoneasauniquelyeffectivecombination ofdrugsusedinthetreatmentofOUD.

Importance

AstheprevalenceofOUDcontinuestorise,itisessential toidentifyatreatmentthateffectivelyreducesoverdoserelatedmortalityandincreasesremissionrates.

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Opioid-relatedmorbidityandmortalityhas increasedintheUS,alongwithfentanyl use.Studiesshowbuprenorphine-naloxoneto beaneffectivetreatmentforopioiduse disorder(OUD).

Whatwastheresearchquestion?

Howarethemortalityandremissionratesof OUDpatientsaffectedbytheprescriptionof buprenorphine-naloxone?

Whatwasthemajor ﬁndingofthestudy?

Patientsprescribedbuprenorphine-naloxone had34%fewerdeaths(CI0.63 – 0.70; P < 0.001)and1.9timesmoreremission (CI1.81 – 1.91;P < 0.001).

Howdoesthisimprovepopulationhealth?

AstheincidenceofOUDandtheavailability offentanylincreases,healthcareinterventions areessential.Buprenorphine-naloxoneisan effectivetreatmentoption.

GoalsofInvestigation

Inthisstudyweusedelectronichealthrecords(EHR) fromtheUnitedStatesCollaborativeNetworkinTriNetXto performananalysiscomparingpatientsprescribed buprenorphine-naloxonewithinoneyearoftheirOUD diagnosiswiththosewhodidnotuseopioidagonisttherapy. Theinvestigationexaminedmortalityandremissionrates betweenthesetwopatientcohorts.

METHODS StudyDesign

TriNetXisaglobalcollaborativenetworkconsistingof de-identifiedpatientEHRdatafromaroundtheworld.All cohortandoutcomedefinitionswerebasedonthe InternationalClassificationofDiseases,10th Rev,Procedure CodingSystem,ClinicalModification(ICD-10-CM)entered intothehealthrecordsystems.Weidentifiedmedications prescribedusingRxNormcodes(Table1).RxNormisa standardizednomenclatureforclinicaldrugsanddrug deliverydevicesintheUS,developedandmaintainedbythe NationalLibraryofMedicine.UsingtheUSCollaborative NetworkofTriNetX,whichcontainsapproximately92 millionpatientsfrom56healthcareorganizations(HCO)in theUS,weestablishedtwocohorts.

Table1. InternationalClassiﬁcationofDiseasesandRxNormcodesforbuprenorphine-naloxoneandopioidusedisorder. TypeNameCodingsystemCode DiagnosisOpioiddependence,uncomplicatedICD-10-CMF11.20 DiagnosisOpioiddependence,inremissionICD-10-CMF11.21 DiagnosisOpioidabuse,inremissionICD-10-CMF11.11 DiagnosisOtherlong-term(current)drugtherapyICD-10-CMZ79.899 MedicationBuprenorphineRxNorm1819 MedicationNaloxoneRxNorm7242

ICD-10-CM,InternationalClassiﬁcationofDiseases;10th Rev,ClinicalModiﬁcation; OUD,opioidusedisorder.

CohortDeﬁnition

Cohort1includedallpatientsdiagnosedwith uncomplicatedopioiddependence(ICD-10-CM:F11.20) whoweregivenaprescriptionofbuprenorphine-naloxone (buprenorphine-RxNorm:1819 + naloxone -RxNorm:7242)withinoneyearofanyF11.20diagnosis inputintothehealthrecordsystem.Cohort2(control) consistedofpatientsdiagnosedwithuncomplicatedopioid dependencewithnocurrentorpriorprescriptionof buprenorphine.Thiswasconsideredthestartingeventor “indexevent.” Individualswhohadneverbeenprescribed buprenorphinewereselectedasthecontrolgroup,because deﬁningthecontrolgroupasthosenothaving buprenorphine-naloxonelistedintheTriNetXdatabase couldinadvertentlyhaveexcludedpatientswhohadbeen treatedsolelywithnaloxonefollowinganoverdoseepisode. Forbothcohorts,thetimewindowwasestablishedbetween January1,2017–May13,2022.Thistimeframewaschosen asbuprenorphine-naloxonebecamewidelyaccessibleand prescribedfrom2017onward,andtheenddateof2022 ensuredtherewasatleastoneyearbetweenthedatethe patientwasseenforOUDandoneyearoffollow-up foroutcomes.

Outcomes

Thetwooutcomesmeasuredinthesecohortswere mortalityandremission.Remissionwasdefinedbythe diagnosisofremissionfromopioiddependence(ICD-10CM:F11.21),remissionfromopioidabuse(ICD-10-CM: F11.11),ortheuseofotherlong-termdrugtherapy(ICD-10CM:Z79.899).Thetimewindowfortheseoutcomesranged fromonedaytooneyearfollowingtheindexevent,which wasdefinedastheusageofbuprenorphine-naloxoneornonusageofbuprenorphineforOUD.Weexcludedfromthe studypatientswithremissionpriortotheindexevent. MortalitydatawithintheTriNetXplatformwasobtained fromEHRdataandHCOs,inconjunctionwiththenational deathregistries.Thereispotentialformisseddeathevents whenapatientistreatedatanHCOnotaffiliatedwiththe TriNetXnetworkandsubsequentlyexperiencesafatal outcomeoutsidethisnetwork.However,thisrepresentsonly

aminorissue,ascurrently,94%ofHCOswithintheTriNetX networkarealsolinkedtotheUSdeathregistries.This percentageissteadilyincreasingasmoreHCOscontinueto belinkedwiththeregistries.

SecondaryAnalysisonSocioeconomicStatus

Weperformedasecondaryanalysistoevaluatetheimpact ofsocioeconomicstatusontheprescriptionof buprenorphine-naloxoneandoutcomesinOUDpatients. ThepresenceoftheICD-10-codesZ56.0(Unemployment, unspeciﬁed)orZ59(Problemsrelatedtohousingand economiccircumstances)wasusedasamarkerforprior historyoflowersocioeconomicstatusandwasextracted fromeachcohort.

Apost-hocanalysiswasperformedfromJune2,2004to June2,2023,inwhichweevaluatedOUDpatientswhowere prescribedsuboxonewithinonemonthoftheOUD diagnosisandexcludedthoseonmethadoneornaltrexone. TheOUDdeﬁnitionwasexpandedtoincludeadditional ICD09/10-codesassociatedwithopioidabuseandopioiduse aswell(SupplementaryTable1).Propensitymatchingwas slightlymorerobust,includingadditionalcovariatessuchas socialdeterminants,disordersrelatedtodrugsofabuse,and nicotine(SupplementaryTable2).

StatisticalAnalysis

A1:1propensityscorematchwasproducedviaTriNetX, usinglogisticandlinearregression.Weusedgreedynearestneighbormatchingwithatoleranceof0.1andadifference betweenpropensityscoreslessthanorequalto0.1.10 Propensitymatchingwasperformedfordemographics includingageattheindexevent,race,ethnicity,andgender, usingtheBalanceCohortstoolinTriNetX.Alldemographic datawasself-reportedbypatientsandrecordedbyHCOsto HL7administrativestandards.WeusedtheMeasureof AssociationAnalysistoolinTriNetXtocalculateriskratio (RR),95%conﬁdenceinterval(CI),and P -values(P)for outcomecomparisonsthroughunivariateanalysis.We calculatedthenumberneededtotreatforbeneﬁt(NNTB) manuallyforeachoutcome.Patientswhomettheoutcome priortothevisitwereexcludedfromtheirrespectivecohorts

fortheoutcomeanalysistoensurethatpatientswhohadthe outcomepriortotheindexeventwereexcluded.The final datawasobtainedandanalysisconductedonMay13,2023. Statisticalsignificancewasdeterminedatatwo-sidedalpha <0.05.Becauseweusedde-identifieddatafromTriNetX,this studywasdeterminedtobeexemptbytheUniversityof TexasMedicalBranch(UTMB)IRB.TheUTMBIRB determinedthatthisprojectdidnotinvolveinterventionor interactionwithhumansubjects,andthedatawasdeidentifiedperthede-identificationstandarddefinedin Section§164.514(a)oftheHIPAAPrivacyRule.This formaldeterminationbyaqualifiedexpertrefreshedon December2020.10

RESULTS

Inthisanalysis,therewereatotalof221,967patientsprior topropensitymatching,with62,041patientsinCohort1and 159,926patientsinCohort2.Regardingtheoutcomeof mortality,afterexclusions,Cohort1consistedof61,656 patients,andCohort2comprised159,061patients.Forthe outcomeofremission,therewereamuchlargernumberof patientsexcludedbecauseofpriorhistoryofremission. Cohort1included37,199patients,andCohort2contained 110,726patients.Afterpropensitymatching,Cohort1 maintainedthesamenumberofpatientsforbothoutcomes. Whenpropensitymatchingwasapplied,Cohort2included 61,746and44,284patientsfortheoutcomesofdeathand remission,respectively,afterexclusions.(Table2).

Afterpropensitymatching,OUDpatientsexperienced 13%lessdeaths(2.6%vs3.0%,RR0.87,95%CI0.81–0.93, P < 0.001)and81%greaterremissionsrates(18.8%vs10.4%,

RR1.81,95%CI1.75–1.88, P < 0.001)intheyearfollowing prescriptionofbuprenorphine-naloxonewhencomparedto thosewhowerenotprescribedbuprenorphine.Before propensitymatching,trendsweresimilar;however,theeffect ofmortalitywasmorepronounced(RR0.66)(Table3).The NNTBwas249fordeathand12forremissionwithinone yearofindexevent.

Thesecondaryanalysisfortheimpactofsocioeconomic statusshowedpatientsprescribedbuprenorphine-naloxone weremorelikelytohaveahistoryofunemploymentthan thosenotprescribedbuprenorphine(6.36%vs2.54%, P < 0.001).Similarly,patientsprescribedbuprenorphinenaloxoneweremorelikelytohavehadproblemsrelatedto housingandeconomiccircumstancesthanthosenot prescribedbuprenorphine(13.93%vs7.82%, P < 0.001).The posthocanalysisshowedthatafterpropensitymatchingfor socialdeterminantsofhealthandothersubstance-related disorders,therelativeriskofdeathinoneyearwas0.78(95% CI0.74–0.83; P < 0.001)andremissionwas2.15(95%CI 2.09–2.22; P < 0.001)forthepatientsprescribed buprenorphine-naloxonewithinonemonthofOUD diagnosis(SupplementaryTable3).Trendsweresimilar beforepropensitymatching.NNTBwas137fordeathand 12forremission.

DISCUSSION

Thismulticenter,retrospectivestudyhasdemonstrated thatbuprenorphine-naloxoneusewasassocaitedwith signiﬁcantlylowermortalityratesandhigherremissionrates incomparisontonotreatmentinpatientswithOUD.The utilizationoftheUnitedStatesCollaborativeNetworkin

Table2. DemographicsbeforeandafterpropensityscorematchingwithCohort1buprenorphine-naloxoneandCohort2opioiduse disordercontrols.

AI/AN564(0.91%)807(0.51%) <0.0010.05564(0.91%)412(0.66%) <0.0010.03

NHPI42(0.07%)140(0.09%) =0.140.0142(0.07%)15(0.02%) <0.0010.02

Unknownethnicity10,945(17.6%)49,014(30.6%) <0.0010.3110,945(17.6%)10,922(17.6%) =

NotHispanicorLatino47,161(76.0%)101,782(63.6%) <0.0010.2747,161(76.0%)47,249(76.2%) =0.56 <0.01 HispanicorLatino3,935(6.34%)9,130(5.71%) <0.0010.033,935(6.34%)3,870(6.24%) =0.45 <0.01

Black6,999(11.3%)22,711(14.2%) <0.0010.096,999(11.3%)6,993(11.3%) =0.96 <0.01

Asian193(0.31%)810(0.51%) <0.0010.03193(0.31%)169(0.27%) =0.21 <0.01

Unknownrace7,873(12.7%)24,643(15.4%) <0.0010.087,873(12.7%)7,916(12.8%) =0.71 <0.01

OUD,opioidusedisorder; AI/AN,AmericanIndianorAlaskanNative; NHPI,NativeHawaiianorotherPaciﬁcIslander.

Table3. OutcomesbeforeandafterpropensityscorematchingwithCohort1buprenorphine-naloxoneandCohort2opioiduse disordercontrols.

Cohort1Cohort2RR(95%CI) P-value

MortalitybeforePSM1,628(2.64%)6,352(3.99%)0.66(0.63–0.70)P < 0.001

MortalityafterPSM1,628(2.64%)1,878(3.04%)0.87(0.81–0.93)P < 0.001

RemissionbeforePSM6,984(18.78%)11,163(10.08%)1.86(1.81–1.91)P < 0.001

RemissionafterPSM6,984(18.78%)4,589(10.36%)1.81(1.75–1.88)P < 0.001

OUD,opioidusedisorder; PSM,propensityscorematching; RR,relativerisk; CI,conﬁdenceinterval.

TriNetXprovidesthelargestsamplesizeinthecurrent literature(221,967beforepropensitymatching)comparing buprenorphine-naloxonetonointervention.Thislargedatasetapproachhasmitigatedpotentialconfounding variables,includingsocialdeterminants,thatmighthave inﬂuencedthe ﬁndingsofpreviousresearch.

The ﬁndingsofourstudyareconsistentwithother literatureexaminingoutcomesofmedicationsforOUD.The PrescriptionOpioidAddictionTreatmentStudytrial demonstratedmoresuccessfuloutcomes(abstinenceornearabstinencefromopioids)inprescriptionopioidusers maintainedonbuprenorphine-naloxonethaninthosewho weretaperedoffbuprenorphine-naloxone.11 Only7%of studyparticipantswhoweretaperedoffbuprenorphinenaloxoneachievedsuccessfuloutcomescomparedto49% whoweremaintainedonbuprenorphine-naloxone.11 This populationofprescriptionopioidusersisslightlydifferent fromourpopulationofpatientswithOUD.Otherstudies havecomparedtheuseofbuprenorphine-naloxonewith othermethods,suchasextended-releasenaltrexone,or methadone,andhavefoundeitherinterventiontobeequally safeandeffective.12–14 Patientswereevenfoundtohave greatershort-termsuccessintreatmentwithbuprenorphinenaloxonecomparedtoclonidineinasmall,multicenter, randomizedtrial.15

Anotherstudyconsiderednotreatment,inpatient detoxiﬁcation,behavioralhealth,buprenorphineor methadone,naltrexone,andnon-intensivebehavioralhealth interventionsinindividualswithOUD.Onlytreatmentwith buprenorphineormethadonedemonstratedasigniﬁcantly reducedriskofoverdoseandopioid-relatedacutecareat 3-and12-monthfollow-ups.16 Ourstudyfurtherstrengthens theargumentformedicationuseintreatingOUDand demonstratestheeffectivenessofbuprenorphine-naloxone.

The ﬁndingsofthisstudycarrysigniﬁcantimplicationsfor theacutemanagementofOUD.Withtheglobalrisein OUD,exacerbatedbytheincreasedavailabilityoffentanyl, mitigatingmortalityratesforindividualswithOUDremains amajorpublichealthchallenge.17 Theseindividualsmay requiremedicalinterventionsforthetreatmentof complicationsofOUDsuchasautonomicinstability, hypoxia,endocarditis,ischemia,orcardiacarrest.18

Proactivepharmacologicmanagementisakeycomponentin preventingthelife-threateningconsequencesofOUD.

LIMITATIONS

Itisimportanttoacknowledgetheinherentlimitationsof aretrospectivestudy,especiallyinthatcausationcannotbe established.Thisstudyincludedasampleof221,967patients from56HCOsbeforepropensitymatching,whichismore extensivethanpreviousstudies,therebyprovidingincreased statisticalpower.Thelargesamplesizeanduseofpropensity matchinghelpmitigatesomeoftheselimitations.Propensity matchingwasperformedforcovariatesincludingage, gender,race,andethnicity,buttheremaybeother unaccountedvariablesthatcouldhaveinfluencedthe mortalityratesorremissioninthisgroup.DuetodeidentificationandprivacypoliciesinherenttotheTriNetX database,granularandnon-codabledatasuchassocial determinantsarelimited,whilegeographicidentifierssuchas ZIPcodesareunobtainable.Moreover,thisstudydidnot takeintoaccountcomorbidmedicalconditionsor psychiatricillnesses,whichmightalsohaveimpacted theoutcomes.

Thedatabaserecordsprescriptionsinthehealthrecords butdoesnotreportthedosagesofbuprenorphine-naloxone, exacttimingofthisadjuncttherapy,orpatientadherenceto regimen.Thisomissionissigniﬁcantbecausedifferent medicationdosagesandcompliancemayaffecttheefﬁcacy ofbuprenorphine-naloxone.Forthecontrolgroup, alternativemethodsformanagingOUDmayhavebeen presentbutwerenotconsideredintheanalysis.Additionally, withoutknowinghowlongapatientwasdrugfree,wecould notruleoutpotentialrelapsesasacausefortheindexvisitin theexcludedgroup.Furthermore,theparameterofthedata collectedrestrictsthegeneralizationofthestudyfor internationalpopulationsandlong-termresultsbeyond ayear.

Finally,thestudyevaluatedtheparametersofmortality rateandremission,butothermeasuresweren’tincluded.We didnotconsiderqualityoflifemeasures,whichcouldprovide amorecomprehensiveunderstandingoftheeffectsof buprenorphine-naloxoneonlifestyleinthosewithOUD. Futureresearchshouldfocusonotherlifeoutcomemeasures,

socialdeterminants,andthelong-termimpactsof buprenorphine-naloxoneontheindividual.

CONCLUSION

Thismulticenterretrospectiveanalysisshowsthat buprenorphine-naloxoneuseisassociatedwithsignificantly improvedmortalityratescomparedtonointerventionin patientswithopioidusedisorder.Furthermore,thestudy highlightsanassociationwithhigherremissionratesinthis population.Whilethese findings,alongwithprevious studies,suggestthatbuprenorphine-naloxoneisaneffective treatmentoptionforOUD,furtherprospectivestudies comparingtoothertreatmentmodalitiesshouldbe consideredtoconfirmefficacy.

AddressforCorrespondence:KrishnaK.Paul,BS,Research Coordinator,DepartmentofEmergencyMedicine,Universityof TexasMedicalBranch,301UniversityBlvd.,Galveston,1173. Email: kkpaul@utmb.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Thisresearchwassupportedbythe InstituteforTranslationalSciencesatTheUniversityofTexas MedicalBranch,supportedinpartbyaClinicalandTranslational ScienceAward(UL1TR001439)fromtheNationalCenterfor AdvancingTranslationalSciencesattheNationalInstitutesofHealth (NIH).Thecontentissolelytheresponsibilityoftheauthorsanddoes notnecessarilyrepresenttheofficialviewsoftheNIH.Thereareno otherconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.YoungS,HayashiK,GrantC,etal.Buprenorphine/naloxoneassociated withareducedoddsoffentanylexposureamongacohortofpeoplewho usedrugsinVancouver,Canada. DrugAlcoholDepend. 2021;228:109006.

2.MattsonCL,TanzLJ,QuinnK,etal.Trendsandgeographicpatternsin drugandsyntheticopioidoverdosedeaths-UnitedStates,2013–2019. MorbMortalWklyRep. 2021;70(6):202–7.

3.MartinM,HeckerJ,ClarkR,etal.Chinawhiteepidemic:aneastern UnitedStatesemergencydepartmentexperience. AnnEmergMed. 1991;20(2):158–64.

4.CentersforDiseaseControlandPrevention.Understandingtheopioid overdoseepidemic.2024.Availableat: https://www.cdc.gov/

overdose-prevention/about/understanding-the-opioid-overdoseepidemic.html.AccessedJune24,2024.

5.LarochelleMR,BernsonD,LandT,etal.Medicationforopioiduse disorderafternonfatalopioidoverdoseandassociationwithmortality: acohortstudy. AnnInternMed. 2018;169(3):137–45.

6.WattsBV,GottliebDJ,RibletNB,etal.Associationofmedication treatmentforopioidusedisorderwithsuicidemortality. AmJPsychiatry. 2022;179(4):298–304.

7.BurnsM,TangL,ChangCH,etal.Durationofmedicationtreatment foropioid-usedisorderandriskofoverdoseamongMedicaidenrollees in11states:aretrospectivecohortstudy. Addiction. 2022;117(12):3079–88.

8.PearceLA,MinJE,PiskeM,etal.Opioidagonisttreatmentandriskof mortalityduringopioidoverdosepublichealthemergency:population basedretrospectivecohortstudy. BMJ. 2020;368:m772.

9.VelanderJR.Suboxone:rationale,science,misconceptions. OchsnerJ. 2018;18(1):23–9.

10.MurphyLR,HillTP,PaulK,etal.Tenecteplaseversusalteplasefor acutestroke:mortalityandbleedingcomplications. AnnEmergMed. 2023;82(6):720–8.

11.WeissRDandRaoV.Theprescriptionopioidaddictiontreatment study:whathavewelearned. DrugAlcoholDepend. 2017;173Suppl1(Suppl1):S48–54.

12.LeeJD,NunesEVJr,NovoP,etal.Comparativeeffectivenessof extended-releasenaltrexoneversusbuprenorphine-naloxoneforopioid relapseprevention(X:BOT):amulticentre,open-label,randomised controlledtrial. Lancet. 2018;391(10118):309–18.

13.NeumannAM,BlondellRD,HoopsickRA,etal.Randomizedclinicaltrial comparingbuprenorphine/naloxoneandmethadoneforthetreatment ofpatientswithfailedbacksurgerysyndromeandopioidaddiction. JAddictDis. 2020;38(1):33–41.

14.D’OnofrioG,O’ConnorPG,PantalonMV,etal.Emergencydepartmentinitiatedbuprenorphine/naloxonetreatmentforopioiddependence:a randomizedclinicaltrial. JAMA. 2015;313(16):1636–44.

15.LingW,AmassL,ShoptawS,etal.Amulti-centerrandomizedtrialof buprenorphine-naloxoneversusclonidineforopioiddetoxiﬁcation: ﬁndingsfromtheNationalInstituteonDrugAbuseClinicalTrials Network. Addiction. 2005;100(8):1090–100.[publishedcorrection appearsin Addiction.2006;101(9):1374].

16.WakemanSE,LarochelleMR,AmeliO,etal.Comparativeeffectiveness ofdifferenttreatmentpathwaysforopioidusedisorder. JAMANetw Open. 2020;3(2):e1920622.

17.RitchieH,ArriagadaP,RoserM.Opioids,cocaine,cannabisandother illicitdrugs.Ourworldindata.2022.Availableat: https://ourworldindata. org/illicit-drug-use#introduction.AccessedJuly30,2023.

18.LavonasEJandDezfulianC.Impactoftheopioidepidemic. CritCare Clin. 2020;36(4):753–69.

ORIGINAL RESEARCH

OpioidTreatmentIsAssociatedwithRecurrentHealthcareVisits, IncreasedSideEffects,andPain

CarolineE.Freiermuth,MD,MS*†

JennyA.Foster,BS‡

PratikManandhar,MS§

EvangelineArulraja,MD∥

AlaattinErkanli,PhD§

CharlesV.Pollack,MD,MA¶

StephanieA.Eucker,MD,PhD∥

*UniversityofCincinnati,DepartmentofEmergencyMedicine,Cincinnati,Ohio

† UniversityofCincinnatiCollegeofMedicine,CenterforAddictionResearch, Cincinnati,Ohio

‡ DukeUniversitySchoolofMedicine,Durham,NorthCarolina

§ DukeUniversity,DepartmentofBiostatisticsandBioinformatics, Durham,NorthCarolina

∥ DukeUniversity,DepartmentofEmergencyMedicine,Durham,NorthCarolina

¶ UniversityofMississippiMedicalCenter,DepartmentofEmergencyMedicine, Jackson,Mississippi

SectionEditor:PatrickMeloy,MD

Submissionhistory:SubmittedJune22,2024;RevisionreceivedJuly16,2024;AcceptedJuly16,2024

ElectronicallypublishedSeptember24,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem

DOI: 10.5811/westjem.18380

Introduction: Painisamajordriverofvisitstotheemergencydepartment(ED).Cliniciansmustconsider notonlytheefﬁcacyoftreatmentoptionsbutalsosubsequenthealthcareutilizationandpatient-centered outcomessuchassideeffectsfromprescribedmedications.Ourgoalinthisstudywastodetermine whethertherewasanassociationbetweenacutepaintreatmentregimen(opioids,intranasal non-steroidalanti-inﬂammatorydrugs[NSAIDs],orboth)andunscheduledhealthcarevisitsfollowing EDdischarge.

Methods: ThisstudywasasecondaryanalysisoftheAcuteManagementofPainfromtheEmergency Department(AMPED)prospective,observationalcohortstudy.WeusedCoxproportionalhazards analysistoassesstherelationshipbetweentreatmentregimenandtimeto ﬁrstunscheduledhealthcare visit.Repeatedmeasureslogisticregressionanalyseswereusedtodeterminetherelationshipbetween treatmentregimenandanyunscheduledvisits,andtoevaluatewhetherthisrelationshipwasmediated bypainseverityand/ormedicationsideeffects.

Results: Of831totalenrolledparticipants,141(16.9%)experiencedanunplannedhealthcarevisit within ﬁvedaysofEDdischarge.AmajorityofthesevisitshappenedonedayaftertheEDvisit.Those whoweretreatedwithintranasalNSAIDsonlywerelesslikelytohaveanunscheduledhealthcarevisit comparedtothosewhoreceivedopioidsonly,withanadjustedoddsratio(AOR)of0.63.Thehigherodds ofunscheduledhealthcarevisitswithopioidsweremediatedbyboththepresenceofsideeffectsand higherpainlevels,withAORsof2.24and1.33,respectively.

Conclusion: Opioidtreatmentforacutepainisassociatedwithincreasedunscheduledhealthcarevisits comparedtothosetreatedwithintranasalketorolac.Thisdifferencecanbeexplainedbyhigherlevelsof ongoingpainandgreatermedicationsideeffects.[WestJEmergMed.2024;25(6)875–882.]

INTRODUCTION

Acutepainisthemostfrequentchiefcomplaintfor emergencydepartment(ED)visitsintheUnitedStates.1 Inadequatetreatmentofpaincontributestotheevolution

fromacutetochronicpain,whichcaninﬂictexcessive personalandeconomicburdensonpatients.2–4 Outsidethe limitednumberofrecommendationsguidingtreatmentofa fewspeciﬁcacuteconditions,suchaslowerbackpain,5 there

arenoevidence-basedguidelinesthataddressthe managementofacutepain,resultinginwidevariability inpractice.6–8

Inparticular,thenumberofopioidprescriptionsincreased duringthelate1990stoearly2010s,yetopioidswerenot foundtocorrelatewithimprovedpain-relatedoutcomesor patientsatisfaction.9–11 Moreover,theadventofthecurrent opioidcrisisandriseinopioid-relateddeathsmotivatednew guidelinesdiscouragingtheuseofopioidsforacuteand chronicpainmanagement.12–14 Whileopioid-prescribing ratessubsequentlybegantodeclinefrom21.5%ofpatients dischargedfromanEDin2011to8.1%in2019,thisrate remainscomparativelyhigh.15 Inaddition,opioidreductions havenotappearedtobematchedbysufﬁcientincreasesin non-opioidprescribing.16

Inparticular,ongoingconcernsaboutthesideeffectsof alternativestoopioidssuchasnon-steroidalantiinflammatorydrugs(NSAID)havehinderedtheiruse.17 However,datashowsthateveninhigh-riskpatients,those prescribedNSAIDsdidnotexperiencehigherratesofserious adverseevents.18,19 Specifically,intranasalketorolachas beenfoundtobeaneffective,well-tolerated,andsatisfactory medicationforadultspresentingtotheEDwithacutepain, withonlyminorsideeffectssuchasnasalburningthat resolveswithinseveralminutes.20,21 Bycomparison,the adverseeffectsofopioidsoutsideopioidusedisorder-related morbidityhavereceivedrelativelylittleattention,withonly recentdatasuggestinggreatersideeffectswithopioidsthan withNSAIDs.19,22 Thus,morestudiesareneededtobetter understandtherisk-benefitcomparisonbetweenopioidsand NSAIDsforpaintreatment.

Meanwhile,therehasbeenafocusonpreventingrecurrent visitstotheED23–25 forpainaswellasothercomplaints. Revisitssubstantiallyimpactboththepatientandhealthcare systemandoftenreﬂectunmetpatientneeds.Recentdata suggestsopioidprescriptionsintheoutpatientsettingare associatedwithanincreasednumberofsubsequent healthcarevisits.26–29 Unplannedvisitscanresultinmissed daysofwork,30–32 disruptingthedailylivesofpatients,as wellasincreasedcosts,oftenhigherwiththerepeatthanthe initialEDvisit.33 Failuretoaddresstheinitialpresenting problemoritssequelae,aswellaslackofcommunication withpatientsregardingthetherapeuticplanand recommendedfollow-up,contributetoEDrecidivism.32,34,35 Emergencyphysiciansmustunderstandhowthetreatment theyprovideintheEDmayaffectdownstreamhealthcare utilizationtofullyconsidertheimpactoftheircareonboth patientsandhealthcaresystems.

Inthissecondaryanalysis,wesoughttodetermine whethertherewasanassociationbetweentreatmentregimen (opioidonly,intranasalNSAIDonly,orintranasalNSAID + opioid)followinganEDvisitforacutepainand subsequentdailyhealthcareutilization.Wefurther characterizedthisassociationbywhethermedicationside

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Widevariabilityinpaintreatmentinacute caresettingsisdueinparttolimitedevidence onpatientexperiencesrelatedtochoice ofmedication.

Whatwastheresearchquestion?

Whatistheassociationbetweenchoiceof acutepaintreatmentandunscheduled healthcarevisitsafterEDdischarge?

Whatwasthemajor ﬁndingofthestudy?

Comparedtoopioids,NSAIDshadalower adjustedoddsratioof0.63forpost-ED unscheduledhealthcarevisits(P=0.03,95% CI0.42 – 0.95,33.3%ofpatientsvs.39.7%).

Howdoesthisimprovepopulationhealth?

Ourstudyfurtherunderscoresand contextualizesthewidercomplexityofthe impactoftreatmentdecisions(eg,medication sideeffects,comparativeef ﬁ cacy)on patientoutcomes.

effectsorongoingpainseveritymediatedtheeffecton healthcareutilization.

METHODS

StudySettingandPopulation

Thiswasasecondaryanalysisofdataobtainedfroma multisite,prospective,observationalcohortstudy,Acute ManagementofPainfromtheEmergencyDepartment (AMPED).36 Theoriginalstudywasapprovedbythe institutionalreviewboardsatallenrollmentsites.A conveniencesampleofadultparticipants ≥18yearsoldwith acutemusculoskeletalorvisceralpainnotrequiring admissionwererecruitedfrom13EDsintheUSbetween September2012–February2014.Patientswereprescribed oneofthreetreatmentregimensfollowingEDdischarge: NSAID(intranasalketorolac)only;oralopioidonly;or both(intranasalketorolacwithopioidasarescuetherapy). Treatmentregimenwasatthediscretionofthe treatingclinician.

DataCollection

Datacollectionincludedpatientdemographics, employmentstatus,paintype(visceralormusculoskeletal) andlocation,painscores(0–10numericratingscale)atED triageandEDdischarge,medicationsgivenduringtheED

visit,anddischargemedicationregimen.Patientswere contacteddailyforfourdaysfollowingEDdischargefor follow-upoutcomesincludingunplannedhealthcarevisits, medicationuse,dailyhighestandlowestpainscores,adverse eventsandsymptoms,overallqualityoflife,andoverall satisfactionwiththeprescribedpainmedication.Detailsof theunplannedhealthcarevisitswerenotrecorded,including reasonforthevisit.

StatisticalAnalysis

All831patientswereincludedforanalysis.Weused descriptivestatisticstosummarizepatientdemographicsfor thosewithorwithoutunplannedhealthcarevisitsafterED discharge.Categoricalvariableswerereportedas frequencies,andcontinuousvariableswerereportedas medianvalueswithinterquartileranges(25th,75th).Fewer than ﬁvepatientshadmissingdemographicdata.The missingdataoncategoricalvariableswereimputedtothe highestfrequencycategory,andcontinuousvariableswere imputedtomedianvalues.

Weperformedstatisticalanalysestoinvestigatewhether treatmentregimenimpactedthetimetoasubsequent unscheduledhealthcareutilizationfollowingEDdischarge andthenumberofanyunscheduledvisits,andtoevaluate possiblemediatingfactors.Thosewhoreceivedopioid-only treatmentswereusedasthereferencecategoryfortreatment regimen.Weadjustedallmodelsforage,gender,race, primaryEDdiagnosis,initialpainscore,anddischargepain score.Atwo-tailed P -valueof0.05wasusedforstatistical signiﬁcance.WeperformedallanalysesusingSAS9.4(SAS InstituteInc,Cary,NC).

Toevaluatetherelationshipbetweentreatmentregimen andtimeto ﬁrstunscheduledvisitfollowingEDdischarge, weusedaCoxproportionalhazardsmodel.Unscheduled visit(UV)datawascollectedoverthefour-dayfollow-up period.The ﬁrstoccurrenceofanunscheduledvisit(UV = YES)wasusedasthedayoftheunscheduledvisit,regardless ofUV = MISSINGorUV = NOvaluesinpriordays.Ifthe UVdatawasmissingforallfourdays,thepatientwas censoredatday4.WegeneratedaKaplan-Meiercurveof unscheduledvisitwithlog-rank P -value.Toevaluatethe relationshipbetweentreatmentregimenandany unscheduledvisitsfollowingEDdischarge,weuseda repeatedmeasureslogisticregressionmodel.Missingdataon unscheduledvisitsoverthefour-dayperiodwereimputedas UV = NOvalue.Additionally,weperformedasensitivity analysiswhereallmissingdataonunscheduledvisitswere imputedasUV = YESvalue.

Toevaluateformediationoftheassociationbetween treatmentregimenandanyunscheduledvisits,weperformed threeseparaterepeatedmeasureslogisticregressionmodels toassesstheeffectof1)anymedicationsideeffect;2)daily maximumpainscores;or3)bothsideeffectsandpainscores.

RESULTS

ParticipantCharacteristics

Atotalof831participantswereincludedintheoriginal AMPEDstudy.Asparticipantshadtocompleteatleast threeofthefourfollow-upcallstobeincludedintheoriginal analysis,wedidnotexcludeanypatientsfromthissecondary analysis.Thedistributionofparticipantsineachdischarge treatmentregimengroupwasunevenintheoriginalstudy, withcombinationtherapy(intranasalketorolacwithopioid rescue)beingtheleastprescribedregimen.36 Asreportedin theparentstudy,characteristicsofparticipantsineach dischargetreatmentgroupwerecomparable,exceptthat participantsintheopioid-onlygroupwereslightlyolderand morelikelytohaveachronicpainhistory;participantswho hadfracturesorvisceralpainweremorelikelytoreceive opioidsaspartoftheirregimen;andparticipantsinthe intranasalketorolac-onlygrouphadslightlyloweraverage painscoresatinitialEDpresentation.36 Overall,141(16.9%) participantshadatleastoneunscheduledvisitoverthefourdayfollow-uptimeperiod.Baselineanddemographic characteristicsofparticipantsinthissub-analysisareshown byattendancetounscheduledvisits(nonevsatleastonevisit) in Table1.

ModelResults

Acrossalltreatmentregimens,thehighestnumberof unscheduledvisitsoccurredonthe firstdayofthefour-day follow-upperiod(Figure1).However,afteradjustmentof thehazardratios(P > 0.05),theredidnotappeartobea statisticallysignificantassociationbetweenpainmedication prescribedandtimeto firstunscheduledvisit(Table2). Kaplan-Meiercurveofunscheduledvisitwithlog-rank P -valueisprovidedin Figure2

Whenweconsiderallunscheduledvisitsoverthefour-day follow-upperiod,participantswhoweretreatedwith intranasalketorolaconlywerelesslikelytohavean unscheduledhealthcarevisitcomparedtothosewhoreceived opioidsonly,withanadjustedoddsratioof0.63(Table3). Thesensitivityanalysisshowedthatmodeleffectswerenot meaningfullyalteredbymissingdata.Themediationanalysis demonstratedthattheassociationbetweentreatment regimenandunscheduledhealthcarevisitsiscompletely explainedbyboththepresenceofsideeffectsandtheseverity ofpain,withoddsratiosof2.24and1.33,respectively (P =<0.001),andsideeffectscontributingthelarger impact(Table4).

DISCUSSION

Inthisstudyevaluatingtherelationshipbetween treatmentregimens(opioids,rapidlyactingintranasal NSAIDs,orboth)followinganEDvisitforacutepainand subsequentdailyhealthcareutilization,wefoundthat treatmentwithopioidsonlywasassociatedwithincreased subsequentunscheduledhealthcarevisitsduringthe

Table1. BaselinecharacteristicsoftheAMPED*participantsbyunscheduledvisits.

Level

Overall (N = 831)

Age(median,IQR)37.0 (27.0,48.0)

Unscheduledvisit

Nounscheduledvisit (n = 690) Atleastoneunscheduled visit(n = 141)

36.0(27.0,49.0)39.0(31.0,48.0)

Missing,n(%)1(0.1)1(0.1)0(0.0)

Treatmentregimen,n(%)Intranasalketorolaconly353(42.5)306(44.3)47(33.3)

Intranasalketorolac + opioid201(24.2)163(23.6)38(27.0)

OpioidOnly277(33.3)221(32.0)56(39.7)

Missing,n(%)0(0.0)0(0.0)0(0.0)

Employmentstatus,n(%)Other92(11.1)78(11.3)14(9.9)

Unemployed128(15.4)109(15.8)19(13.5)

Parttime129(15.5)103(14.9)26(18.4)

Fulltime481(57.9)399(57.8)82(58.2)

Missing,n(%)1(0.1)1(0.1)0(0.0)

Gender,n(%)Male434(52.2)373(54.1)61(43.3)

Female396(47.7)316(45.8)80(56.7)

Missing,n(%)1(0.1)1(0.1)0(0.0)

Race,n(%)Other169(20.3)148(21.4)21(14.9)

Caucasian251(30.2)193(28.0)58(41.1)

Black407(49.0)345(50.0)62(44.0)

Missing,n(%)4(0.5)4(0.6)0(0.0)

PrimaryEDdiagnosis,n(%)Musculoskeletal707(85.1)593(85.9)114(80.9)

Visceral124(14.9)97(14.1)27(19.1)

Missing,n(%)0(0.0)0(0.0)0(0.0)

PainscoreinED (median,IQR)

Painscoreatdischarge (median,IQR)

8.0(7.0,10.0)8.0(7.0,10.0)8.0(7.0,10.0)

Missing,n(%)1(0.1)1(0.1)0(0.0)

5.0(3.0,8.0)5.0(3.0,8.0)5.0(3.0,8.0)

Missing,n(%)1(0.1)1(0.1)0(0.0)

4-dayanysideeffect,n(%)Yes621(74.7)493(71.4)128(90.8)

No210(25.3)197(28.6)13(9.2)

Missing,n(%)0(0.0)0(0.0)0(0.0)

4-daynausea,n(%)Yes284(34.2)215(31.2)69(48.9)

No542(65.2)470(68.1)72(51.1)

Missing,n(%)5(0.6)5(0.7)0(0.0)

4-dayvomited,n(%)Yes82(9.9)57(8.3)25(17.7)

No744(89.5)628(91.0)116(82.3)

Missing,n(%)5(0.6)5(0.7)0(0.0)

4-dayconstipation,n(%)Yes272(32.7)215(31.2)57(40.4)

No554(66.7)470(68.1)84(59.6)

Missing,n(%)5(0.6)5(0.7)0(0.0)

4-daynasalirritation,n(%)Yes208(25.0)168(24.3)40(28.4)

No618(74.4)517(74.9)101(71.6)

Missing,n(%)5(0.6)5(0.7)0(0.0) (Continuedonnextpage)

Level Overall (N = 831)

Unscheduledvisit

Nounscheduledvisit (n = 690) Atleastoneunscheduled visit(n = 141)

4-dayrash/hives,n(%)Yes33(4.0)17(2.5)16(11.3)

No793(95.4)668(96.8)125(88.7)

Missing,n(%)5(0.6)5(0.7)0(0.0)

4-dayabdominalpain,n(%)Yes167(20.1)122(17.7)45(31.9)

No659(79.3)563(81.6)96(68.1)

Missing,n(%)5(0.6)5(0.7)0(0.0)

4-daydrowsiness,n(%)Yes412(49.6)323(46.8)89(63.1)

No414(49.8)362(52.5)52(36.9)

Missing,n(%)5(0.6)5(0.7)0(0.0)

*AMPED,AcutePainManagementfromtheEmergencyDepartment; IQR,interquartilerange; ED,emergencydepartment.

Figure1. Dailyunscheduledvisitratesbydischarge treatmentregimen.

immediatepost-EDdischargeperiodcomparedtotreatment withintranasalketorolac.Similarassociationshavebeen reportedpreviouslybetweenopioidtreatmentforchronicpain andincreasedfrequencyofEDvisits,opioidprescriptionsand recurrentEDvisitsforlowerbackpain,aswellasopioiduse andincreasedhealthcareutilizationfollowingelectivespine surgery. 26–29 Despitethisrecentattentionontherelationship

betweentreatmentchoiceandhealthcareutilization,thereisa disconcertingpaucityofliteratureinvestigatingpossible mediatingfactors.Inparticular,severalrecentsystematic reviewshaveshownbothlowerpaintreatmentefﬁcacyand higherratesofadverseeventswithopioidscomparedwith NSAIDs,emphasizingtheimportanceofthesefactorswhen weighingtherisksandbeneﬁtsoftreatmentchoiceforacute painmanagement.19,22,37

Ourstudyisanimportantcontributiontotheliterature,as itisthe firsttoourknowledgetodemonstratethatthe associationbetweenpaintreatmentregimensandsubsequent healthcareutilizationpost-EDdischargeismediatedbyboth thepresenceofmedicationsideeffectsandongoing,poorly controlledpain.These findingshighlightadditionalrisksand benefitsrelatedtotheseacutepaintreatmentregimensthat mayimpactpatientoutcomes.Consideringpreviousresearch demonstratingthegreater financialburdenofrecurrent healthcareutilizationonthepatientcomparedtotheinitial visit33 aswellasthereasonspatientsreturntotheED which rangefromlackofsymptomaticimprovementtoadditional questionsorconcerns32,34,35 ourstudyservestofurther contextualizeanddemonstratethewidercomplexityofthe impactoftreatmentdecisionsonpatientoutcomes.Thus, whenchoosingapainmanagementplan,itisimportantto considertheimpactitmayhaveonmultipleaspects Table1. Continued.

Table2. Unadjustedandadjustedhazardratiosoftimeto ﬁrstunscheduledvisit.

OutcomeParameterUnadjustedHRUnadjusted

UnscheduledvisitsOpioidonly(reference)

HR,hazardratios(95%conﬁdenceintervals).

*Adjustedforage,gender,race,primaryEDdiagnosis,painscorepriortotreatmentandpainscoreatdischarge.

Figure2. Kaplan-Meiercurveofunscheduledvisitsfortreatment regimenwithnumberofeventsandatrisk.

ofthepatient’squalityoflifeinconjunctionwith treatmentefﬁcacy.

LIMITATIONS

Studylimitationsincludethatthiswasasecondary analysisofobservationaldataand,therefore,notall potentialconfoundersmayhavebeenmeasuredor controlledfor.Additionally,thestudywasnotrandomized, whichmayhadledtoselectionbiaswithconfounding factorsthatimpactedchoiceoftreatmentregimen,which mayhaveaffectedthelikelihoodofunscheduledvisits.

Further,thedatacollectionoccurredbetweentheyears 2012–2014,whichmaylimittheapplicabilityofthese ﬁ ndings,aspaintreatmentregimensandopioidusehave sinceevolved.

Despitethefactthatoverallrateofopioidprescriptions issuedatdischargefromtheEDdecreasedfrom14.6%in 2017to8.1%in2020,therearestillalargeoverallnumberof peoplewhoarereceivingopioids.15 Inaddition,prescribing patternsshouldnotchangetheunderlyingside-effectprofile ortheirinherentassociationswithunscheduledvisits.Itis alsoimportanttonotethatwhileour findingssupporta strongassociationbetweentreatmentregimenand unscheduledhealthcarevisitsmediatedbythepresenceof sideeffectsanddegreeofpainrelief,thereasonsfor unscheduledvisitswerenotconfirmedbypatientreport. Thus,itispossiblethatpatientssoughthealthcareforreasons unrelatedtotheiracutepain,currenttreatmentplan,orside effects.Finally,thetreatmentregimenwasnotstandardized acrossallgroupsapartfromtheintranasalketorolac;thus, thereportedsideeffectsandcomplicationsmayalsobe relatedtodifferencesinopioidtypeordosing.

CONCLUSION

Outpatienttreatmentwithopioidsonlyforacutepain afteranEDvisitisassociatedwithincreasedunscheduled healthcarevisitscomparedtothosetreatedwithintranasal NSAIDsalone.Thisdifferencecanbeexplainedbyhigher levelsofongoingpainandthepresenceofmedicationside effects.Understandingtheimpactthatpainmedication choicehasonrecurrenthealthcareutilizationandthefactors

Table3. Unadjustedandadjustedoddsratiosofanyunscheduledvisit.

OR,oddsratios(95%conﬁdenceintervals).

*Adjustedforage,gender,race,primaryEDdiagnosis,painscorepriortotreatmentandpainscoreatdischarge.

Table4. Oddsratioformediators – anysideeffectandmaximumpainscore.

thatmediatethisrelationshipaddstothebodyofknowledge regardingrisksandbeneﬁtsofthesetreatments,allowing emergencyphysicianstomakebetterinformeddecisions.

AddressforCorrespondence:StephanieA.Eucker,MD,PhD, DepartmentofEmergencyMedicine,DukeUniversity,2301Erwin Rd.,DUMC3096,Durham,NC27710.Email: stephanie.eucker@ duke.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.CentersforDiseaseControlandPrevention(CDC).NationalHospital AmbulatoryMedicalCareSurvey:2017emergencydepartment summarytables.2017.Availableat: https://www.cdc.gov/nchs/data/ nhamcs/web_tables/2017_ed_web_tables-508.pdf AccessedJune25,2023.

2.FriedmanBW,ConwayJ,CampbellC,etal.Painoneweekafter anemergencydepartmentvisitforacutelowbackpainisassociated withpoorthree-monthoutcomes. AcadEmergMed. 2018;25(10):1138–45.

3.StevansJM,DelittoA,KhojaSS,etal.Riskfactorsassociatedwith transitionfromacutetochroniclowbackpaininUSpatientsseeking primarycare. JAMANetwOpen. 2021;4(2):e2037371.

4.CohenSP,VaseL,HootenWM.Chronicpain:anupdate onburden,bestpractices,andnewadvances. Lancet. 2021;397(10289):2082–97.

5.OliveiraCB,MaherCG,PintoRZ,etal.Clinicalpracticeguidelinesfor themanagementofnon-speciﬁclowbackpaininprimarycare:an updatedoverview. EurSpineJ. 2018;27(11):2791–803.

6.USDepartmentofHealth.PainManagementBestPracticesInteragencyTaskForcereport:updates,gaps,inconsistencies,and recommendations.2019.Availableat: https://www.hhs.gov/sites/ default/ﬁles/pain-mgmt-best-practices-draft-ﬁnal-report-05062019.pdf AccessedJune25,2023.

7.NationalAcademiesofSciences,Engineering,andMedicine. Framing OpioidPrescribingGuidelinesforAcutePain:DevelopingtheEvidence WashingtonDC:NationalAcademiesPress,2019.

8.KeatingLandSmithS.Acutepainintheemergencydepartment:the challenges. RevPain. 2011;5(3):13–7.

9.SitesBD,BeachML,DavisMA.Increasesintheuseofprescription opioidanalgesicsandthelackofimprovementindisabilitymetrics amongusers. RegAnesthPainMed. 2014;39(1):6–12.

10.NewcombP,WilsonM,BaineR,etal.Inﬂuencesonpatientsatisfaction amongpatientswhouseemergencydepartmentsfrequentlyforpainrelatedcomplaints. JEmergNurs. 2017;43(6):553–9.

11.SchwartzTM,TaiM,BabuKM,etal.LackofassociationbetweenPress Ganeyemergencydepartmentpatientsatisfactionscoresand emergencydepartmentadministrationofanalgesicmedications. AnnEmergMed. 2014;64(5):469–81.

12.ManchikantiL,HelmS,FellowsB,etal.OpioidepidemicintheUnited States. PainPhysician. 2012;15(3Suppl):ES9–38.

13.Shaﬁ S,CollinsworthAW,CopelandLA,etal.Associationofopioidrelatedadversedrugeventswithclinicalandcostoutcomesamong surgicalpatientsinalargeintegratedhealthcaredeliverysystem. JAMASurg. 2018;153(8):757–63.

14.DowellD.CDCclinicalpracticeguidelineforprescribingopioidsfor pain UnitedStates,2022. MorbMortalWklyRecommRep 2022;71(3):1–95.

15.SantoLandSchappertSM.Opioidsprescribedtoadultsatdischarge fromemergencydepartments:UnitedStates,2017–2020. NCHSData Brief. 2023;(461):1–8.

16.GoldstickJE,GuyGP,LosbyJL,etal.Patternsinnonopioidpain medicationprescribingafterthereleaseofthe2016guidelinefor prescribingopioidsforchronicpain. JAMANetwOpen. 2022;5(6):e2216475.

17.CastellsagueJ,Riera-GuardiaN,CalingaertB,etal.IndividualNSAIDs anduppergastrointestinalcomplications:asystematicreviewand meta-analysisofobservationalstudies(theSOSproject). DrugSaf. 2012;35(12):1127–46.

18.BouckZ,MecredyGC,IversNM,etal.Frequencyandassociationsof prescriptionnonsteroidalanti-inﬂammatorydruguseamongpatients withamusculoskeletaldisorderandhypertension,heartfailure,or chronickidneydisease. JAMAInternMed. 2018;178(11):1516–25.

19.BusseJW,SadeghiradB,OparinY,etal.Managementofacutepain fromnon-lowback,musculoskeletalinjuries:asystematicreviewand networkmeta-analysisofrandomizedtrials. AnnInternMed. 2020;173(9):730–8.

20.GaulE,BarbourT,NowackiAS,etal.Intranasalketorolacforacute paininadultemergencydepartmentpatients. WestJNursRes. 2022;44(11):1047–56.

21.WattsK,BalzerS,DrumM,etal.Ibuprofenandacetaminophenversus intranasalketorolac(Sprix)inanuntreatedendodonticpainmodel:a randomized,double-blindinvestigation. JEndod. 2019;45(2):94–8.

22.JonesCMP,LinCWC,JamshidiM,etal.Effectivenessofopioid analgesicmedicinesprescribedinoratdischargefromemergency departmentsformusculoskeletalpain:asystematicreviewandmetaanalysis. AnnInternMed. 2022;175(11):1572–81.

23.AtzemaCLandMaclaganLC.Thetransitionofcarebetween emergencydepartmentandprimarycare:ascopingstudy. AcadEmergMed. 2017;24(2):201–15.

24.KripalaniS,TheobaldCN,AnctilB,etal.Reducinghospitalreadmission rates:currentstrategiesandfuturedirections. AnnuRevMed. 2014;65:471–85.

25.WarcholSJ,MonestimeJP,MayerRW,etal.Strategiestoreduce hospitalreadmissionratesinanon-Medicaid-expansionstate. PerspectHealthInfManag. 2019;16(Summer):1a.

26.ShergillY,RiceD,SmythC,etal.Characteristicsoffrequent usersoftheemergencydepartmentwithchronicpain. CJEM. 2020;22(3):350–8.

27.GerlachEB,PlantzMA,SwiatekPR,etal.Thedriversofpersistent opioiduseanditsimpactonhealthcareutilizationafterelectivespine surgery. GlobSpineJ. 2024;14(2):370–9.

28.GinsbergZ,GhaithS,PollockJR,etal.Relationshipbetweenpain managementmodalityandreturnratesforlowerbackpaininthe emergencydepartment. JEmergMed. 2021;61(1):49–54.

29.MegallaM,OgedegbeC,SandersAM,etal.Factorsassociatedwith repeatemergencydepartmentvisitsforlowbackpain. Cureus. 2022;14(2):e21906.

30.EuckerSA,ManandharP,ArulrajaK,etal.Higherdailypainseverity afteremergencydepartmentvisitsisassociatedwithlowerreturn-toworkrates. AmJEmergMed. 2021;45:48–53.

31.LöttersFJB,FoetsM,BurdorfA.Workandhealth,ablindspotincurative healthcare?Apilotstudy. JOccupRehabil. 2011;21(3):304–12.

32.HutchinsonCL,McCloughenA,CurtisK.Incidence,characteristicsand outcomesofpatientsthatreturntoemergencydepartments.An integrativereview. AustralasEmergCare. 2019;22(1):47–68.

33.DusejaR,BardachNS,LinGA,etal.Revisitratesandassociated costsafteranemergencydepartmentencounter:amultistateanalysis. AnnInternMed. 2015;162(11):750–6.

34.ProudMEandHowardPK.Whydopatientsreturntotheemergency department? AdvEmergNursJ. 2016;38(2):83–8.

35.NunezS,HexdallA,Aguirre-JaimeA.Unscheduledreturnstothe emergencydepartment:anoutcomeofmedicalerrors? QualSafHealth Care. 2006;15(2):102–8.

36.PollackCV,DiercksDB,ThomasSH,etal.Patient-reportedoutcomes fromanational,prospective,observationalstudyofemergency departmentacutepainmanagementwithanintranasalnonsteroidal anti-inﬂammatorydrug,opioids,orboth. AcadEmergMed. 2016;23(3):331–41.

37.MooreRA,DerryS,AldingtonD,etal.Singledoseoralanalgesics foracutepostoperativepaininadults-anoverviewof Cochranereviews. CochraneDatabaseSystRev. 2015;2015(9):CD008659.

ORIGINAL RESEARCH

ImpactofCOVID-19PandemiconEmergencyDepartment

VisitsforOpioidUseDisorderAcrossUniversityof CaliforniaHealthCenters

MatthewHeshmatipour,BS* DingQuanNg,BS*

EmilyYi-WenTruong,BS† JianweiZheng,PhD‡ AlexandreChan,PharmD,MPH* YunWang,PhD‡

SectionEditor: PatrickMeloy,MD

*UniversityofCaliforniaIrvine,SchoolofPharmacyandPharmaceutical Sciences,Irvine,California

† UniversityofCaliforniaIrvine,DonaldBrenSchoolofInformationandComputer Sciences,Irvine,California

‡ ChapmanUniversity,SchoolofPharmacy,Irvine,California

Submissionhistory:SubmittedSeptember15,2023;RevisionreceivedAugust30,2024;AcceptedSeptember6,2024

ElectronicallypublishedOctober2,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18468

Introduction: Coronavirus2019(COVID-19)hashadadevastatingimpactonmentalhealthandaccess toaddictiontreatmentintheUnitedStates,includinginCalifornia,whichresultedinthehighestratesof emergencydepartmentvisits(ED)foropioidpoisoningin2020.AsCaliforniaslowlyreturnstoprepandemicnormalcy,itremainsuncertainwhethertheratesofopioid-relatedeventshavesloweddown overtime.Wehypothesizedthatthenumberofopioid-relatedEDvisitswereexacerbatedaftertheperiod oftheCOVID-19pandemicandcontinueatahighrateinthepresent.

Methods: InthisanalysiswesearchedtheUniversityofCalifornia(UC)HealthDataWarehouse a databaseofelectronichealthrecordsfromsixacademicmedicalcenters foropioidrelatedEDvisits, identifiyingusingthefollowingInternationalClassificationofDiseases,10th Ed,ClinicalModification codes:F11codes,andT40.0*,T40.1*,T40.2*,T40.3*,T40.4*,T40.6*.Opioidoverdose-associated visitswereclassifiedbytypesofopioidsinvolved:heroin(T40.1*);prescriptionopioids(T40.2*or T40.3*);andsyntheticopioids(T40.4*).Weperformedinterruptedtimeanalysistoestimatethe immediate(level)changeandchange-in-timetrend(trendchange),frombefore(January2018–October 2019)andduringthepandemic(April2020–December2022).Monthlyvisitrateswereevaluatedwith negativebinomialregressionadjustedfor first-orderautoregressionandusingall-causeEDcountsas theoffset.Wepresenteffectsizesasrateratios(RR)and95%confidenceintervals(CI),testedat α = .05.

Results: WeobservedadecreaseinoverallEDvisitsfrom28,426to25,121visitsinDecember2019and June2021,respectivelyacrossallsixUCHealthCenters.BeforeCOVID-19,wefoundthatEDvisitrates steadilyincreasedforalloutcomes(P < 0.05)exceptsyntheticopioids.Totalopioid-relatedEDvisitrates increasedby15%(RR1.15,95%CI1.02–1.29, P = 0.20)immediatelyafterMarch2020beforedecreasing by0.5%everymonth,albeitwithoutstatisticalsigniﬁcance(RR.995,95%CI.991–1.00, P = 0.06).Opioidrelatedeventsacrossthesixacademicmedicalcentersincreasefrom232inDecember2019,representing asinglemonth’stotal,andpeakedat315inJune2021.Similartrendswereobservedwithprescriptionopioid overdoses,withastepincreaseof44%(RR1.44,95%CI1.10–1.89, P = .008)beforeplateauingafterMarch 2020(RR1.01,95%CI.998–1.02, P = 0.12).Speciﬁcally,thetotalnumberofprescriptionopioid-relatedED visitsmorethandoubledbetweenDecember2019(22visits)andJune2021(49visits).AfterMarch2020, EDvisitratesforsyntheticopioidoverdoseswereincreasingsteadilyby4%everymonth(RR1.04,95%CI 1.02–1.06, P = .001),unlikewithheroin,whichwasobservedwithan8%monthlyreduction(RR.92,95%CI .90–.93, P < .001).Noimmediateincreaseinvisitrateswasobservedforeitheropioid.

Conclusion: Whileopioid-relatedEDadmissionsamongtheUChealthcentersshowedanoverall decrease,prescriptionandsyntheticopioidoverdosesremainedsigniﬁcantlyhigherthanpre-pandemic trendsasofDecember2022.Amultilevelapproachtoimproveawarenessofnewopioidhealthpolicies couldamelioratethesealarmingrisesinthepost-pandemicera.[WestJEmergMed.2024;25(6)883–889.]

INTRODUCTION

Opioidsplayamajorroleinhealthcareasanimportant prescriptionmedicationforpainrelief.Opioidanalgesicsare abeneficialinterventionwhenproperlyadministered,1 thus creatingchallengeswhenregulatingtheiruseastheyhavethe potentialforlong-termadverseeffects.Althoughearlier phasesoftheopioidcrisiswerecharacterizedbythemisuseof prescriptionopioids(firstwave),recenttrendsrevealthat heroin(secondwave)andillicitsyntheticopioids(thirdwave) havebecomecrucialtocharacterizingtheopioidepidemic.2 Syntheticopioidssuchasfentanylhavebeenshowntobethe mostdevastatingcontributorstothecurrentrisingopioidrelatedcasesduetoitsassociatedpositivesupplyshock, allowingforitspricetoreducesignificantly.3 Predictedtrends overlayingscatterplots.

Beginninginearly2020,theadventofthecoronavirus 2019(COVID-19)pandemicledtoanabrupt,worldwide disruptiontosocietalfunctionsandtypicaldailylifeasstayat-homeorderswereimplementedtocurbthespreadofthe virusandpreservemedicalfacilitiesandequipmentforthe mostsevereinfections.Thesubsequentriseinunemployment ratesandsocialisolationledtoincreasedpsychological distress,4 whichwaspostulatedtohavecausedanationwide increaseinopioidoverdosecasesacrossmultiplestatesin 2020.5–8 Associetyslowlybegantoreturntopre-pandemic normalcyin2021and2022,itremainsuncertainwhetherthe ratesofopioid-relatedeventshaveslowedovertime,given thechallengesofweaningoffchronicopioiduse.Thus,a deeperexplorationintothetrendsrelatedtoopioid-related eventsin2021and2022andareviewofcurrentinterventions andsolutionsisnecessarytoallocateresourcesforenhancing ourmanagementoftheopioidepidemic.

Inthispaperwereportour findingsforemergency department(ED)visitratesassociatedwithopioid-related casesfrom2018–2022acrossthesixUniversityofCalifornia (UC)healthcenters.Wecomparedtheratesfromtheperiod beforethepandemic(January2018–December2019)with thoseduringthepandemic(April2020–December2022). AwashoutperiodbetweenJanuary–March2020was implementedduetowidespreadpublicuncertaintyregarding thenatureofthepandemicduringthattimeframe.These datestakereferencefromtheUS’ s firstconfirmedlaboratory caseonJanuary20,2020,andCalifornia’sstatewidestay-athomeorderonMarch19,2020.9 Wehypothesizedthat opioid-relatedEDvisitratescontinuedtoworsenafter

Whatdowealreadyknowaboutthisissue?

Theopioidusedisorder(OUD)epidemicand COVID-19pandemicaretwopublichealth crisesthatsigni ﬁ cantlyincreasedEDvisitsin theUS.

Whatwastheresearchquestion?

HowdidtheCOVID-19pandemicaffect opioid-relatedEDvisitsinCalifornia?

Whatwasthemajor ﬁndingofthestudy?

Opioidprescriptionsandheroin-relatedED visitsincreasedafterthepandemicacross UChospitals.

Howdoesthisimprovepopulationhealth?

Our ﬁ ndingsshowtrendsofopioidusein Californiaandidentifykeycommunity elementsthatcouldcontributetoOUD interventionsthroughfutureresearch.

2021andremainedhigherthanpre-pandemicratesas ofDecember2022.

MATERIALSANDMETHODS

DataSource

ForthisanalysisweusedtheUCHealthDataWarehouse (UCHDW),adatabaseofelectronichealthrecordsfromthe sixUCacademichealthcenters:UCDavis;UCIrvine;UC LosAngeles;UCRiverside;UCSanDiego;andUCSan Francisco.TheUCHDWcontainsclinicaldataof8.7million patientsseenataUCfacility,totalingapproximately378 millionencountersincludingofﬁcevisits,inpatient admissions,andEDvisits.10 Alldataisorganizedbasedon theObservationalMedicalOutcomesPartnership(OMOP) commondatamodel(CDM),anopencommunitydata standardforstandardizingthestructureofreal-worldclinical dataacrossinstitutionsdespitedifferencesintheunderlying clinicaldatainfrastructure.11 AdoptingtheOMOPCDM facilitatesefﬁcient,multicenteredstudiesandgenerationof

PopulationHealthResearchCapsule

reproducibleevidence.Institutionalreviewboard(IRB) reviewwasnotrequiredforthisdataanalysisaswedeidentiﬁedallaccesseddataelementspriortoreceipt.12 All dataquerieswerecompletedonFebruary8,2023.

EmergencyDepartmentVisits

FollowingthestructureoftheOMOPCDM,weidentiﬁed EDvisitsfromthe “visitoccurrence” domainusingvisit conceptidentiﬁcations(IDs)of “9203 – EmergencyRoom Visit” or “262 – EmergencyRoomandInpatientVisit.”13

Opioid-relatedEvents

TheEDvisitsassociatedwithopioid-relatedevents, labelledas “allopioid-relatedvisits,” wereidentifiedifthey hadatleastoneofthefollowingInternationalClassification ofDiseases,10th ed,ClinicalModification:F11codes,and T40.0*,T40.1*,T40.2*,T40.3*,T40.4*,T40.6*.Opioid overdose-associatedvisitswerethenfurtherclassifiedby typesofopioidsinvolved:heroin(T40.1*);prescription opioids(T40.2*orT40.3*);andsyntheticopioidsotherthan methadone(T40.4*).14

OutcomesMeasures

Wesummarizedandanalyzedtheoutcomesasmonthly EDvisitratesofopioid-relateddisordersandoverdoses(per 100,000all-causeEDvisits).Thiswasjustiﬁedbythelower all-causeEDvisitratesacrosstheUSduringtheCOVID-19 pandemic15 andwasimplementedinpreviouslypublished studies.8,16 WealsoplottedmonthlyEDvisitcountsacross theUChealthcentersfrom2018–2022toverifythistrend. ActualEDvisitcountsforopioid-relatedvisitswerealso maskedasperinstitutionalpolicy.

StatisticalAnalysis

Weperformedinterruptedtimeanalysistoestimatethe immediate(level)changeandchange-in-timetrend(trend change)withnegativebinomialregressionwithrobust standarderrors(Huber-Whitesandwichestimator)foroverdispersedoutcomes.Themodelwasfurtheradjustedfor ﬁrstorderautoregressionusingalag-1variable(outcomevalue fromthepreviousmonth),withall-causeEDvisitcountsas theoffsetvariable.Themodelspeciﬁcationisasfollows: log EytðÞ ðÞ = β 0 + β 1

• yt wasthecountoutcomeofinterestinmonth t

• covid t wasabinarydummyvariableindicatingwhether month t wasbefore(January2018–December2019, assignedwithavalue “0”),orduring(April 2020–December2022,assignedwithavalue “1”)the COVID-19pandemic.Awashoutperiodbetween January–March2020wasimplementeddueto

widespreaduncertaintyregardingthenatureofthe pandemicduringthattimeframe.

• timet wasadiscretetimevariablecountingmonths startingfromJanuary2018,takingthevaluesof 1,2,3,etc.

• posttimet wascoded “0” fortimepointsbeforethe COVID-19pandemic(January2018–December2019) andthenbecomeadiscretetimevariablecounting monthsduringtheCOVID-19pandemic(April 2020–December2022),takingthevaluesof1,2,3,etc.

• lag 1t wasthecountoutcomeofinterestinthemonth prior(ie, yt 1 ).Thisallowscontrolofautocorrelationin atime-seriesdataset.

• offsett takesthenumberofall-causeEDvisitsinmonth t.Thisvariablewasexcludedforthetrendanalysisof all-causeEDvisitcounts.

• Thekeycoefﬁcientsofinterestincludethefollowing:

◦ β 1 ,whichrepresentstheimmediate(level)changein log EytðÞ ðÞ duetotheCOVID-19pandemic.

◦ β 2 ,whichrepresentsthemonthlychangein log EytðÞ ðÞ beforetheCOVID-19pandemic(January 2018–December2019).

◦ β 2 + β 3 ,whichrepresentsthemonthlychangein log EytðÞ ðÞ duringtheCOVID-19pandemic(April 2020–December2022).

Effectsizesforcountandrateoutcomesarepresentedas ratiosand95%conﬁdenceintervals(CI)viaexponentiatingthe coefﬁcients.Predictedtrendsoverlayingscatterplotswere plottedtoillustratethetrendsinthecountandrateoutcomes permonth.Allanalysesweretestedwith α = .05andcompleted withStatav16.1(StataCorpLLC,CollegeStation,TX).

RESULTS

All-CauseEDVisits

AtthebaselineofDecember2019thetotalnumberofED visitswas28,426acrossthesixhealthcenters,whereasitfell slightlyto25,121visitsinJune2021.Monthlyall-causeED visitcountsremainedrelativelyconstantbeforetheCOVID19pandemic(P = 0.30).Duringthepandemic,therewasa suddendeclineby34%comparedtopre-pandemictrends (95%CI15–49%, P = 0.002)beforeincreasingby1.5%per month(95%CI0.5–2.5%, P = 0.002)(Figure1).

AllOpioid-relatedEvents

AtthebaselineofDecember2019thenumberoftotal opioid-relatedEDvisitswas232acrossthesixhealthcenters, whereasitjumpedto315visitsinJune2021.TheEDvisitrates ofallopioid-relatedvisitsincreasedby1%permonthduring thepre-COVID-19period(95%CI0.6–2.0%, P < 0.001).This wasfollowedbya15%immediateincreaseduringthe pandemic,comparedtopre-pandemiclevels(95%CI2–29%, P = 0.020),andastatisticallynon-signiﬁcantmonthlydecline by0.5%(95%CI 0.9to0.0%, P = 0.06)(Figure2A).

Figure1. Monthlycountsofall-causeemergency(ED)visitsfrom 2018–2022acrosssixUniversityofCaliforniahealthcenters.The periodfromJanuary2018–December2019wasdesignatedas “pre-COVID,” whiletheperiodfromApril2020–December2022 wasdesignatedas “COVID.” Awashoutperiodfrom January–March2020isrepresentedbytheredlinesinthegraph.

PrescriptionOpioidOverdoses

AtthebaselineofDecember2019thenumberof prescriptionopioid-relatedEDvisitswas22acrossthesix healthcenters,whereasitjumpedto49visitsinJune2021. TheEDvisitratesofprescriptionopioidoverdosesincreased by1%everymonthbeforetheCOVID-19pandemic(95%CI

0.2–2.0%, P = 0.02).DuringCOVID-19,therewasan immediateincreaseinvisitratesof44%(95%CI10–89%, P = 0.008)andastatisticallynon-signiﬁcantincreaseof1% everymonth(95%CI 0.2to2.0%, P = 0.12)(Figure2B).

SyntheticOpioidOverdoses

AtthebaselineofDecember2019thenumberofsynthetic opioid-relatedEDvisitswaslessthan11acrossthesixhealth centers,whereasitjumpedslightlyto14visitsinJune2021. Countslessthan11aremaskedfollowingtheUCCenterfor Data-drivenInsightsandInnovation(CDI2)policyon limitedresearchusesusingtheUCHDW.TheEDvisitrates ofsyntheticopioidoverdosesincreased5%everymonth, albeitwithoutstatisticalsigniﬁcance(95%CI0.2–11.0%, P = 0.06),beforethepandemic.DuringCOVID-19,there wasastatisticallynon-signiﬁcantincreaseof69%(95%CI 32to316%, P = 0.26)comparedtopre-pandemictrends, followedbyamonthlyincreaseof4%(95%CI2–6%, P = 0.001)(Figure2C).

HeroinOverdoses

AtthebaselineofDecember2019thenumberofheroinrelatedEDvisitswas18acrossthesixhealthcenters,whereas itfellslightlytolessthan11visitsinJune2021.Aswiththe countsofsyntheticopioid-relatedEDvisits,countslessthan 11aremaskedperCDI2policy.MonthlyEDvisitratesfor

Figure2. Monthlyemergencydepartment(ED)visitratesofopioid-relateddisordersandoverdoses(per100,000all-causeEDvisits)from 2018–2022acrosssixUniversityofCaliforniaHealthcenters.TheperiodfromJanuary2018– December2019wasdesignatedas “preCOVID,” whiletheperiodfromApril2020–December2022wasdesignatedas “COVID.” AwashoutperiodfromJanuary–March2020was implementedandisrepresentedbytheredlinesinthegraph.

heroinoverdosesincreasedby2%monthlyduringthepreCOVID-19period(95%CI0.6–3.0%, P = 0.006). Subsequently,thepandemicledtoanimmediatebut statisticallynon-signiﬁcantincreaseby20%(95%CI 14to 67%, P = 0.28)comparedtothepre-COVID-19period.This wasfollowedbyamonthlydeclineof8%(95%CI 10to 7%, P < 0.001)(Figure2D).

DISCUSSION

TheCOVID-19pandemicbroughtlighttotheurgentneed formultilevelinnovativeapproachestoaidagainsttheopioid epidemicinCalifornia.Inparticular,thepandemichas facilitatedtheworseningoftrendsrelatedtoprescriptionand syntheticopioidoverdosesacrosstheUChealthcenters,with littlesignsofimprovementasrecentasDecember2022. Consideringthatpastresearchonthesamesubjectwasonly updatedasof2020andthatour ﬁndingsweresimilartothose papers, 5–8 ourstudymayprovidethemostrecentoutlookof thecurrentopioidcrisisacrosstheUSandreﬂectthatmore couldbedonetoarresttheproblem.

Interestingly,wesawadecreasingtrend,albeitnonstatisticallysigniﬁcant,inEDvisitratesrelatedtoalltypesof opioid-relatedeventsafteraninitialspikeinApril2020, whichcontrastedwiththetrendswesawwithopioid overdoses.Whilethekeyreasonscouldnotbedetermined basedontheavailabledata,wehypothesizethatthis decreasingtrendcouldbeanartifactofarecoveryinnumbers ofall-causeEDvisits.Additionally,thistrendpossibly reﬂectsaparadigmshiftintheseekingoftreatmentforopioid usedisorderfromEDsduringthepandemic.Akey considerationisthatopioid-relatedeventscomprisenotonly opioidoverdosecomplicationsbutalsoopioidwithdrawal episodes,andbothcanbetreatedattheEDwithlifesaving naloxoneandbuprenorphine,respectively.17

TheEDalsoprovidesthesemedicationsforopioiduse disorder(MOUD)ondischargeasstand-bymedicationsfor useasrequiredandconnectspatientswithfollow-upOUD managementinthecommunity.18,19 InCalifornia,theCA Bridgeprogramhasspearheadedthecoordinated implementationoflow-thresholdbuprenorphinetreatment foropioidwithdrawals(anapproachthatreducesasmany barriersaspossibleregardingaccesstobuprenorphine20)at EDsof52hospitals,includingfourUChealthcenters,since 2018.21 Unfortunately,theprogramreportedasteepdecline inED-initiatedbuprenorphineduringthe ﬁrstmonthsofthe pandemic.22 Ourstudymayprovidepreliminaryevidence thatthelowerratesofOUDtreatmentatEDs,especiallyfor withdrawalepisodes,havecontinuedinto2021and2022. Furtherresearch,suchasexaminingratesofadministration andprescriptionofdifferentMOUDattheED,willbe requiredtoconﬁrmthishypothesis.

Attheinitialstagesofthepandemic,Currieetalreported thatopioidprescriptionsacrosstheUSweregenerally maintainedamongexistingopioidusersbutdecreasedbrieﬂy

fornewusers.23 AnotherstudybyZhengetalreporteda declineinopioidprescriptionsandprescribersbeforeand duringtheCOVID-19pandemicperiodinCalifornia.24 Therewerelittletonosignsofasuddenincreaseinopioid prescriptionsthatcouldexplainthehigherratesof prescriptionopioidoverdoseratesseeninourstudy.A multisitereportstudyingdrugusebehaviorsduringthe pandemicfoundthatCOVID-19-relatedstressors,suchas jobloss,increasedhousinginsecurity,andloneliness,were commonlycitedreasonsforincreasingdrugusebehaviors.25 Thesamestudyfurtherreportedthatsyntheticopioids, particularlyfentanyl,hadsaturatedtheillicitdrugmarket duringthistimeduetoitsinexpensivenatureandhigh accessibilitycomparedtomethamphetamineandheroin.25 Thesetrendswerefurtherveriﬁedinthe CaliforniaHealth CareAlmanac featureonCaliforniasubstanceuse.26 Other thanstrategizingtoreducethesupplyofillicitfentanyland fentanyl-lacedpills,improvingthementalhealthofopioid users,andtherebyreducingdemandforbothprescription andsyntheticopioids,willalsohelptomitigatetheopioid epidemicmoreeffectively.

Recentpolicychangeshavebeenmadetopurposefully bringMOUDclosertopatientsandpreventdeathsfrom fatalopioidwithdrawalsandoverdoses:1)theremovalofthe X-waiver(issuedbytheSubstanceAbuseandMentalHealth ServicesAdministrationandtheDrugEnforcement Administration)enabledphysicianstoprescribe buprenorphineinclinicswithoutadministrativebarriersand extensivecertiﬁcation27,28;and2)theapprovalbytheFood andDrugAdministrationofNarcan,the ﬁrstover-thecounternaloxonenasalspray.29 However,withthe emergenceofthefourthwaveoftheopioidcrisis characterizedbyopioidandstimulantco-use,30 thewar againsttheopioidepidemicisnotyetwon,andeffective policychangeswillnecessitatekeychangesincurrent practicesandoperations.

Variscoetalidentiﬁedlegacybarriersinstate-level regulationsandwholesalerpolicythatcouldlimit buprenorphinesuppliesanddispensingatpharmacies.31 Educationandtheintentionaldistributiontothecommunity willberequiredtoensurethatNarcanisalwaysavailableto respondtoanopioidoverdoseevent.Finally,thereremains evidenceofhigh-riskopioidprescribinganddispensing behaviorsassociatedwithayoungerandlesseducated demographic,whichmayprecipitategreaterratesofopioidrelatedcases.32 Thereisstillmuchtodoasclinicians, researchers,andcommunityadvocatestoensurethe effectivemanagementoftheopioidepidemic,givenour availableresources.

STRENGTHSANDLIMITATIONS

Onelimitationofthestudyistheuseofdeidentiﬁeddata, whichpreventedusfromevaluatingindividual-level characteristicsthatarepredictiveofhigheropioid-related

admissionsorreadmissions.Suchdatawillneedtobe navigatedaroundexistingandnewregulatoryframeworks frommultisiteIRBsacrossmultipleUChealthcenters. Toourknowledge,thisisahurdlethathasyettobe tackledattheUC-widelevel,andwesuspectthatmore timewillbeneededtoestablishtheprocess.Onceestablished, wewillstrivetodeterminesociodemographicand geolocationfeaturesthatcouldhelpwithadvisingthe optimalallocationofresourcesinthiswaragainstthe opioidepidemic.

Regardless,thestrengthofthiscurrentstudyliesintheuse ofreal-worldEDdatatorapidlyassessthechangesinopioidrelatedEDtrendsduringtheCOVID-19crisiswhenresearch datacollectionhadbeenhaltedduetolockdownmeasures. Ourstudyhasalsodemonstratedthefeasibilityofusingthis methodologytocreateareal-timeOUDdashboard33 to facilitatethetimelydisseminationoftrenddatainopioidrelatedEDvisitsandadmissionsintheeverchanging,opioid uselandscapeinCalifornia.Additionally,wesoughtto includecommunity-basedprograms,suchasOUD-based harmreductionorganizations,asabeneﬁcialinterventionto reducethenumberofopioid-relatedEDvisits.However,the UCHDWdidnotincludedatathatwasrelevanttothese programs,andthereisyettobearobuststudydisplaying theirsigniﬁcance.

CONCLUSION

Ourdataprovidessupportthatopioid-relatedEDvisit ratesinCaliforniahaveexceededpre-pandemicratesand havecontinuedtoworsenafter2021.AlthoughtheCOVID19pandemicsawanoveralldecreasingtrendinthenumber ofopioid-relatedEDvisitsinalltheUCHealthcenters,the numberofvisitsduetoprescriptionandsyntheticopioidsrelatedoverdoseremainshigh.Moresigniﬁcantly,this increasingtrendprovidesagreatpublichealthconcern, especiallyastheUSentersthefourthwaveoftheopioid epidemic,characterizedbypolysubstanceuse.Notonlymust wereducethesupplyofillicitopioids,weshouldalsoaimto reducethedemandforprescriptionandsyntheticopioids, whichislikelythefunctionofworseningmentalhealth duringtheCOVID-19pandemic.WhiletheCOVID-19 pandemiccontinuestobecomelessnotabletothepublic, theconsequentialchanginglandscapeoftheopioid epidemicremainsanuphillbattlethatcallsformultilevel, proactive,innovative,andcollaborativeapproachesacross theUS.

ACKNOWLEDGMENTS

TheauthorsthanktheCenterforData-drivenInsights andInnovationatUCHealth(CDI2; https://www.ucop.edu/ uc-health/functions/center-for-data-driven-insights-andinnovations-cdi2.html),foritsanalyticalandtechnical supportrelatedtouseoftheUCHealthDataWarehouse.

AddressesforCorrespondence:AlexandreChan,PharmD,MPH, UniversityofCaliforniaIrvine,SchoolofPharmacyand PharmaceuticalSciences,802WPeltasonDr.,Irvine,CA92697. Email: a.chan@uci.edu.YunWang,PhD,ChapmanUniversity, SchoolofPharmacy,9401JeronimoRd.,Irvine,CA92618.Email: yunwang@chapman.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.VolkowNDandBlancoC.Thechangingopioidcrisis: development,challengesandopportunities. MolPsychiatry. 2021;26(1):218–33.

2.CiccaroneD.Thetriplewaveepidemic:supplyanddemanddriversof theUSopioidoverdosecrisis. IntJDrugPolicy. 2019;71:183–8.

3.CiccaroneD.FentanylintheUSheroinsupply:arapidlychangingrisk environment. IntJDrugPolicy. 2017;46:107–11.

4.RajA,JohnsNE,BarkerKM,etal.TimefromCOVID-19shutdown, gender-basedviolenceexposure,andmentalhealthoutcomesamonga staterepresentativesampleofCaliforniaresidents. EClinicalMedicine. 2020;26:100520.

5.RoddaLN,WestKL,LeSaintKT.Opioidoverdose-relatedemergency departmentvisitsandaccidentaldeathsduringtheCOVID-19 pandemic. JUrbanHealth. 2020;97(6):808–13.

6.HallGT,CruzDS,LankPM,etal.Opioid-relatedemergencydepartment visitsduringCOVID-19inalargehealthsystem. JAddictMed. 2021;15(4):345–8.

7.SlavovaS,RockP,BushHM,etal.Signalofincreasedopioidoverdose duringCOVID-19fromemergencymedicalservicesdata. DrugAlcohol Depend. 2020;214:108176.

8.SoaresWE,MelnickER,NathB,etal.Emergencydepartmentvisitsfor nonfatalopioidoverdoseduringtheCOVID-19pandemicacrosssixUS healthcaresystems. AnnEmergMed. 2022;79(2):158–67.

9.CentersforDiseaseControlandPrevention.CDCmuseumCOVID-19 timeline.Availableat: https://www.cdc.gov/museum/timeline/covid19. html.AccessedAugust28,2023.

10.UniversityofCaliforniaOfﬁceofthePresident.Centerfordata-driven insightsandinnovation(CDI2).Availableat: https://www.ucop.edu/ uc-health/departments/center-for-data-driven-insights-andinnovations-cdi2.html.AccessedAugust28,2023.

11.ObservationalHealthDataSciencesandInformatics(OHDSI). Standardizeddata:theOMOPcommondatamodel.Availableat: https:// www.ohdsi.org/data-standardization/.AccessedAugust28,2023.

12.PetersonTA,FontilV,KoliwadSK,etal.Quantifyingvariationin treatmentutilizationfortype2diabetesacross ﬁvemajoruniversityof Californiahealthsystems. DiabetesCare. 2021;44(4):908–14.

13.ObservationalHealthDataSciencesandInformatics(OHDSI).OMOP CDMv5.3.Availableat: https://ohdsi.github.io/CommonDataModel/ cdm53.html#VISIT_OCCURRENCE.AccessedAugust28,2023.

14.DiRicoR,NambiarD,StoovéM,etal.DrugoverdoseintheED:arecord linkagestudyexaminingemergencydepartmentICD-10coding practicesinacohortofpeoplewhoinjectdrugs. BMCHealthServRes. 2018;18(1):945.

15.HartnettKP,Kite-PowellA,DeViesJ,etal.ImpactoftheCOVID-19 pandemiconemergencydepartmentvisits-UnitedStates,January1, 2019-May30,2020. MorbMortalWklyRep. 2020;69(23):699–704.

16.JohnsonE,AxeenS,VosooghiA,etal.Interruptedtimeseriesanalysis: patientcharacteristicsandratesofopioid-use-disorder-related emergencydepartmentvisitsintheLosAngelesCountypublichospital systemduringCOVID-19. Healthcare(Basel). 2023;11(7):979.

17.HuhnASandDunnKE.Whyaren’tphysiciansprescribingmore buprenorphine? JSubstAbuseTreat. 2017;78:1–7.

18.HuntleyK,EinsteinE,PostmaT,etal.Advancingemergency department-initiatedbuprenorphine. JAmCollEmergPhysiciansOpen. 2021;2(3):e12451.

19.D’OnofrioG,McCormackRP,HawkK.Emergencydepartmentsa24/7/365optionforcombatingtheopioidcrisis. NEnglJMed. 2018;379(26):2487–90.

20.JakubowskiAandFoxA.Deﬁninglow-thresholdbuprenorphine treatment. JAddictMed. 2020;14(2):95–8.

21.SnyderH,KalminMM,MoulinA,etal.Rapidadoptionoflow-threshold buprenorphinetreatmentatCaliforniaemergencydepartments participatingintheCAbridgeprogram. AnnEmergMed. 2021;78(6):759–72.

22.HerringAA,KalminM,SpeenerM,etal.Sharpdeclineinhospitaland emergencydepartmentinitiatedbuprenorphineforopioidusedisorder duringCOVID-19stateofemergencyinCalifornia. JSubstAbuseTreat. 2021;123:108260.

23.CurrieJM,SchnellMK,SchwandtH,etal.Prescribingofopioid analgesicsandbuprenorphineforopioidusedisorderduringthe COVID-19pandemic. JAMANetwOpen. 2021;4(4):e216147.

24.ZhengJ,RakovskiC,VariscoT,etal.EPH198thesustainedimpactof COVID-19onopioiddispensinginCalifornia. ValueinHealth. 2023;26(6S):S200.

25.BolinskiRS,WaltersS,Salisbury-AfsharE,etal.Theimpactof theCOVID-19pandemicondrugusebehaviors,fentanylexposure, andharmreductionservicesupportamongpeoplewhouse drugsinruralsettings. IntJEnvironResPublicHealth. 2022;19(4):2230.

26.CaliforniaHealthCareFoundation.SubstanceuseinCalifornia: prevalenceandtreatment.2022.Availableat: https://www.chcf.org/ publication/2022-edition-substance-use-california/ AccessedAugust29,2023.

27.UnitedStatesDrugEnforcementAdministration.DEAapplaudsthe repealoftheX-Waiver.2023.Availableat: https://www.dea.gov/ documents/2023/2023-01/2023-01-17/dea-applauds-repeal-x-waiver AccessedAugust28,2023.

28.HouseTW.WhiteHousehostseventtomarkremovalofbarriersto addictiontreatment.2023.Availableat: https://www.whitehouse. gov/ondcp/brieﬁng-room/2023/01/25/white-house-hosts-event-tomark-removal-of-barriers-to-addiction-treatment/ Accessed August28,2023.

29.UnitedStatesFederalDrugAdministration.FDAapproves ﬁrstover-thecounternaloxonenasalspray.2023.Availableat: https://www.fda.gov/ news-events/press-announcements/fda-approves-ﬁrst-overcounter-naloxone-nasal-spray.AccessedAugust28,2023.

30.JenkinsRA.ThefourthwaveoftheUSopioidepidemicandits implicationsfortheruralUS:afederalperspective. PrevMed. 2021;152(Pt2):106541.

31.VariscoTJ,WanatM,HillLG,etal.TheimpactoftheMainstreaming AddictionTreatmentActandassociatedlegislativeactiononpharmacy practice. JAmPharmAssoc(2003). 2023;63(4):1039–43.

32.LuH,ZhengJ,WangY.Geospatialanalysisofopioiddispensing patternsinCalifornia:a2021real-worldstudy. Healthcare(Basel). 2023;11(12):1732.

33.WuE,VillaniJ,DavisA,etal.Communitydashboardstosupport data-informeddecision-makingintheHEALingcommunitiesstudy. DrugAlcoholDepend. 2020;217:108331.

ORIGINAL RESEARCH

AnAssessmentofthePresenceof Clostridiumtetani intheSoilandonOtherSurfaces

MichaelShalaby,MD*†

AlessandroCatenazzi,PhD‡

MelissaF.Smith,MD†

RobertA.FarrowII,DO*†

DavidFarcy,MD*†

OrenMechanic,MD*†

TonyZitek,MD*†

SectionEditor:MarkI.Langdorf,MD,MHPE

*HerbertWertheimCollegeofMedicineatFloridaInternationalUniversity, DepartmentofEmergencyMedicineandCriticalCare,Miami,Florida † MountSinaiMedicalCenterMiamiBeach,DepartmentofEmergencyMedicine, MiamiBeach,Florida

‡ CatenazziLabatFloridaInternationalUniversity,DepartmentofBiologicalSciences, Miami,Florida

Submissionhistory:SubmittedJanuary16,2024;RevisionreceivedSeptember3,2024;AcceptedSeptember6,2024

ElectronicallypublishedOctober22,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18702

Introduction: Standardemergencymedicinepracticeincludestetanusvaccineadministrationaspartof woundcaremanagementforpatientswhoarenotfullyimmunized.Sincetherehavebeennoavailable studiesintheUnitedStatesreafﬁrmingtheprevalenceof Clostridiumtetani(Ctetani) since1926,we soughttoidentifyitsprevalenceinamajorurbancountyintheUS.

Methods: Wesampledsoil,rustedmetal,concrete,anddogfecestodeterminetheprevalenceof Ctetani inasinglemetropolitancountyintheUnitedStates.Soilsamplesandswabswerecollectedfrom fourlocations:thesoilofapublicparkandanelementaryschool;dogfecesfromasinglepublicdogpark; andrustedsurfaces(metalandconcrete)incommonstudentareasofauniversitycampus.The presenceof Ctetani ineachsamplewasdeterminedusingaquantitativepolymerasechainreaction.

Results: Intotal,200sampleswerecollected,ofwhich37(18.5%)testedpositivefor Ctetani DNA. Amongthe140samplestakenfromthesoil,justone(0.7%)testedpositivefor Ctetani DNA.Ofthe40 samplesofrustedmetalandconcretesurfaces,30(75%)testedpositivefor Ctetani,andsix(30%)ofthe 20samplesfromdogfecestestedpositivefor Ctetani.

Conclusion: Wefoundthat Ctetani isfrequentlypresentonrustedmetalandconcretesurfacesbutrarely insoilsamples.Minorwoundscontaminatedwithsoilmaybeconsideredlowriskfortetanus.However, futurestudiesshouldassesstheburdenof Ctetani inothersimilarurban,suburban,andruralenvironments tohelpdeterminethethreatof Ctetani moreexactly.[WestJEmergMed.2024;25(6)890–893.]

INTRODUCTION

Background

Clostridiumtetani (Ctetani ),aGram-positiveobligate anaerobe,isthecausativeagentoftetanus,adisorderthat inducesuncontrollablemusclespasms(knownastetany)and carrieshighmortality.1 Itispreventedbyacommonly administeredtoxoidvaccine.1 Ctetani isthoughttoinhabit soil,mostofteninthesporeform,throughwhichitcan withstandextremetemperaturesandvolatileenvironments.2 Afterinoculationofcontaminatedwounds,thespores

proliferateandspawnvegetativebacteria,whichrelease toxinsthatprecipitatethedisease’scharacteristicsymptoms oftetany.1

Tetanusposesaconsiderableriskindevelopingcountries withlittleaccesstovaccination.In2015therewerenearly 57,000casesoftetanusreportedworldwide,with79% originatinginSouthAsiaandsub-SaharanAfrica.3 The annualincidenceoftetanusintheUnitedStates,incontrast, isverylow.Sincetheintroductionofthetetanustoxoid vaccineinthe1930s,therateofinfectionhassteadily

declinedfromapeakof500in1950tonomorethan30cases yearly.4 Documentedcasesaretypicallyininjectiondrug usersandtheelderly,whohaveahigherriskofinsufﬁcient antibodytitersdespiteupdatedvaccinationstatus.1,4

TheUSCentersforDiseaseControlandPrevention AdvisoryCommitteeonImmunizationPractices(ACIP) recommendstetanusvaccinationaspartof “routinewound caremanagement” forpatientswhoarenotuptodatewith theirvaccinationaftersustainingawound.5 Protocolsfor woundcharacteristics(ie,abrasionvslaceration)arenot specifiedbytheACIP.6 However,theAmericanCollegeof EmergencyPhysiciansguidelinesdifferentiatebetween “minorwoundsandsuperficialburns” and “otherwounds” : minorwoundsrequireaboosterwithin10years,while “other wounds” requireaboosterwithin fiveyears. 7 Thus,tetanus toxoidisadministeredliberallyinUSemergency departments(ED)aspartofroutinewoundcare,whetherfor simpleabrasionsormorecomplexwounds.

Theprevalenceof Ctetani inthesoilhasnotbeen measuredintheUSsince1926,whenDamonetalfed culturedsoilspecimensobtainedmostlyfromfarmlandto pigsandsubsequentlymonitoredthemforsignsofdisease.8 Morerecentlyin2008,BukaretalsampledsoilinNigeria, andviaincubationofspecimenstheydemonstrateda60% incidenceof Ctetani 9 However,thesestudiesmaynotbe generalizabletomodernUSpopulations.Newer,more robustmethodsfordeterminingthepresenceof Ctetani exist today;furthermore,83%ofpeopleintheUStodayresidein urbanenvironments10 wheretheburdenof Ctetani inthe environmentmaydiffer.Thus,thetrueprevalenceof Ctetani inthemodern,urbanUSenvironmenthasyetto beelucidated.

Wesoughttodeterminethefrequencywithwhich Ctetani ispresentinthesoilascomparedtoconcrete,metal,anddog fecesinasingle,majorurbancountyintheUS.

MATERIALSANDMETHODS

Weassessedenvironmentalsamplesforthepresenceof Ctetani DNAinMiami-DadeCounty,Florida,whichhasa populationofapproximately2.7millionpeople.11 Thisstudy didnotincludehumansubjectsandthuswasexemptfrom reviewbytheinstitutionalreviewboard.Thisresearch receivednooutsidefunding.

EightysoilsampleswerecollectedinsterilizedWhirl-Pak bags(FiltrationGroupCorp,Madison,WI)fromanurban publicparkandanelementaryschool.Thesesiteswerechosen basedontheirdistancefromeachotherandtheirlikelihoodto representisolatedsoil ﬂorawithinthesamecounty,butnot withincloseenoughproximitytobeexpectedtosharesimilar ﬂora.Eachsoil-samplebagcontainedthreeseparatesamples fromwithinafewinchesofsoilusingthreeseparateplastic spoonsthatweredisposedofaftereachuse.Wecollected samplesthiswaysuchthateachbagwaslargeenoughandthat oneindividualspoonmightnot “contaminate” theothertwo

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Clostridiumtetaniisassumedtobewidely presentwithinthesoil.Thelastsoilsample studyperformedintheUnitedStateswasin 1926,whichonlyshowedonepositivesample of62collected.

Whatwastheresearchquestion?

WhatistheprevalenceofC.tetaniinthesoil, indogfeces,andonconcreteandmetal surfaceswithinasingleurbancounty?

Whatwasthemajor ﬁndingofthestudy?

Amongthe140samplestakenfromthesoil, just1(0.7%)testedpositiveforC.tetani DNA.Ofthe40samplesofrustedmetal andconcretesurfaces,30(75%)tested positive,asdid6(30%)of20samples fromdogfeces.

Howdoesthisimprovepopulationhealth?

Certainwoundtypes(ie,soilcontamination) maycarryalowerriskforCtetani,and theelevatedcostoftetanustoxoid administrationintheemergencydepartment maybeforgoneforoutpatientvaccination whichismuchcheaper.

sampleswithinthesamebag.Theothersamples,collectedby DRYSWABbrush(MedicalWire&EquipmentLtd, Corsham,UK),included20samplesofdogfecesfromonedog park,and60combinedsamplesofconcreteandrustedmetal surfaces(suchasmetalsignsandrailings,andconcrete walkways)atasinglepublicuniversity.Asubsequentsetof60 soilsamples,alsocollectedbydryswab,wereagaintakenfrom thesamelocationsastheoriginalsoilsamplesinthesterilized bags.Sampleswereimmediatelytakentotheprocessing laboratoryaftercollection.

Sampleswereanalyzedinauniversitymicrobiologylab usingquantitativepolymerasechainreaction(qPCR) followingastandardizedpreviouslydescribedmethod (Akbulutetal).12 Thisassayampliﬁesa160-basepair fragmentoftheteNTgene(tetanustoxin)of Ctetani (GenBankAccessionNumberX06214,X04436).Thetipof eachswabwasremovedandplacedintoa1.5milliliter Eppendorftubecontaining50microliters(μl)PrepMan UltraSamplePreparationReagent(ThermoFisher Scientiﬁc,Waltham,MA).Thetubewasthenincubatedfor

10minutesat100°Cinadrybath,afterwhichitwas centrifugedat13,000revolutionsperminuteforthree minutes.Theinitial80soilsampleswereanalyzedwith 500milligramsofmixedsoil,whilethedryswabsweremixed withareagentintheabsenceofsigniﬁcantamountsofsoilin thesample.

Five μlofDNAwasthenextractedfromthetubeand transferredtoa0.6-milliliter(mL)qPCRtubecontaining 45 μlof1/10TEbuffer.Aftermixing,5 μlofthisdiluted DNAproductwasaddedtoanewqPCRtubealongwith 20 μlofa62.5:35.5:1:1mixtureofSYBRGreenmastermix (ThermoFisherScientific,Waltham,MA):purifiedwater: forwardprimerTeNT-F(CCTAGTTTCAAAACTTAT TGGCTTATGTAA):reverseprimerTeNT-R (CATAATTCTCCTCCTAAATCTGTTAATGAT).The qPCRwasperformedonaQuantStudioTM3Real-Time PCRInstrument(96-well0.1mLBlock)(Applied BiosystemsInc,FosterCity,CA)asfollows:twominutesat 50°C,followedbytwominutesat90°C,followedby51cycles of15secondsat95°C/1minuteat56°C,followedbya final15 secondsat95°C.OurqPCRwasspecifiedforthe first160 basepairfragmentsoftheteNTgeneof Ctetani . Theplateincludedthreedistilledwaternegativecontrols, onePrepMannegativecontrol,andthreeserialdilutionsof double-stranded,syntheticDNA(gBlock,IntegratedDNA TechnologiesInc,Coralville,IA)oftheteNTgeneof Ctetani (GenBankAccessionNumberX06214,X04436). ResultswereanalyzedinQuantStudioDesignandAnalysis Softwarev1.5.1(AppliedBiosystemsInc,FosterCity,CA). Nopowercalculationwasperformed.Withnoexternal funding,theinvestigatorsdeterminedwehadfundsforthree plates.Themaximumnumberofsamplesperplatewas82. Weanalyzed200totalsamplesfromtheenvironmentto assessforthepresenceof Ctetani DNA.Inthelaboratory setting,theassayweusedhasnearly100%sensitivityand specificityfor Ctetani DNA,butitispossiblethatin environmentalsamplesthepresenceofadditionalsubstances mayinterferewithit.Therefore,wetagged42soilsamples with Ctetani toassesstheaccuracyofourassay(thecontrol group).Wecalculatedthenumber(%)ofsamplesthatwere positivefor Ctetani overall,foreachtypeofenvironmental sample,andforthecontrolgroup.Wecomparedthethree groups(soil,concreteandmetal,anddogfeces)inapairwise fashionwithregardtothepercentageofsampleswith Ctetani DNAusingFisherexacttests.

RESULTS

Overall,ofthe200samplescollectedforanalysis,37 (18.5%)testedpositivefor Ctetani DNA(Table).The ﬁrst80 samplesconsistedofsoilcollectioninplasticbagsfrom publicparks.These80sampleswereallnegativefor Ctetani DNA byouranalysis.However,giventhepossibilityof interferenceofsoilhumicacidwithqPCRanalysis,13 another 60soilsampleswereobtainedviadryswabsfromthesame

Table. Swabsandsoilsampleresultsforpresenceof Clostridiumtetani

Samplelocationandtype

Allsoil:

• Park samples

• Elementaryschoolsamples

• Parkswabs

• Elementaryschoolswabs

Positive samples

1/140(0.7%)

• 0/20 (0%)

• 0/60(0%)

• 1/15(1.7%)

• 0/45(0%)

Dogparkfecesswabs 6/20(30%)

Undergraduate campus,oxidizedmetaland concreteswabs 30/40(75%)

locations(urbanparkandelementaryschool)astheoriginal 80.Repeatanalysisofthe60dryswabsofthesoilrevealed onepositivefor Ctetani DNA.Therefore,amongthe140 analyzedsoilsamples(withtwodifferentmethodsof collection),one(0.7%)wasfoundtohave Ctetani DNA.To assessthesensitivityofourassay,wetagged22samplesof soilinsterilizedplasticbagswith Ctetani DNA,and16were identiﬁedaspositives(72.7%).Wetagged20dryswabsofsoil with Ctetani DNA,and11(55%)werepositive.

Wecollected40swabsofcombinedpublicoxidized concreteandmetalsurfacesfromauniversitycampus,aswell as20swabsofdogfecesatasinglepublicdogpark.Ofthese, six(30%)of20dogfecessampleswerepositivefor Ctetani DNA,and30(75%)of40swabsofoxidizedconcrete andmetalwerepositive.Acontrolwascreatedandevaluated bytagging20ofthemetalandconcretesampleswith Ctetani DNA.Ofthe20controlsamples,15were positive(75%).

Therewasastatisticallysignificantdifferenceinthe percentageofsampleswith Ctetani DNAfromtheconcrete andmetalgroup(75%)ascomparedtothesoilgroup(0.7%), P < 0.001.Additionally,therewasastatisticallysignificant differenceinthepercentageofsampleswith Ctetani DNA fromthedogfecesgroup(30%)ascomparedtothesoilgroup (0.7%), P < 0.001.Finally,therewasastatisticallysignificant differenceinthepercentageofsampleswith Ctetani DNAintheconcreteandmetalgroup(75%)as comparedtothedogfecesgroups(30%), P < 0.001.

DISCUSSION

Weundertookthisstudytohelpascertainthefrequency withwhich Ctetani isfoundinthesoilandonother substancesintheenvironmentinanurbanareaintheUS.Our resultssuggestthat Ctetani ismuchmorecommonon oxidizedmetalandconcrete,aswellasdogfecesthanitisin soil.OurresultsareconsistentwiththelastassessmentofUS soilfrom1926,whichdidnotstronglysuggestthat Ctetani waspresent.Withthepaucityof Ctetani isolatedinthis

sampleofUSsoilfromasinglecounty,itissuggestedthat furtherinvestigationintotheprevalenceofthebacteriumis needed.Therearebroaderimplicationstoidentifying Ctetani .Forone,thisstudyfoundthatnon-soilmediamay provideamorefavorablegrowthenvironmentfor Ctetani , andsoilitselfmaynotbeanabundantsource.Education effortsforexposuremayneedtobeconcentratedoninjuries duetorustedmetal,concrete,dogfeces,andpotentially othersourcesof Ctetani .Giventhatourstudysample wassmall,wedonotbelievethatthisdataalonemeritsa changeinEDvaccinationguidelines,especiallysincethe tetanustoxoidvaccinealsoprovidesimmunizationagainst diphtheriaandpertussis,bothofwhicharealsosigniﬁcant publichealththreats.Rather,moredatafromsimilarstudies isrequired.

LIMITATIONS

Therearesomelimitationstothissmallsingle-UScounty study.Firstly, Ctetani DNAsamplesweretestedviaqPCR analysis,ratherthanincubated,andreagents,suchashumic acid,withinthesoilmayhaveinterferedwithPCRanalysis.13 Repeatanalysisofdryswabsaimedtomitigatesucherror, andsimilarresultswereproduced.Furthermore,the prevalenceof Ctetani shouldbestudiedinotherurbanareas beforepublichealthconclusionsaremade.Farmlandand non-urbanareaswerenotstudiedand,therefore,thiscannot begeneralized.Oxidizedmetalandconcretesurfaceswere analyzedtogether,andthustheextenttowhich Ctetani is presentoneithersurfacewasnotfullyassessedinthis study.Lastly,whilethisstudycontainsmoredatasamples thanprevioussimilarundertakings,200soilsamplesfrom onlyafewseparatesitesinasingle,urbancountylikely donotfullyrepresentthetrueextentofthepresenceof Ctetani inotherenvironments,suchasothersitesfrom withinthesameurbancountyandotherurban,suburban, andruralenvironments.

CONCLUSION

Tetanusposesasignificantpublichealththreat.Yetits presenceinthesoilmaynotbeassignificantasiscurrently assumed,atleastnotinurbanareas,asour findingssuggest. Inourstudy,wetestedsoil,concrete,metal,anddogfecesfor Ctetani inasingleurbancounty.Theresultssuggestthat Ctetani ismoreabundantinoxidizedmetalandconcrete,as wellasindogfecesthanitisinsoil.However,several questionsabouttheprevalenceandvirulenceof Ctetani remain.Furtherstudiesshouldelucidatetheprevalenceof Ctetani inotherurban,suburban,andruralsites.

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.YenLMandThwaitesCL.Tetanus. Lancet(London,England). 2019;393(10181):1657–68.

2.CampbellJI,LamTMY,HuynhTL,etal.Microbiologiccharacterization andantimicrobialsusceptibilityof Clostridiumtetani isolatedfrom woundsofpatientswithclinicallydiagnosedtetanus. AmJTropMed Hyg. 2009;80(5):827–31.

3.KyuHH,MumfordJE,StanawayJD,etal.Mortalityfromtetanus between1990and2015: ﬁndingsfromtheglobalburdenofdisease study2015. BMCPublicHealth. 2017;17(1):179.

4.CentersforDiseaseControlandPrevention.Abouttetanusdisease (lockjaw).2022.Availableat: https://www.cdc.gov/tetanus/about/index. html.RetrievedApril27,2023.

5.LiangJL,TiwariT,MoroP,etal.Preventionofpertussis,tetanus, anddiphtheriawithvaccinesintheUnitedStates:recommendations oftheadvisorycommitteeonimmunizationpractices(ACIP). MMWRRecommRep. 2018;67(2):1–44.

6.HaversFP,MoroPL,HunterP,etal.Useoftetanustoxoid,reduced diphtheriatoxoid,andacellularpertussisvaccines:updated recommendationsoftheAdvisoryCommitteeonImmunizationPractices UnitedStates,2019. MorbMortalWklyRep. 2020;69(3):77–83.

7.Tetanus.Availableat: https://www.acep.org/siteassets/new-pdfs/ policy-statements/immunization-of-adults-and-children-in-theemergency-department.pdf.2020.Accessed:June12,2023.

8.DamonSRandPayabalLB.Distributionofthesporesofbacillus botulinusandbacillustetaniinthesoil:1.InMaryland, JInfectDis. 1926;39(6):491–501.

9.BukarA,MukhtarMD,AdamSA.Currenttrendinantimicrobial susceptibilitypatternof Clostridiumtetani isolatedfromsoilsamplesin Kano. BayeroJPureApplSci. 2008;1(1):112–5.

10.CenterforSustainableSystems.U.S.Citiesfactsheet.Availableat: https://css.umich.edu/publications/factsheets/built-environment/ us-cities-factsheet.Accessed:April27,2023.

11.CensusReporter.Miami-DadeCounty,FL.Availableat: http:// censusreporter.org/proﬁles/05000US12086-miami-dade-county-ﬂ/ Accessed:April27,2023.

AddressforCorrespondence:MichaelShalaby,MD,Herbert WertheimCollegeofMedicine,DepartmentofEmergencyMedicine andCriticalCare,4300AltonRd.,MiamiBeach,Florida33140. Email: michael.shalaby@msmc.com

12.AkbulutD,DennisJ,GentM,etal.Woundbotulismininjectors canofdrugs:upsurgeincasesinEnglandduring2004. EuroSurveill. 2005;10(9):172–4.

13.MathesonCD,GurneyC,EsauN,etal.AssessingPCRinhibitionfrom humicsubstances. OpenEnzymInhibJ. 2010;3(1):38–45.

EDUCATIONAL SPECIAL ISSUE:ORIGINAL

Yi-JuHo,MD*

Pei-ISu,MD*

Chien-YuChi,MD†

Min-ShanTsai,MD,PhD*‡

Yih-SharngChen,MD,PhD§∥

Chien-HuaHuang,MD,PhD*‡

*NationalTaiwanUniversityHospital,DepartmentofEmergencyMedicine, Taipei,Taiwan

† NationalTaiwanUniversityHospitalYun-LinBranch,DepartmentofEmergency Medicine,Yun-Lin,Taiwan

‡ NationalTaiwanUniversityHospital,DepartmentofInternalMedicine, Taipei,Taiwan

§ NationalTaiwanUniversityHospital,DepartmentofSurgery,Taipei,Taiwan

∥ NationalTaiwanUniversity,CollegeofMedicine,DepartmentofSurgery, Taipei,Taiwan

SectionEditor:EricMelnychuk,MD

Submissionhistory:SubmittedNovember28,2028;RevisionreceivedAugust1,2024;AcceptedAugust1,2024

ElectronicallypublishedOctober4,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18601

Background: Extracorporealcardiopulmonaryresuscitation(ECPR)improvestheprognosisofinhospitalcardiacarrest(IHCA).Thesix-factorRESCUE-IHCAscore(resuscitationusingECPRduring IHCA)wasdevelopedtopredictoutcomesofpost-IHCAECPR-treatedadultpatients.Ourgoalwasto validatethescoreinanAsianmedicalcenterwithahighvolumeandexperienceofECPRperformance andtocomparethedifferencesinpatientcharacteristicsbetweenthecurrentstudyandtheoriginal cohortina2022observationalstudy.

Method: Forthissingle-center,retrospectivecohortstudyweenrolled324ECPR-treatedadultIHCA patients.Theprimaryoutcomewasin-hospitalmortality.Weusedtheareaunderthereceiveroperating curve(AUROC)toexternallyvalidatetheRESCUE-IHCAscore.Thecalibrationofthemodelwastested bythedecilecalibrationplotaswellasHosmer–Lemeshowgoodness-of-ﬁtwithanassociated P-value.

Results: Ofthe324participants,231(71%)diedbeforehospitaldischarge.Thediscriminative performanceoftheRESCUE-IHCAscorewascomparablewiththeoriginallyvalidatedcohort,withan AUCof0.63.Aprolongeddurationofcardiacarrestwasassociatedwithanincreasedriskofmortality (oddsratio[OR]1.02,95%confidenceinterval[CI]1.01–1.03, P = .006).Aninitialrhythmofventricular tachycardia(OR0.14,95%CI0.04–0.51, P = .003),ventricular fibrillation(OR0.11,95%CI0.03–0.46, P = .003),andpalpablepulse(OR0.26,95%CI0.07–0.92, P = 0.04)wereassociatedwithareduced mortalityriskcomparedtoasystoleorpulselesselectricalactivity.Incontrasttotheoriginalstudy,age (P = 0.28),resuscitationtiming(P = 0.14),diseasecategory(P = 0.18),andpre-existingrenal insufficiency(P = 0.12)werenotassociatedwithin-hospitaldeath.

Conclusion: Inexternalvalidation,theRESCUE-IHCAscoreexhibitedperformancecomparabletoits originalvalidationwithinthesingle-centerpopulation.Furtherinvestigationonhospitalexperience,timeof-dayeffect,andspeciﬁcdiseasecategoriesiswarrantedtoimprovetheselectioncriteriaforECPR candidatesduringIHCA.[WestJEmergMed.2024;25(6)894–902.]

INTRODUCTION

Extracorporealcardiopulmonaryresuscitation(ECPR)is apromisingmodalitythatcombinesextracorporeal membraneoxygenation(ECMO)withtraditionalCPR techniquesforimprovingtheoutcomeaftercardiacarrest. Predictionmodelsdevelopedtoestimatethesurvival likelihoodofpatientswithrefractorycardiogenicshockor cardiacarrestwhoreceivedECPRhaverarelyfocusedon patientswhohavesustainedanin-hospitalcardiacarrest (IHCA),1,2 A2010observationalstudyshowedthatECPR leadstomorefavorableoutcomesinIHCAthaninout-ofhospitalcardiacarrest(OHCA),3 possiblyowingtothe shorterno-flowandlow-flowduration.Otherobservational studieshaveattemptedtoascertainstrongpredictorsthatcan helpidentifyIHCApatientswhowouldbenefitmost fromECPR.4,5

TheRESCUE-IHCAscoringsystem(resuscitationusing ECPRduringIHCA)wasdevelopedtopredictoutcomesof ECPR-treatedadultIHCApatientsandwasexternally validatedusingpatientdatafromtheExtracorporealLife SupportOrganization(ELSO)Registry.6 RESCUE-IHCAis asimplifiedscorethatcomprisessixvariables:1)age; 2)pre-existingrenalinsufficiency;3)timeoftheday (7 AM – 2:59 PM);4)diseasecategory(cardiac,ornon-cardiac, surgicalormedical,aspertheCurrentProcedural TerminologyandInternationalClassificationofDiseases); 5)initialrhythm;and6)thedurationofarrest,allofwhich canbeeasilycollecteduponhospitalarrival.However,inthe validationgroup,theRESCUE-IHCAscoringsystemonly demonstratedacceptablediscrimination.

Despiteitsmodestclinicalperformance,theRESCUEIHCAistheonlymodelavailableforpredictingoutcomesof ECPR-treatedIHCApatients.Therefore,furtherevaluation oftheRESCUE-IHCA’sreproducibilitybyusingexternal datasetsiswarrantedforwiderapplicationofthisscoring system.OurobjectivewastovalidatetheRESCUE-IHCA scoreusingdatafromadifferentpopulationandtoidentify potentialpredictorsthatmaydifferfromthoseintheoriginal study.Weaimedtoenhanceclinicaldecision-makingby providingmoreaccurateoutcomepredictionsforECPR initiationinIHCApatients.

METHODS

StudyDesignandSetting

Thisretrospectivecohortstudywasconductedovera seven-yearperiod(January2012–December2019)ata tertiary,extracorporeallife-supportreferralmedicalcenterin Taiwan,oneofthelargestmedicalcentersinAsiawith2,600 beds,including220bedsinintensivecareunits.Mostpatients areTaiwaneseresidents,withforeignersoccasionally admittedthroughinternationaltransfer.Overthepast decade,themedicalcenterhasperformedmorethan100 ECPRproceduresannuallyundertheguidanceofcardiac surgeons. 7 Thisstudy,approvedbytheinstitutionalreview

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Extracorporealcardiopulmonary resuscitationimprovestheprognosisofinhospitalcardiacarrest.TheRESCUE-IHCA scorepredictsoutcomesforthesepatients.

Whatwastheresearchquestion?

WeaimedtovalidatetheRESCUE-IHCA scoreandtocomparedifferencesinpatient characteristicsbetweenourstudyanda2010 observationalstudy.

Whatwasthemajor ﬁndingofthestudy?

TheRESCUE-IHCAscoreshowed compromiseddiscriminationcomparedtothe originalstudy,withanAUCof0.63(95%CI 0.56 – 0.70).

Howdoesthisimprovepopulationhealth?

TheRESCUE-IHCAscoredidnotpredict outcomesbetterthantheoriginallyvalidated cohort.Methodofdiseasecategorizationmay havein ﬂ uencedoutcomes.

board[202306052RIND],demonstratesarobustadherence tomethodologicalstandardsinhealthrecordreviewstudies.8 Therequirementofinformedconsentwaswaivedduetothe retrospectivenatureoftheresearch.Thesamplingpatients wereidentiﬁedthroughchartreviewofelectronic healthrecords(EHR)bymedicalandemergency physicianswhocollectedcovariatedataanddeﬁnedthe post-IHCAoutcome.

CaseSelection

BetweenJanuary2012–December2019,study participantswereenrolledbasedontheseselectioncriteria: 1)patientshadundergoneECPRfollowingIHCA;2)were aged ≥18years;and(3)hadnohistoryofOHCApriorto admission.Weusedacriticalscreeningprocesstoexclude ineligiblepatientsbasedonthefollowingcriteria:1)transfer toanotherhospitalafterreturnofspontaneouscirculation; 2)traumaticarrest;3)historyofOHCA;and4)missing outcomesintheEHR.

DataCollectionandProcessing

Covariatedatafromeachmedicalchartweredeﬁned clearlyandreviewedbyindependentphysicians,and monthlymeetingswereheldtoensureconsistencyofthe collecteddata.Tominimizepotentialbiasesorerrors,the

studydesignanddataanalysiswereundertakenbya physicianwhowasblindedtothedatacollectionprocess.We discussedanydisputesorambiguousrecordswith cardiologistsandemergencyphysicians,anddecisionswere maderegardingeachcontroversialhealthrecord.Individuals wholackedoutcomevariableswereexcludedinitially.The onlymissingdatainthecurrentcohortswasthepre-arrest laboratorydata,whichwasnotincludedinthe RESCUE-IHCAscoreorthe ﬁnalanalysis.

Variables

Wecategorizedthestudyvariablesintodemographics, pre-existingdiseases,intra-arrestcharacteristics,and presumedetiologyofcardiacarrest.Demographicsincluded age,gender,bodyweight,andbodymassindex(BMI).Preexistingdiseasesincludedhypertension,diabetesmellitus, chronicobstructivepulmonarydisease,coronaryartery disease,congestiveheartfailure,chronickidneydisease, cerebralvasculardisease,andcancer,andthediagnoseswere confirmedfromtheEHRbasedonregularmedication prescriptions,treatment,andoutpatientfollow-up.Intraarrestcharacteristicsincludedinitialcardiacrhythmdefined asasystole,pulselesselectricalactivity,ventricular tachycardia(VT),ventricular fibrillation(VF),orwitha palpablepulseinitially;timeofday,durationofcardiac arrest;andintra-arresttreatmentsincludingdefibrillation andmedicationsadministered.Thepresumedetiologyof IHCAwasdeterminedbyreviewingtheEHR.We categorizedparticipantsintofourgroupsbasedon whethertheIHCAwascardiogenicornon-cardiogenic andwasrelatedtoasurgicalormedicalillness (SupplementaryTable1).

Outcomes

AswiththeoutcomeoftheoriginalRESCUE-IHCA study,theprimaryoutcomeinthisstudywasin-hospital death.WecalculatedtheRESCUE-IHCAscoreinour datasetsfortheexternalvalidationprocess.

StatisticalAnalysis

Continuousdatawereassessedfornormalityusingthe Kolmogorov–Smirnovtestandwereexpressedasthemean (standarddeviation)ifnormallydistributedormedian (interquartilerange[IQR])ifnon-normallydistributed.We presenteddichotomousandcategoricalvariablesasthe frequency(percentage).Wecomparedcontinuousvariables usingtheMann-Whitney U test,whereasdichotomousand categoricalvariableswereexaminedusingthechi-square test.WeusedtheHosmer-Lemeshowtesttoshowthe goodnessof ﬁt.

TheexternalvalidationoftheRESCUE-IHCAscorewas performedinthestudycohort,andweassessed discriminatoryperformanceusingtheareaunderthereceiver operatingcharacteristiccurve(AUROC)with95%

conﬁdenceinterval(CI).Themodelcalibrationwastested usingacalibrationplotbasedon10deciles,aswellasthe Hosmer–Lemeshowgoodness-of-ﬁttestwithanassociated P -value.Wetestedindividualvariableswithabinarylogistic regressionmodelandadjustedtheminthemultivariate regressionmodelusingtheforce-entrymethod.Theresults werepresentedasadjustedoddsratioand95%CI.

WeperformedstatisticalanalysisusingtheStatistical PackagefortheSocialSciencesversion26.0(IBMCorp, Armonk,NY)andR4.3.0(RFoundationforStatistical Computing,Vienna,Austria).Atwo-sided P -valuelessthan 0.05wasconsideredstatisticallysigniﬁcant.

RESULTS

PatientCharacteristics

Duringthestudyperiod(January2012–December2019), 324eligiblepatientswhoreceivedECPRafterIHCAwere enrolledinthisstudy,andamongthem231(71%)diedbefore hospitaldischarge(Figure1). Table1 presentsthebaseline characteristicsoftheparticipants.Intheoverallcohort,121 patients(37.3%)hadaninitialshockablerhythm,and265 patients(82%)werepresumedtohaveamedicalillness. PatientswhosurvivedtodischargeafterreceivingECPRfor IHCA,whencomparedwiththenon-survivors,hadahigher frequencyofhypertension(62/93[66.7%]vs121/231[52.3%], P = 0.03),presentedmorefrequentlywithaninitial shockablerhythm(50/93[53.8%]vs71/231[30.7%], P < 0.001),hadashorterlow ﬂowduration(28minutesvs38 minutes, P < 0.001),andthecardiacarrestwasmore frequentlypresumedtobeofamedicalcardiogenicor surgicalcardiogenicorigin(62/93[66.7%]vs125/231[54.1%]; 7/93[7.5%]vs16/231[6.9%], P = 0.03).Thedurationof ECMOsupportwasshorter(2daysvs4days, P = .002),and thetotalhospitallengthofstaywaslonger(15daysvs3days, P < .001)inthesurvivalgroup.Nosigniﬁcantintergroup differencesbetweensurvivorsandnon-survivorswere detectedintermsofage,gender,bodyweight,BMI,history ofcomorbiditiesbesideshypertension,witnessedarrest,or timeofday.

Figure1. Studyenrollment ﬂowchart.

Table1. Comparisonofbasiccharacteristicsofin-hospitalcardiacarrestpatientsreceivingextracorporealcardiopulmonaryresuscitation withorwithoutsurvivaltodischarge.

Patientsdiedbeforedischarge(n = 231)Patientssurvivedtodischarge(n = 93) P

Pastcomorbidities

HTN121(66.1)62(33.9)0.03

DM91(67.9)43(32.1)0.27

Renalinsufﬁciency50(79.4)13(20.6)0.12 CVA13(81.2)3(18.8)0.57

Dichotomousvariableswerereportedasnumber(percentage)whilecontinuousvariableswerereportedasmedian(interquartilerange). BMI, bodymassindex; BW,bodyweight; CAD,coronaryarterydisease; CHF,congestiveheartfailure; CPR,cardiopulmonaryresuscitation; CVA,cerebrovascularevent; DM,diabetesmellitus; ECMO,extracorporealmembraneoxygenation; ECPR,extracorporealcardiopulmonary resuscitation; IHCA,in-hospitalcardiacarrest; IQR,interquartilerange; kg,kilograms.

ExternalValidationoftheRESCUE-IHCAScore

TheRESCUE-IHCApredictivemodelwasexternally validatedamongthe324participants.Themodel discriminationwaspoortoacceptable(areaunderthecurve 0.63[95%CI0.56–0.70]).Thepredictedprobabilityof mortalityrangedfrom38–93%accordingtotheRESCUEICHAscore(Figure2).Modelcalibrationindicatedgood ﬁt withtheHosmer-Lemeshowgoodness-of-ﬁttest(P = 0.91). Theobservedmortalityinthestudycohortvsthepredicted mortalitycalculatedfromtheRESCUE-IHCAscoreis

presentedin Figure3.Otherbinsizeswerelikewisetested withoutfurtherimprovementin ﬁt.

SigniﬁcantFactorsAssociatedwithIn-HospitalDeathin theCurrentCohort

Toassesspotentialpredictivefactorsinourcohortand comparethemwiththeoriginalRESCUE-IHCAscore,we conductedunivariatelogisticregressionforallvariables, followedbymultivariateregressionforthosevariableswith P < 0.1(Table2).Theresultshowedthatthemortalityrisk

Figure2. PredictedprobabilityofdeathacrossRESCUE-IHCAscore.

waspositivelyassociatedwithlongerlow-flowduration (oddsratio[OR]1.02,95%CI1.01–1.03, P = .006),and negativelyassociatedwithaninitialcardiacrhythmofVT (OR0.14,95%CI0.04–0.51, P = .003),VF(OR0.11,95%CI 0.03–0.46, P = .003),orpalpablepulse(OR0.26,95%CI 0.07–0.92, P = 0.04).Patient’sage,pre-existingrenal insufficiency,timingofresuscitation,anddiseasecategory didnotshowsignificantassociationswithmortality.The Hosmer-Lemeshowtestresultsshowedagood fit(P = 0.66).

DISCUSSION

ValidationoftheRESCUE-ICHAScore Inthepresentstudy,weperformedtemporaland geographicalexternalvalidationoftheRESCUE-IHCA scoringsysteminanAsianmedicalcenterequippedwith standardizedprotocolsforECMOinitiation.The performanceofmodeldiscrimination(AUC0.63)modestly decreasedascomparedwiththeoriginalderivationand validationcohorts(AUC0.72[95%CI0.68–0.76]and0.68

Figure3. Calibrationplotofobservedvspredictedmortalityfromvalidateddataset.

Table2. Logisticregressionmodelofriskfactorsassociatedwithin-hospitaldeath. N(%)/median(IQR)CrudeOR(95%CI) P AdjustedOR(95%CI) P

Male 227(70.1) 0.75(0.44–1.29) 0.30

Age(year)59.7(45.7–73.7)1.01(0.99–1.03)0.28

BW(kg)66.55(57.60–76.45)1.00(0.99–1.02)0.94

BMI(kg/m2)24.64(22.20–27.36)1.01(0.96–1.06)0.80

Pastcomorbidities

HTN183(56.5)0.55(0.33–0.91)0.020.70(0.41–1.20)0.19

DM134(41.4)0.76(0.47–1.23)0.26

COPD7(2.2)2.45(0.29–20.66)0.41

CAD118(36.4)0.72(0.44–1.18)0.19

CHF66(20.4)1.23(0.76–1.99)0.41

Renalinsufﬁciency63(19.4)1.70(0.88–3.30)0.12

CVA16(4.9)1.79(0.50–6.43)0.37

Cancer29(9)2.70(0.91–7.99)0.07

CPR

Witnessedarrest301(92.9)0.67(0.24–1.87)0.45

Presentingrhythm

Asystole42(13)referencereference

PEA19(5.9)0.29(0.06–1.45)0.130.32(0.06–1.70)0.18

VT83(25.6)0.11(0.03–0.37) <0.0010.14(0.04–0.51).003

VF34(10.5)0.10(0.03–0.38)0.0010.11(0.03–0.46).003

Pulse(+)146(45)0.24(0.07–0.81)0.020.26(0.07–0.92)0.04

Timeofday

07:00–14:59148(45.7)reference 15:00–10:59109(33.6)0.69(0.37–1.29)0.24

23:00–06:5967(20.7)1.19(0.60–2.35)0.63

Low ﬂowduration(min)37.6(17.0–58.3)1.03(1.01–1.04)0.0011.02(1.01–1.03).006

Presumeddiseasecategory

Dichotomousvariableswerereportedasnumber(percentage)whilecontinuousvariableswerereportedasmedian(interquartilerange). Variables withP < 0.1inunivariablelogisticregressionwereadjusted. BMI,bodymassindex; BW,bodyweight; CAD,coronaryarterydisease; CHF,congestiveheartfailure; CPR,cardiopulmonaryresuscitation; CVA,cerebrovascularevent; DM,diabetesmellitus; IQR,interquartilerange; kg,kilograms; N,casenumber; OR,oddsratio; PEA,pulseless electricalactivity; VF,ventricular ﬁbrillation; VT,ventriculartachycardia.

[95%CI0.61–0.75],respectively).Themodel’sperformance maybeattributedtothelackofsignificanceofcertain variables,includingage,timingofresuscitation,the presumeddiseasecategory,andpre-existingrenal insufficiency.Wefoundthatthelow flowdurationandthe initialcardiacrhythmserveassignificantpredictorsforthe outcome,consistentwith findingsfromprevious observationalstudiesandmeta-analyses.5,9–12,19 Despitethe single-centerfocus,thehospitalisgloballyrecognizedasthe

second-largestfacilityforECPRprocedures,managing hundredsofcaseseachyear.Thisstudyprovidesnovel insightswithinauniqueethniccontext.

IndividualPredictorsofIn-HospitalDeath

Agewasnotasigniﬁcantpredictorinourstudy,possibly attributabletothesmallsamplesize.Whencomparing patientcharacteristicsbetweenstudies,weobservedasimilar agedistributionamongnon-survivorsandsurvivorsinthe

twocohorts(studycohort:63.2vs59.9;RESCUE-IHCA cohort:61vs58).

Incontrasttothenotable ﬁndingfromtheRESCUEIHCAstudy,theinﬂuenceoftimeofdayonsurvivalno longerpersisted.Duringlatenightandearlymorningfrom 11 PM – 5 AM,fewersurvivorswereobservedintheRESCUEIHCAcohortcomparedtothecurrentcohort(studycohort: 16/93[17.2%];RESCUE-IHCAcohort:27/306[8.8%]).The currentstudywasconductedatatertiarymedicalcenter whereECPRinitiationwasprotocolizedregardlessof timing.Whenacardiacarrestoccurs,astandardized hospital-wideemergencycallactivatesarapidresponseteam ofmedicalcardiologists,surgeons,andemergencyphysicians whopromptlydeterminetheneedforECPR.

AcomprehensivediscussiononECPRimplementation wasprovidedbyaprospectiveobservationalstudy conductedinTaiwan.7 Experiencedsurgeonsandteam memberssubsequentlyestablishECMOcannulationatthe bedsideifindicated.Despitereducedwardstaffingduringthe night,survivalrateswerenotsignificantlyaffected.Aprior studyconductedatamedicalcenterinTaiwanrevealedthat thetimeofdayhadnoimpactonthesurvivaloutcome followinganin-hospitalcardiacarrest.13 Ourstudysuggests thatanexperiencedhealthcaresystemwithtrainedcrew membersoperatinginanestablishedsystemcaneffectively mitigatetheincreasedworkloadfromdecreasedstaffing duringoff-hours.Furtherstudiesarewarrantedtodetermine theimpactofhospitalcaseloadandexperiencewhilefocusing ontheoutcomesofcardiacarrestpatientswho receiveECPR.

TheRESCUE-IHCAstudyfoundthatsurgicalcardiac, surgicalnon-cardiac,andmedical-cardiacdiseaseswere predictivefactorsforsurvival.However,theresultsofthe currentstudyonlyobservedarelationshipbetweensurgical non-cardiacdiseaseandin-hospitalmortality,althoughthis associationdidnotreachstatisticalsignificance(Table1). Thesurgicalnon-cardiacdiseasesinourstudyincluded aorticdissection,hypovolemiaorhemorrhage,and intracranialhemorrhage,whichmaypotentiallyderiveless benefitfromECPR(SupplementaryTable1).The2010 RESCUE-IHCAstudyincludedahigherproportionof surgicalpatients(610/1,075[56.7%]),whereasthecurrent studycomprisedalowerpercentageofpatientswithsurgical illnesses(59/324[18.2%]).Thediseasecategoryintheoriginal RESCUE-IHCAstudywasautomaticallyretrievedfrom CurrentProcedureTerminology(CPT)andInternational ClassificationofDiseases(ICD)codes,whereasinthisstudy wemanuallyreviewedthehealthchartstoassesstheultimate etiologyofthecardiacarrest.

Previousstudieshaveidentiﬁed “cardiacorigin” or “presumedreversiblecardiogenicetiology” ascritical selectioncriteriaforIHCApatientsreceivingECPR,without specifyingthediagnosticprocessordisease-categorization method.7–9,15,16–18 Thesedivergent ﬁndingshighlightthe

complexityofthepresumedetiologyofarrest.Concerns persistregardingthepotentialmisinterpretationof diagnoses,whichcouldresultintheinappropriateinitiation ofECPRinacutescenarios.Establishingandvalidatinga standardizeddisease-categorizationsystemforIHCA patientsreceivingECPRisacrucialchallengethatmay signiﬁcantlyimproveoutcomesinthefuture.

Renalinsufficiencywasnotidentifiedasasignificant predictorformortalityinthisstudy.Ourstudyincluded fewerpatientswithpre-existingrenalinsufficiencyinboththe mortalityandsurvivalgroups(non-survivorsvssurvivorsin thestudycohort:50/231[21.6%]vs13/93[14.0%];RESCUEIHCAcohort:193/769[25.1%]vs49/306[16.0%]).Previous studiesconductedinanEastAsianpopulationexhibited similarproportionofpatientswithrenalinsufficiencytoour study.20,21 Theproportionofpatientswithrenaldisease influencestheresults;however,renalfunctionstillplaysa crucialroleinprognostication.

ApplicationoftheRESCUE-IHCAscorein real-worldscenarios

TheoverallsurvivalrateofIHCApatientstreatedwith ECPRwas28.7%inthestudy,whichapproximatestorealworldconditionsaccordingtorecentobservationalstudies andsystemicreviewswhereinreportedsurvivalratesranged from23.1–40%.22–25 Inconclusion,theperformanceofthe RESCUE-IHCAscorewasmodestlycompromisedinthis single-centercohort.Althoughasuboptimalvalidationof RESCUE-IHCAscoreinthiscohortmightnotindicateits prognosticperformanceinotherpopulationswithdifferent characteristics,suggestionsweremadeforpersonalized decisionsconsideringthepatient’sarrestetiology,clinical status,andtheinstitutionalcapacityandexperiences.

LIMITATIONS

Therewereseverallimitationstothisstudy.Firstly, missingdataisacommonissueinretrospectivelycollected variables,andthisstudywasnoexception.Patientswith missingoutcomeswereexcludedfromthebeginning, compromisingthesizeofthecohort.Secondly,thesmall samplesizemayhaveledtohighervariability,whichmight notaccuratelyreﬂectthereal-worldsituation.Thirdly, neurologicaloutcomeswerenotassessedduetoa considerableamountofmissingdata.Furtherlarge-scale validationstudiesshouldbeperformedtoconductedto assesstheuniversalapplicabilityofthisscore.

CONCLUSION

Inthecurrentstudycohort,theRESCUE-IHCAscoredid notpredictoutcomesbetterthantheoriginallyvalidated cohort,withlow ﬂowdurationandinitialrhythmspersisting asconsistentpredictivefactors.Themethodofdisease categorizationinIHCApatientsandthedifferencesin hospitalexperiencemayhaveinﬂuencedtheseoutcomes.

Althoughthesix-factorscorecarriedsomeadvantages, signiﬁcantlimitationswerepresent.Furtherresearchis neededtoexploretheimpactofhospitalexperienceand standardizeddiagnosticcriteriaforcardiacoriginIHCAon theECPRoutcomes.

Ethicscommitteeapproval:Thisresearchisapprovedby theInstitutionalReviewBoard(202306052RIND).The requirementforinformedconsentwaswaivedbecauseofthe retrospectivenature.

Availabilityofdataandmaterials:Thedatasetsusedor analyzedduringthecurrentstudyareavailablefromthe correspondingauthoronreasonablerequest.

Acknowledgments: Notapplicable.

AddressforCorrespondence:Chien-HuaHuang,MD,PhD,National TaiwanUniversityHospital,DepartmentofEmergencyMedicine, No.7,ZhongshanS.Rd.,ZhongzhengDist.,TaipeiCity100,Taiwan. Email: chhuang5940@ntu.edu.tw

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.SchmidtM,BurrellA,RobertsL,etal.PredictingsurvivalafterECMOfor refractorycardiogenicshock:thesurvivalafterveno-arterial-ECMO (SAVE)-score. EurHeartJ. 2015;36(33):2246–56.

2.OkadaY,KiguchiT,IrisawaT,etal.Developmentandvalidationofa clinicalscoretopredictneurologicaloutcomesinpatientswithout-ofhospitalcardiacarresttreatedwithextracorporealcardiopulmonary resuscitation. JAMANetworkOpen. 2020;3(11):e2022920.

3.KagawaE,InoueI,KawagoeT,etal.Assessmentofoutcomesand differencesbetweenin-andout-of-hospitalcardiacarrestpatients treatedwithcardiopulmonaryresuscitationusingextracorporeallife support. Resuscitation. 2010;81(8):968–73.

4.ShinTG,JoIJ,SimMS,etal.Two-yearsurvivalandneurological outcomeofin-hospitalcardiacarrestpatientsrescuedbyextracorporeal cardiopulmonaryresuscitation. IntJCardiol. 2013;168(4):3424–30.

5.ParkIH,YangJH,JangWJ,etal.Clinicalsigniﬁcanceoflow-ﬂowtimein patientsundergoingextracorporealcardiopulmonaryresuscitation: resultsfromtheRESCUEregistry. JClinMed. 2020;9(11):3588.

6.TonnaJE,SelzmanCH,GirotraS,etal.ResuscitationusingECPR duringin-hospitalcardiacarrest(RESCUE-IHCA)mortalityprediction

scoreandexternalvalidation. JACCCardiovascInterv. 2022;15(3):237–47.

7.ChenYS,LinJW,YuHY,etal.Cardiopulmonaryresuscitationwith assistedextracorporeallife-supportversusconventional cardiopulmonaryresuscitationinadultswithin-hospitalcardiacarrest: anobservationalstudyandpropensityanalysis. Lancet. 2008;372(9638):554–61.

8.WorsterA,BledsoeRD,CleveP,etal.Reassessingthemethodsof medicalrecordreviewstudiesinemergencymedicineresearch. Ann EmergMed. 2005;45(4):448–51.

9.D’ArrigoS,CacciolaS,DennisM,etal.Predictorsoffavourable outcomeafterin-hospitalcardiacarresttreatedwithextracorporeal cardiopulmonaryresuscitation:asystematicreviewandmeta-analysis. Resuscitation. 2017;121:62–70.

10.GrandboisvanRavenhorstC,SchluepM,EndemanH,etal.Prognostic modelsforoutcomepredictionfollowingin-hospitalcardiacarrestusing pre-arrestfactors:asystematicreview,meta-analysisandcritical appraisal. CritCare. 2023;27(1):32.

11.OhbeH,TagamiT,OguraT,etal.Low-ﬂowdurationandoutcomesof extracorporealcardiopulmonaryresuscitationinadultswithin-hospital cardiacarrest:anationwideinpatientdatabasestudy. CritCareMed. 2022;50(12):1768–77.

12.TonnaJE,SelzmanCH,GirotraS,etal.Patientandinstitutional characteristicsinﬂuencethedecisiontouseextracorporeal cardiopulmonaryresuscitationforin-hospitalcardiacarrest. JAmHeart Assoc. 2020;9(9):e015522.

13.ChenCT,ChiuPC,TangCY,etal.Prognosticfactorsforsurvival outcomeafterin-hospitalcardiacarrest:anobservationalstudy oftheorientalpopulationinTaiwan. JChinMedAssoc. 2016;79(1):11–6.

14.PatlollaSH,PajjuruVS,SundaragiriPR,etal.Hospital-leveldisparities inthemanagementandoutcomesofcardiacarrestcomplicatingacute myocardialinfarction. AmJCardiol. 2022;169:24–31.

15.DaubinC,BrunetJ,HuetJ,etal.Extracorporealcardiopulmonary resuscitationandsurvivalafterrefractorycardiacarrest:isECPR beneﬁcial? AsaioJ. 2021;67(11):1232–9.

16.StubD,BernardS,PellegrinoV,etal.Refractorycardiacarresttreated withmechanicalCPR,hypothermia,ECMOandearlyreperfusion(the CHEERtrial). Resuscitation. 2015;86:88–94.

17.TonnaJE,SelzmanCH,GirotraS,etal.Patientandinstitutional characteristicsinﬂuencethedecisiontouseextracorporeal cardiopulmonaryresuscitationforin-hospitalcardiacarrest. JAmHeart Assoc. 2020;9(9):e015522.

18.OhbeH,OguraT,MatsuiH,etal.Extracorporealcardiopulmonary resuscitationforacuteaorticdissectionduringcardiacarrest:a nationwideretrospectiveobservationalstudy. Resuscitation. 2020;156:237–43.

19.RadeschiG,MinaA,BertaG,etal.Incidenceandoutcomeofin-hospital cardiacarrestinItaly:amulticentreobservationalstudyinthePiedmont Region. Resuscitation. 2017;119:48–55.

20.WangCH,HuangCH,ChangWT,etal.Prognosticperformanceof simpliﬁedout-of-hospitalcardiacarrest(OHCA)andcardiacarrest hospitalprognosis(CAHP)scoresinanEastAsianpopulation:a prospectivecohortstudy. Resuscitation. 2019;137:133–9.

21.WangCH,ChenYS,MaMH.Extracorporeallifesupport. CurrOpinCrit Care. 2013;19(3):202–7.

22.AbramsD,MacLarenG,LorussoR,etal.Extracorporeal cardiopulmonaryresuscitationinadults:evidenceandimplications. IntensiveCareMed. 2022;48(1):1–15.

23.BlumensteinJ,LeickJ,LiebetrauC,etal.Extracorporeallifesupportin cardiovascularpatientswithobservedrefractoryin-hospitalcardiac arrestisassociatedwithfavourableshortandlong-termoutcomes:

apropensity-matchedanalysis. EurHeartJAcuteCardiovascCare. 2016;5(7):13–22.

24.LasaJJ,RogersRS,LocalioR,etal.Extracorporealcardiopulmonary resuscitation(E-CPR)duringpediatricin-hospitalcardiopulmonary arrestisassociatedwithimprovedsurvivaltodischarge:a reportfromtheamericanheartassociation’sgetwiththe guidelines-resuscitation(GWTG-R)registry. Circulation. 2016;133(2):165–76.

25.GravesteijnBY,SchluepM,DisliM,etal.Neurologicaloutcomeafter extracorporealcardiopulmonaryresuscitationforin-hospital cardiacarrest:asystematicreviewandmeta-analysis. CritCare. 2020;24(1):505.

EDUCATION SPECIAL ISSUE:BRIEF EDUCATIONAL ADVANCES

TeachingtheNewWays:ImprovingResidentDocumentation fortheNew2023CodingRequirements

NathanZapolsky,MD*

AnnemarieCardell,MD†

RiddhiDesai,DO*

StaceyFrisch,MD,MSEd‡

NicholasJobeun,DO,MSEd§

DanielNovak,DO*

MichaelSilver,MS*

ArleneS.Chung,MD,MACM∥

*MaimonidesMedicalCenter,DepartmentofEmergencyMedicine, Brooklyn,NewYork

† EmoryUniversityHospital,DepartmentofEmergencyMedicine,Atlanta,Georgia ‡ NewYorkUniversity,GrossmanSchoolofMedicine,NewYork,NewYork

§ IcahnSchoolofMedicineatMountSinai,NewYork,NewYork

∥ UniversityofVermontMedicalCenter,Burlington,Vermont

SectionEditor: BenjaminHoldenSchnapp,MD

Submissionhistory:SubmittedJune1,2024;RevisionreceivedJune12,2024;AcceptedJuly8,2024

ElectronicallypublishedSeptember19,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.21183

BACKGROUND

InJanuary2023,significantchangestothestructureofthe CurrentProceduralTerminologyCPT(R)evaluationand management(E/M)codes(hereforwardcalledthe2023E/M changes)wereimplementedforemergencydepartment(ED) encounters.Thesemodificationsaimtolessenadministrative workloadandaccuratelymatchcodingwithcontemporary patientcarepractices.1 Theyareanticipatedtoimpact roughly85%oftherelativevalueunitsofEDcare2 and,thus, alsohavesignificant financialimplicationsforEDs. Residentsprovidefront-linecareanddocumentationfor millionsofpatientsseeninUnitedStatesEDsannually.The ModelofClinicalPracticeofEmergencyMedicine3 identifies financialprinciples,suchasbillingandcoding,toberequired corecontentforboardcertification.Furthermore,the AccreditationCouncilforGraduateMedicalEducation (ACGME)includesqualityclinicaldocumentationtobeone ofthemilestonesthatdetermineadvancementinresidency training.4 Interventionsthatalleviatedocumentationburden arealsoassociatedwithimprovedphysicianwell-beingper theexistingliterature.5 However,researchsuggeststhatmost emergencymedicine(EM)residentsdonotreceiveformal traininginbillingandcodingandhaveknowledgegapsin thisarea.6–8

Historically,documentationofencountersintheED focusedonthenumberofelementswithinhistoryofpresent illness(HPI),reviewofsystems(ROS),physicalexam(PE), andmedicaldecision-making(MDM).Thesenewcoding guidelinesshiftthefocusalmostentirelytoMDM.They emphasizedocumentationofdifferentialdiagnoses; independentinterpretationofmedicaltesting;justiﬁcationof testingnotpursued;socialdeterminantsofhealth;chronic diseases;histories;communicationwithconsultants,

ancillarystaff,andprimarycare;andreviewof externalrecords.

OBJECTIVES

Wesoughttoimproveresidentunderstandingofand compliancewiththe2023E/Mchanges.Objectivesincluded identificationofthekeyelementsrequiredateachE/Mlevel, chartingandreceivingfeedbackofsampleencounters,and appreciationfortheimportanceofaccurateandhigh-quality documentation.Wesecondarilysoughttoinvestigate whetherourinterventionimprovedresidentwellness specificallyviabenefitsinconfidencetoperformaccurateand expeditiousdocumentationandcompletionofchartingina timelymanner.

CURRICULARDESIGN

OurcurriculumwasdevelopedusingKern’ssix-step approachtocurriculumdesign9 asapartoftheeducational qualityimprovementprocessatasingleEMresidency programbasedatasingle,large,tertiary-care,urbanhospital withanapproximateannualEDpatientcensusof110,000 fromOctober1,2022–February28,2023.Priortostudy initiationaneedsassessmentwasperformed.Key stakeholdersindepartmentalbillingandcodingwere identiﬁedandinterviewed,andrelevantliteraturewas reviewed.1–10 Thisincludedthehospitalchiefmedical informationofﬁcer,EDvicechair,andbillingandcoding leadership.Theinterviewsrevealedasharedopinionthat oftenthedocumentationtoreachtheappropriateexpected levelofservice(4or5)waslackingtosupportthatlevelof billingandmostofthatdocumentationshouldbecapturedin theMDMportionofthenote.Thus,theMDMportionof thenotewastargetedfortheintervention.

Oureducationalmethodsprimarilyusedin-person, ﬂipped-classroomsessions.Wedecidedtousea ﬂippedclassroomapproachforseveralreasons:1)toallowresidents togainexposuretothenewbillingcriteriapriortotheinpersonsessions;2)asamechanismtoassessresident understandingandskills,bothindividuallywithhomework responsesaswellasinagroupsetting;and3)tousefaculty’ s in-persontimeforoversightandfeedback.11 Wealsoapplied aspacedlearningapproachtomaximizeknowledge acquisitionandretention.12 Thesessionswereheldon December14,21,and28,2022.

Forpre-sessionhomeworkeachweek,allresidentswere providedasamplepatientHPI,ROS,andPEcomponents. Alllearnerswereprovidedthesamecase,andcaseswere changedeachweek.Caseswereformulatedtoinclude elementsthatcouldbeexpandeduponintheMDM, Residentswerealsoprovidedwiththe “CPTEvaluationand Management(E/M)CodeandGuidelineChanges” document.10 TheywerethenaskedtocreateanMDM sectioninaccordancewiththeabovedocument.Homework responseswerereviewedbyfaculty,andfeedbackwasgiven individuallyviaemail.Writtenfeedbackforresidentswas generatedusingatemplatebasedlargelyonthe2023E/M guidelineschanges.10 AnideallydocumentedsampleMDM sectionwasalsosuppliedforreference(Supplement1).13

Duringeach30-minutesession,residentsweredivided intosmallgroupsoffourandprovidedanexamplepatient case,whichincludedonlytheHPI,ROS,andPE components.ResidentsthencollaborativelywroteanMDM sectionforthecase.Allgroupswereprovidedthesamecase, andcaseswerechangedeachweektofocusondifferent aspectsoftheMDMsection.Eachsmallgroupofresidents sharedtheirresponsewiththelargergroupandwere providedpeerfeedbackundertheguidanceofafaculty facilitatorselectedfortheiradvancedknowledgeineither educationoroperations.Facilitatorswereprovidedin advancewithanexampleofanideallydocumentedMDM section,whichresidentswerealsoprovidedwithatthe conclusionoftheexercise.

IMPACT/EFFECTIVENESS

Weemployedapre-postinterventionalstudydesignusing aconveniencesampleofresidents,inwhichgroup assignmentwasbasedonthenumberoftrainingseach residentcouldattendduetoschedulingfactorsoutsidethe scopeofthisstudy.Thisstudywasdeterminedtobeexempt bytheinstitutionalreviewboardofMaimonidesMedical Center.Participationwasvoluntaryandanonymous.We evaluatedtheimpactofourbriefeducationalinterventionon subjectivemeasuresofEMresidentknowledge,skills,and attitudesviasurveyandonobjectivemeasuresofskillsand behaviorsbyassessingaggregatechartdata.

Surveysweredevelopedthroughagroupiterativeprocess thatincludedoneauthor(ASC)withexpertiseinsurvey

designmethodology.RedCap,14,15 hostedat[Maimonides MedicalCenter]wasusedtoanonymouslydistributeboth pre-andpost-interventionsurveys.Bothsurveysconsistedof sixLikert-scalequestions,threeregardingtheirreporteduse ofdocumentationshortcuts,andthreeassessingattitudes abouttheirownunderstandingofandpredictedskillwiththe newE/Mcodingchanges.Sixadditionalmultiple-choice questionsassessedknowledgeaboutdocumentationrules.A ﬁnalquestionwasforfeedbackandrequestedideasforother E/Mbillingandcodingeducation.Thepre-intervention survey,distributedNovember30,2022,differedfromthe post-interventionsurveyofDecember28,2022, onlyinaskingtheself-reportednumberofsessions attended.(Supplement2).

Residentskillswereassessedusingactualclinical documentation.ResidentaggregateE/Mlevelswereassessed acrossthreemonthspre-intervention(October1–December 31,2022)andtwomonthspost-intervention(January 1–February28,2023).Duetovariationinresidentclinical schedules,wechosetheabovetimeperiodstocapturethe greatestproportionoftheEDencountersdocumentedby residents.WeusedtheKirkpatrickmodeltoevaluateour intervention’simpact.16 Surveyswereusedtoassessresident subjectivereactions,andobjectiveknowledgeby identificationofbillableelementsinaprovidedsample MDM.Weusedactualclinicaldocumentationtoassess changesinbehavior.Specifically,weassessedwhether traineeshadastatisticallysignificantincrease(P < 0.05)in theproportionofE/Mlevel5charts(99285)andlikewisea significantdecreaseinlevel1,2,3,and4charts(99281, 99282,99283,99284).

Weuseddescriptivestatisticsandcomparisonofmeans withtheMann-WhitneyUteststratifiedbynumberof educationalsessionsattendedtoanalyzesignificant differencesinknowledgeandattitudesbeforeandafterthe intervention.Forknowledge,thesecalculationswere summarizedwithmedianandinterquartilerange(IQR)and comparedacrosstimeperiodsusinganexactWilcoxon signed-ranktest.ABonferronicorrectionforthesignificance oftheinterventionchangesthealphato0.01667.Foreach chartlevel(99281–99285),wecreatedlogisticregression modelsusinggeneralizedestimatingequationsforindividual repeatedmeasurestoaccountforpersonalvariability.The numberofattended flipped-classroomsessionswastreatedas theindependentvariable.Zerotrainingswereconsideredto bethepre-periodforanalysis.Allanalysesconsidered alpha ≤ 0.05tobestatisticallysignificantandwereconducted usingSPSSv28.015 (SPSSStatistics,IBMCorp, Armonk,NY).

Forty-sixofthe54EMresidents(85%)eligibleforthe studycompletedbothpre-andpost-interventionsurveys.All 54residentsparticipatedinatleastonesurvey.Duetoclinical schedules,someresidentswerenotpresentatoneormoreof thethreeofferedsessions.The ﬁrstlivesessionwasattended

by33(61%)residents,thesecondby38(70%),andthethird by40(74%).Six(11%)residentsattendedonesession, 15(28%)attendedtwosessions,and25(46%)attendedthree sessions.Eight(15%)didnotattendanysessions.

Residentsdemonstratedasignificantimprovementin knowledgeregardingwhichelementsarethekeytothe MDMwithinthe2023E/Mchanges[6(5–6.5)to8(7.5–8) P < 0.001],andbycorrectlyidentifyingthenumberand complexityofproblems,complexityofdata,risklevel,and theoverallcomplexityofasampleencounter.Therewasno statisticallysignificantimprovementinidentificationofthe importantcodingelements(4[3–5]to5[3.5–5], P = 0.38). Residentsalsoendorsedgreaterconfidenceintheirability todescribe(2[1–3]to4[3–4], P < 0.005),accurately document(3[2–3]to4[3–4], P < 0.005),andbill(2[2–3]to3 [2–3] P < 0.005).Therewerenosignificantchangesintheir opinionoftheirabilitytocompletetheirchartsinatimely manner(P < 0.19,CI0.165–0.215)inthedecisiontouse dictationsoftware(P = 1),shortcuts(P = 1),orcustom preparedtextphrases(P = 1)followingtheintervention. Residentsparticipatinginanynumberof flipped-classroom sessionsshowedsignificantchangesintheirskills,including anincreaseinE/MLevel5codedcharts,andasignificant decreaseinLevel1,2,and3codedcharts(P < 0.005).The increaseinLevel5chartsanddecreaseinLevel3chartswere significantafterjustonesession(Figure).Nosignificant changewasobservedinLevel4charts.

Tothebestofourknowledge,thisisthe ﬁrst studytodate todescribetheimpactofaneducationalinterventiononEM residentdocumentationknowledge,skills,andattitudes withintheframeworkofthe2023E/Mchanges.Naturally, ourexperienceandresultsatoursingleEMresidency

programbasedatalarge,urban,tertiary-carehospitalmay notbegeneralizable.Thisinterventiondataissinglecenter andpreliminary,andtheinterventionshouldundergo repetitionandcomparisonbefore ﬁrmconclusionscan bedrawn.

Wechosetocollectdataduringatimerangetiedtothe sameillnessseasontokeepthecaseacuitymixandattending/ residentstafﬁngcomparable.Weotherwisecouldhave comparedtothesamemonthsofthepreviousyearforpreinterventiondata,tobestmatchtheillnessseason,or alternatively,post-interventiondatacouldhavebeendrawn insteadfromthefollowingyear(2024)tohelpmitigate recencybiasintheinterventiongroup.Thatbeing said,themajordifferencesinresidentandattending stafﬁngbetweentimesayearapartcouldalsohave confoundedresults.

Weconsideredfacultysupplementaldocumentationand itseffectondocumentationoutcomesduringourstudy designandtookapragmaticapproach.Forthedurationof thisstudytheattendingpopulationwasstable,nosignificant changestoattendingeducationwereperformedduringthis period,andattendingstaffingremainedatbaselinewithno changestoratios,shiftdurations,orstandarddistributionsof encountersthroughouttheEDcareareas.Tofurtheraddress thisconcernweattemptedanalysisoftheattending distributionbetweenthesevariousgroups.Noattendinghad agreaterthan1.4%changeintheirbillingfrompre-topostintervention,andtheirsmallcontributionstotheoverall billingforeachinterventiongroupwas,therefore,unlikelyto havebiasedthelargedifferencesseenbetweengroups. However,thedifferencesindistributionofattendingshifts betweenthegroupsvariedstatisticallysignificantly,andbias

Figure. ProportionofeachCurrentProceduralTerminologyevaluationandmanagementlevelbynumberofeducationalsessionsattended.

cannotbeassessedwithoutpatient-levelbillingrecords.This couldbeconsideredinfuturestudies.

Ourprogrammayhaveimplicationsregardingwellnessas well.Residencytrainingmustprepareemergencyphysicians forallaspectsoftheireventualprofessionalexpectations. Residentsreceivingeducationexpressedgreaterconﬁdence intheirabilitytodescribe,accuratelydocument,andbillfor careprovided.Businessliteraturefrequentlynoteshowalack ofclearexpectationsincreasesworkstressandharms employeewellnessandproductivity.18 However,whether thisassociationappliestoemergencyphysiciansdeserves furtherstudy.

CONCLUSION

Overall,weobservedsigniﬁcantimprovementsinresident knowledge,attitudes,skills,andbehaviorsregardingclinical documentation.Wehopetoapplythesesuccessesandlessons learnedtotheformationofenduringeducationmaterialsat ourowninstitutionfordocumentationimprovementfor bothresidentsandattendings.

AddressforCorrespondence:NathanZapolsky,MD,Maimonides MedicalCenter,DepartmentofEmergencyMedicine,4802Tenth Ave.,Brooklyn,11219NY.Email: nizapo@gmail.com

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.ThoughtLeadershipTeam.2023E/Mcodingchanges.2023.Available at: https://www.aapc.com/resources/evaluation-management-codingchanges-2023.AccessedAugust18,2023.

2.McKenzieDAandGranovskyMA.2023documentationguideline changesforEDE/Mcodes99281–99285.2022.Availableat: www. acepnow.com/article/2023-documentation-guideline-changesfor-ed-e-m-codes-99281-99285/.AccessedJanuary25,2023.

3.BeesonMS,BhatR,BroderJS,etal.The2022modelofthe clinicalpracticeofemergencymedicine. JEmergMed. 2023;64(6):659–95.

4.CooneyRR,MuranoT,RingH,etal.Theemergencymedicine milestones2.0:settingthestagefor2025andbeyond. AEMEducTrain. 2021;5(3):e10640.

5.NguyenO,JenkinsN,KhannaN,etal.Asystematicreviewof contributingfactorsofandsolutionstoelectronichealthrecord-related impactsonphysicianwell-being. JAmMedInformAssoc. 2021;28(5):974–84.

6.RomansB,NguyenE,BiddleM,etal.Billingandcoding:disparitiesin healthcareprovidertraining. JAlliedHealth. 2022;51(1):43–6.

7.DawsonB,CarterK,BrewerK,etal.Chartsmart:aneedfor documentationandbillingeducationamongemergencymedicine residents? WestJEmergMed. 2010;11(2):116–9.

8.ChinS,LiA,BouletM,HowseK,etal.Residentandfamilyphysician perspectivesonbilling:anexploratorystudy. PerspectHealthInf Manag. 2022;19(4):1g.

9.ThomasPA,KernDE,HughesMT,etal. CurriculumDevelopmentfor MedicalEducation:ASix-stepApproach,3rded.Baltimore,MD:The JohnsHopkinsUniversityPress,2015.

10.AmericanMedicalAssociation.CPT® evaluationandmanagement (E/M)codeandguidelinechanges.2022.Availableat: https://www. ama-assn.org/practice-management/cpt/cpt-evaluation-andmanagement.AccessedMarch24,2024.

11.RotellarCandCainJ.Research,perspectives,andrecommendations onimplementingthe ﬂippedclassroom. AmJPharmEduc. 2016;80(2):34.

12.WeidmanJandBakerK.Thecognitivescienceoflearning:concepts andstrategiesfortheeducatorandlearner. AnesthAnalg. 2015;121(6):1586–99.

13.AllenA.Medicineclerkship(inpatient)H&Pexamples:H&P5.2018. Availableat: https://www.med.unc.edu/medclerk/wp-content/uploads/ sites/877/2018/10/hp5.pdf.AccessedFebruary24,2024.

14.HarrisPA,TaylorR,ThielkeR,etal.Researchelectronicdatacapture (REDCap) – ametadata-drivenmethodologyandworkﬂowprocessfor providingtranslationalresearchinformaticssupport, JBiomedInform. 2009;42(2):377–81.

15.HarrisPA,TaylorR,MinorBL,etal.TheREDCapconsortium:building aninternationalcommunityofsoftwarepartners, JBiomedInform. 2019;95:103208.

16.KirkpatrickDL. EvaluatingTrainingPrograms:TheFourLevels SanFrancisco,CA:Berrett-KoehlerPublishersInc,1994.

17.IBMCorp.IBMSPSSStatisticsforWindows,Version28.0.2021. Armonk,NY:IBMCorp.

18.BornsteinJ.Howalackofclearexpectationsleavesemployeesanxious andadrift.2023.Availableat: https://www.forbes.com/sites/ forbescoachescouncil/2023/06/05/how-a-lack-of-clear-expectationsleaves-employees-anxious-and-adrift/?sh=70de0f8f5983 AccessedApril23,2024.

TelesimulationUseinEmergencyMedicineResidencyPrograms: NationalSurveyofResidencySimulationLeaders

MaxBerger,MD,MEHP*

JackBuckanavage,MD†

JaimeJordan,MD,MAEd*

StevenLai,MD*

LindaRegan,MD,MEd‡

*UniversityofCaliforniaLosAngeles,DavidGeffenSchoolofMedicine, DepartmentofEmergencyMedicine,LosAngeles,California

† MountSinaiSchoolofMedicine,DepartmentofEmergencyMedicine, NewYork,NewYork

‡ JohnsHopkinsUniversity,DepartmentofEmergencyMedicine,Baltimore,Maryland

SectionEditors: KendraParekh,MDandBenjaminHoldenSchnapp,MD

Submissionhistory:SubmittedJune14,2024;RevisionreceivedSeptember20,2024;AcceptedSeptember26,2024

ElectronicallypublishedOctober22,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.24863

Introduction: Coronavirus2019(COVID-19)acceleratedtheneedforvirtuallearningincluding telesimulation.Manyemergencymedicine(EM)programshaltedin-personsimulationandtrialed telesimulation,butspeciﬁcsonitsutilizationandplansforfutureuseareunknown.Telesimulationhas beendeﬁnedas “aprocessbywhichtelecommunicationandsimulationresourcesareutilizedtoprovide education,training,and/orassessmenttolearnersatanoff-sitelocation.” Ourobjectiveinthisstudywas todescribethepatternsoftelesimulationusageinEMresidencyprogramsduringCOVID-19-induced learningrestrictionsaswellasitsanticipatedfutureutility.

Methods: WeidentiﬁedEMsimulationleadersviatheEMRAMatchwebsite,institutionalwebsites,or personalcontactwithresidencycoordinatorsanddirectors,andinvitedthemtoparticipatebyemail. Participantscompletedaconﬁdential,web-basedsurveyconsistingofmultiple-choiceitemsandone free-responsequestion,developedbyourstudyteamwithconsiderationofsurveyresearchbest practicesandMessick’svalidityframework.WecollecteddatabetweenJanuary–February2022.We calculateddescriptivestatisticsformultiple-choiceitemsandexaminedthefree-responseanswersfor commonthemes.

Results: Weobtainedcontactinformationforsimulationleadersat139EMresidencyprograms.Survey responseratewas65%(91/139).Duringin-personrestrictions,62%(56/91)ofprogramsused telesimulation.Assumingallrestrictionslifted,38%(34/90)ofrespondentsplannedtocontinuetouse telesimulation,comparedto9%(8/91)usingtelesimulationbeforeCOVID-19.Mostrespondents plannedtousetelesimulationformedicalknowledge(26/34,76%)andcommunication/teamworkfocusedcases(23/34,68%).Inresponsetothefree-responsequestionregardingexperiencewithand plansforuse,weidentiﬁedthreemajorthemes:1)telesimulationisavaluablealternativetoin-person learning;2)telesimulationisanoptionforlearnersunabletoparticipateinperson;and3)telesimulationis challengingforproceduraleducation.

Conclusion: DespitetherelativelylimiteduseoftelesimulationinEMresidenciespriortoCOVID-19,an increasednumberofprogramshaveplanstocontinueincorporatingtelesimulationintotheircurricula. Thisplanforcontinueduseopensopportunitiesforfurtherinnovationandscholarshipwithinsimulation education.[WestJEmergMed.2024;25(6)907–912.]

INTRODUCTION

Restrictionsimposedonin-personeducationduringthe coronavirus2019(COVID-19) pandemicacceleratedtheneed forvirtuallearning,includingtelesimulation.1,2 Telesimulation hasbeendeﬁnedas “aprocessbywhichtelecommunication andsimulationresourcesareutilizedtoprovideeducation, training,and/orassessmenttolearnersatanoff-sitelocation.”3 Initialapplicationswereinlowerresourcesettingssuchas developingcountrieswherelearnersdidnothaveaccessto simulationcentersorinstructors.4,5 Withintelesimulation, differentmodalitieshavebeendescribedthatvaryin ﬁdelityas wellaslocationofthelearnersandinstructorsrelativetoeach otherandthesimulationcenter.6–8

Severalpublishedarticlessinceearly2020havedescribed differentinstitutions’ approachestotelesimulationsincethe pandemic.1–2,9–13 Commonthemesincludetheneedtomodify learningobjectivestovirtualenvironmentsandtoselectthe appropriatemodalityoftelesimulationbasedoninstitutional needsandresources.9–12 Differentmodalitiesoftelesimulation havebeendescribed,includingthefollowing:1)learners virtuallyobservinganddebrieﬁngalivesimulation7; 2)learnerspresentwithamanikinwhileinstructorsfacilitate fromaseparatelocation6;3)instructorspresentwitha manikinwhilelearnersremotelyparticipate7;and 4)completelyremoteoptionwherelearnersandinstructors bothparticipateremotelyfromseparatelocations.10,11

Limiteddatacomparingtelesimulationvstraditional simulationsuggeststhatlearnersatisfactionwith telesimulationorhybridvirtualandin-personsimulationis similar,althoughthiswasnotfoundinallstudies.7,13,14

Ascopingreviewfrom2021highlightedthemixeddataon studentperceptionoftelesimulation,withsomeofthe includedstudiesindicatingremotefacilitationofsimulation beingperceivedasequallyormoreeffectivethanlive facilitation,whileothersfoundremotefacilitationtobe inferior.14 Facilitatorperceptionoftelesimulationhasnot beenwellstudied.Limitedlearningoutcomedatahas suggestedsimilarimprovementsbetweenin-person simulationandtelesimulation.8,14

Ourobjectiveinthisstudywastodescribethepatternsof telesimulationusageinemergencymedicine(EM)residency programsduringCOVID-19-induced,in-personlearning restrictionsaswellasitsanticipatedutilitymovingforward. Thisinformationiscrucialtounderstandingthevalueof telesimulationanditsutilityinmedicaleducation.

METHODS

StudyDesign,Setting,andPopulation

Weconductedacross-sectionalsurveystudyoffacultyin chargeofsimulationforEMresidencyprogramsinthe UnitedStates.WecollecteddatafromJanuary–February 2022.AfteridentifyingEMresidencyprogramsandtheir websitesfromtheEMRAMatchdatabase,15 wesearched eachwebsiteforcontactinformationforthedirectorof

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

COVID-19acceleratedtheneedforvirtual learningincludingtelesimulation.

Whatwastheresearchquestion?

Towhatextendwastelesimulationusedby EMresidenciesduringCOVID-19,andwhat isitsanticipatedutilitymovingforward?

Whatwasthemajor ﬁndingofthestudy?

Only9%(8/91)ofprogramsusedtelesimulation beforeCOVID-19.DuringCOVIDrestrictions, 62%(56/91)ofprogramsusedit,whileafter limitationswerelifted,38%(34/90)plannedto continuetelesimulation.

Howdoesthisimprovepopulationhealth?

Asanadjuncttotraditionalin-person simulationcurriculum,telesimulationisa viableoptiontoimprovemedicalknowledge andcommunication-basedcompetencies.

simulationeducation.Iftherewasnodirectordesignated,we emailedeachresidency’sprogramcoordinatorand/or directoraskingforcontactinformationforthefacultyin chargeofresidencysimulation.Eachprogramwasallowed onlyonedesignatedparticipant.Thisstudywasgivenexempt statusbytheUniversityofCalifornia,LosAngeles InstitutionalReviewBoard(IRB#21-001336)andtheJohns HopkinsUniversityHomewoodIRB(HIRB00013694).

SurveyDevelopmentandDissemination

Giventhelackofanypreviouslycreatedsurveyapplicable tothisconstruct,theprimaryauthor(MB)developedawebbasedsurveytoolwithconsiderationofsurveyresearchbest practicesandMessick’svalidityframework.16–19 Forcontent validityevidence,we firstperformedaliteraturereview,and theauthorgroupofexpertsimulationeducatorsandmedical educationresearchersreviewedthesurveyforclarityand relevancetotheconstruct.Wedefinedtelesimulationas includinganysimulationactivitywhere “telecommunication andsimulationresourcesareutilizedtoprovideeducation, training,and/orassessmenttolearnersatanoff-site location.”3 Wepilotedthesurveywithagroupofsimulation educatorswhowerenotincludedinthetargetsampleto gatherresponseprocessvalidityevidence.Afterpiloting,we revisedthesurveyforclarity.The finalsurveyincluded multiple-choiceitemsandonefree-textresponse item(Appendix1).

Weinvitedparticipantsbyemailandsenttwotargeted, follow-upinvitationstonon-respondersatbi-weekly intervals.WeadministeredthesurveyviaQualtrics (Qualtrics,LLC,Provo,UT).Noindividualidentifying informationwascollected.Tomaximizeresponserateand minimizeguessing,wedidnotrequireparticipantsto completeallsurveyitems.Participantswerenot compensatedforparticipatinginthestudy.

DataAnalysis

Wecalculatedandreporteddescriptivestatisticsforitems withdiscreteanswers.Weconductedcalculationsusing QualtricsandMicrosoftExcelforMac(MicrosoftCorp, Redmond,WA).Weexaminedtheanswerstothefree-text responsestoidentifycommonthemesthatwouldbroadenthe reader’sunderstandingofthedata.Successivewaveanalysis wasperformedtoassesstheextentofpossiblenonresponse bias.20 Weexaminedwhetheruseoftelesimulationduringthe pandemic,plannedfutureuseoftelesimulationafterinpersonrestrictions,andrespondentprogramformat (postgraduateyears[PGY]1–3vs1–4)differedbywave. Bivariatechi-squaretestsforeachvariableofinterestby wavewereperformedusingMicrosoftExcelforMac,and P -valueslessthan0.05wereconsideredstatisticallysigniﬁcant. Weusedtheconsensus-basedchecklistforreportingofsurvey studies(CROSS)asreportingguidelines(Appendix2).23

RESULTS

Of139simulationleadersweidentiﬁed,91(65%) completedthesurveywith87(63%)completingallitems.We reportdemographicdataforsurveyrespondentsin Table1, whilerespondents’ experienceandperceptionsof telesimulationareshownin Table2.PriortotheCOVID-19 pandemic,9%(8/91)ofsurveyrespondentswereusing telesimulationintheircurricula.Therewasawidevarietyof priorexperienceswithtelesimulation,withthemostcommon beingthattheyhadheardoftelesimulationbutneverbeen involved(44%).Ninety-twopercent(84/91)ofrespondents reportedthattheirinstitutionprohibitedin-personlearning activitiesatsomepointduringtheCOVID-19pandemic. Duringin-personlearningrestrictions,62%(56/91)used telesimulationinsomeform.

Whensurveyrespondentswereaskedaboutwhat format(s)oftelesimulationwereused,11%(10/90)statedthat theyonlyusedacompletelyvirtualoral-boardsstyleformat, whiletherestofthosewhousedtelesimulationreportedusing ahybridorvirtualformatinvolvingapatientmonitorand/or manikin.Thelargestpercentageofsurveyrespondentsfelt thatmedicalknowledgeandcommunication/teamworkfocusedcaseswerebestsuitedfortelesimulation(72%and 47%respectively),whilemostfeltthatproceduretraining wasnotwellsuitedfortelesimulation(62%).Thirty-eight percent(34/90)ofrespondentsstatedtheyplannedtouse telesimulationinsomeformintheircurriculummoving

Table1. Surveyrespondentdemographics. n(%)

Formatofrespondent’s current residencyprogram

PGY1–362/89(70%)

PGY1–427/89(30%)

Sizeofrespondent’scurrentresidencyprogram (totalnumberofresidentsinallyears)

≤20residents11/90(12%)

21–40residents38/90(43%)

41–60residents31/90(34%)

≥60residents10/90(11%)

Respondent’scurrentresidencyprogram primaryinstitutionsetting

University-based58/90(64%)

Community-based28/90(31%)

County-based13/90(14%)

Priorsimulationtrainingofsurveyrespondent

Fellowshiptraininginsimulation31/90(34%)

Non-fellowshiptraininginsimulation48/90(53%)

Noformaltraininginsimulation17/90(19%)

Respondentyearssinceresidencygraduation

≤522/90(24%)

6–1033/90(37%)

11–1516/90(18%)

16–208/90(9%)

≥2111/90(12%)

PGY, postgraduateyear.

forward,mostlyformedicalknowledgeandcommunication/ teamwork-focusedcases(76%and68%,respectively).

Wereceived14free-textresponses,andidentiﬁedthree majorthemes,describedbelowwithexemplarquotes.

1.Telesimulationisavaluablealternativeto in-personlearning: “Ithasbeenthe ‘betterthannothing’ optionbutaccepted bylearnerswhenotheroptionsarenotfeasible.” “Ithasexceededexpectationsinhowhelpfulithasbeen.”

2.Telesimulationisanoptionforlearnersunableto participateinperson:

“Wefoundthatit’sagreatoptionforresidentswith familiesorwhohaveotherextenuatingcircum-stances whytheycan’tparticipateinperson,ie,breastfeeding moms,newparents,eldercare,etc.Manyofour residentswhoarebetweennightsorbetweenmidshiftswilllogonandparticipate.”

3.Telesimulationischallengingforproceduraleducation: “Difﬁculttolearnmusclememoryforhighacuity,low occurrenceskills.”

Table2. Keysurveyresults. n(%)

EMresidencyprogramuseoftelesimulation

Prior toCOVID-19pandemic 8/91(9%)

Duringin-personlearningrestrictions56/91(62%)

Planneduseafterin-personrestrictionslifted34/90(38%)

DuringanypointintheCOVID-19pandemic, didyourinstitutionprohibitin-person learningactivities?

Yes 84/91(92%)

No 7/91(8%)

Experiencewithtelesimulationprior toCOVID-19

Hadneverheardoftelesimulation 17/91(19%)

Heardoftelesimulationbutneverinvolved40/91(44%)

Attendedapresentation 20/91(22%)

Participatedasaninstructor 19/91(21%)

Participatedasalearner 6/91(7%)

Conductedaresearchproject 5/91(5%)

Readapaperabouttelesimulation 16/91(18%)

Formatsoftelesimulationusedduring COVID-19restrictions

Completelyvirtual;utilizingreal-timepatient monitorand/ormanikin 21/90(23%)

Completelyvirtual;oralboardsstylecases31/90(35%)

Hybrid;instructor,learnersand/orsimtechin simcenterwhileothersremote 31/90(35%)

Whatsimulationactivitieswerebestsuited fortelesimulation?

Medicalknowledgefocusedcases 65/90(72%)

Communication/teamworkfocusedcases42/90(47%)

Procedurefocusedcases 5/90(6%)

Dedicatedproceduretraining 2/90(2%)

Proceduretrainingonhomemademodels10/90(11%)

Whatsimulationactivitieswerenotwellsuited fortelesimulation?

Medicalknowledgefocusedcases 0/87(0%)

Communication/teamworkfocusedcases18/87(21%)

Procedurefocusedcases 52/87(60%)

Dedicatedproceduretraining 54/87(62%)

Proceduretrainingonhomemademodels23/87(26%)

Percentoffuturesimulationcurriculum involvingtelesimulation

0%ofthecurriculum 56/90(62%)

1–25%ofthecurriculum 30/90(33%)

26–50%ofthecurriculum 3/90(3%)

51–75%ofthecurriculum 1/90(1%)

76–100%ofthecurriculum 0/90(0%)

(Continued onnextcolumn)

Table2. Continued. n(%)

Typesoffuturesimulationactivitiesforthose who plantocontinueusingtelesimulation

Medicalknowledge-focusedcases 26/34(76%)

Communication/teamwork-focusedcases23/34(68%)

Procedurefocusedcases 7/34(21%)

Dedicatedproceduraltraining 5/34(15%)

Proceduretrainingonhomemademodels5/34(15%)

EM, emergencymedicine.

“Proceduraltrainingwasthemostdifﬁculttosimulate viatelesim.”

Forthewaveanalysis,thestudyincluded91respondents, including42inwave1(46%),21inwave2(23%),and28in wave3(31%).Resultsoftheexaminedsurveyquestionsdid notstatisticallydifferbywavewithall P -values > 0.05.

(See SupplementalTable.)

DISCUSSION

DespiterelativelylowuseoftelesimulationwithinEM programspriortotheCOVID-19pandemic,wefoundthat manyEMresidencyprograms(62%)quicklyadaptedtoinpersonlearningrestrictionsbyusingtelesimulation.While notallprogramsthattrialedtelesimulationplantocontinue itsuse,38%ofrespondentprogramsdoplantocontinueto usetelesimulation,comparedto9%ofprogramsusing telesimulationpriortoCOVID-19.Thisrepresentsalarge increaseintheoverallusageoftelesimulationwithinEM residencies.Ourstudyalsoshedslightonhowtelesimulation canbeneﬁtEMprograms.Beingabletoincreaselearner participationtoincluderesidentswithfamilyobligationsor betweennightshiftscouldallowforincreasedreturnon investmentforsimulationresourcesandfacultytime.Most respondentswhoplantocontinuetousetelesimulation reportedthattheywilluseitasasmallpercentageoftheir overallsimulationcurriculum,whichhighlightsthat telesimulationisnotreplacingin-personsimulationbut augmentingthetraditionalcurriculum.Thiscouldbeina hybridformatthatallowsforincreasedparticipation,oras partofseparatetelesimulationdaysthatcouldreducethe travelburdensonlearnersandinstructors.

Therewaslargevariationinhowprogramsconducted telesimulationduringin-personrestrictions.Thisisinline withpriorliteratureandlikelyreﬂectsindividualprogram needs,preferences,andavailableresources.1,2,8–11,13,22 Most describedtelesimulationasbestsuitedformedicalknowledge andcommunication/teamwork-focusedcases,ratherthan forprocedureteaching.Thisisinterestinggiventhatearly descriptionsoftelesimulationintheliteraturemostly

involvedproceduralteaching.5,6 Onepossibleexplanation forthisdiscrepancyisthatthoseearlystudiesinvolved duplicatesimulatorsatremotelocations,anexpensethatis likelynotpractical,ornecessary,foraresidencyprogram giventheabilitytohostproceduretrainingaspartofthe in-personcurriculum.Whileitisapparentthatthereare increasedplansfortheuseoftelesimulationcomparedtothe pre-pandemicera,notallresidencyprogramswhoused telesimulationduringtimesofin-personrestrictionsare planningtocontinuetodoso.Thereasonsforthisare unknownbutmayrelatetotelesimulationresource availabilityorlimitedoutcomedataonitsutility.

Basedonourresults,webelievethattelesimulationcan continuetobeavaluableadditiontothetraditionalinpersonsimulationcurriculum,particularlyinallowingfor increasedparticipationoflearnersandinstructors,reducing resourcecostssuchassimulationcenterandstafftime,and allowingforaviableoptiontopracticemedicalknowledge andcommunication-basedcompetencies.Nowthat telesimulationhasbeenestablishedasaninstructional strategythatwillcontinuetobepartofmanyEMresidency curricula,itopensopportunitiesforfutureinnovationand scholarshipwithinsimulation-basedmedicaleducation. Additionalinvestigationcouldcomparedifferentmodalities oftelesimulationonobjectivelearningoutcomes.23 Itwould alsobeinterestingtoexploretheroleofvirtualand augmentedrealitywithintelesimulation.24,25

LIMITATIONS

Despitemultipleattempts,wewerenotabletoobtain contactinformationforasimulationleaderfromallEM programs.However,thebreakdownofPGY1–3vsPGY1–4 programsofsurveyrespondents(70%PGY1–3vs30%PGY 1–4),approximatingtheactualdistributionoftheEM residencyprograms(81%PGY1–3vs19%PGY1–4), suggeststhatthesamplecloselyresemblesthatofthe population.8 Givenourresponserateof65%,itispossible non-responsebiasaffectedourresults,withparticipantswith lessinterestorfamiliaritywithtelesimulationbeinglesslikely torespond.However,theresultsofoursuccessivewave analysisfailedtodetectnon-responsebiasfortheselected surveyquestions.

Theremaybeotherinﬂuencesaffectingaprogram’suseof telesimulationthatwewerenotabletocapture,andinthis surveystudyweexaminedonlytheopinionsoffacultyand notthoseofresidentlearners.Additionally,theliteraturebaseddeﬁnitionoftelesimulationweusedmaybeoverly broadandencompassmorethanwhattypicaleducators mightconsidertelesimulation.Finally,weacknowledgethat thesurveywasadministeredin2022within-personlearning restrictionsjuststartingtobelifted,andhowpeopleareusing telesimulationnowmaybechanging.Futureworkcould examinethisevolvinguseoftelesimulationwithinEM residencyprograms.

CONCLUSION

Thisstudydescribespastandplannedfutureuseof telesimulationwithinEMresidencyprograms.Alarge proportionofEMresidenciestrialedtelesimulationduring COVID-19-inducedrestrictions.Despiterelativelylowuseof telesimulationpriortothepandemic,moreEMprograms plantoincorporatetelesimulationmovingforwardasa limitedportionoftheiroverallsimulationcurriculum. Opportunitiesforfurtherinnovationandscholarshipwithin thisareaofsimulationeducationwillbepossiblegiventhis plannedcontinueduse.

ACKNOWLEDGMENTS

Thisprojectwascompletedinfulﬁllmentofthe requirementsfortheJohnsHopkinsUniversityMasterof EducationintheHealthProfessions.

AddressforCorrespondence:MaxBerger,MD,MEHP,Universityof CaliforniaLosAngeles,DavidGeffenSchoolofMedicine, DepartmentofEmergencyMedicine,1100GlendonAve.,Suite 1200,LosAngeles,CA90024.Email: MRBerger@mednet.ucla.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.PatelSM,MillerCR,SchiaviA,etal.Thesimmustgoon:adapting residenteducationtotheCOVID-19pandemicusingtelesimulation. AdvSimul.2020;5:26.

2.DiazMCGandWalshBM.Telesimulation-basededucationduring COVID-19. ClinTeach.2021;18(2):121–5.

3.McCoyCE,SayeghJ,AlrabahR,etal.Telesimulation:aninnovative toolforhealthprofessionseducation. AEMEducTrain 2017;1(2):132–6.

4.MikrogianakisA,KamA,SilverS,etal.Telesimulation:aninnovative andeffectivetoolforteachingnovelintraosseousinsertiontechniquesin developingcountries. AcadEmergMed.2011;18(4):420–7.

5.OkrainecA,HenaoO,AzzieG.Telesimulation:aneffectivemethodfor teachingthefundamentalsoflaparoscopicsurgeryinresourcerestrictedcountries. SurgEndosc.2010;24(2):417–22.

6.HaydenEM,KhatriA,KellyHR,etal.Mannequin-basedtelesimulation: increasingaccesstosimulation-basededucation. AcadEmergMed 2018;25(2):144–7.

7.McCoyCE,SayeghJ,RahmanA,etal.Prospectiverandomized crossoverstudyoftelesimulationvsstandardsimulationforteaching medicalstudentsthemanagementofcriticallyillpatients. AEMEduc Train.2017;1(4):287–292.

8.YasserNBM,TanAJQ,HarderN,etal.Telesimulationinhealthcare education:ascopingreview. NurseEduc.Today 2023;126:105805.

9.NadirNA,KimJ,CassaraM,etal.Simulation-basedemergency medicineeducationintheeraofphysicaldistancing:lessons learnedintheCOVID-19pandemic. AEMEducTrain 2021;5(3):e10586.

10.RayJM,WongAH,YangTJ,etal.Virtualtelesimulationfor medicalstudentsduringtheCOVID-19pandemic. AcadMed 2021;96(10):1431–5.

11.SanseauE,LavoieM,TayKY,etal.TeleSimBox:aperceived effectivealternativeforexperientiallearningformedicalstudent educationwithsocialdistancingrequirements. AEMEducTrain 2021;5(2):e10590.

12.ThomasA,BurnsR,SanseauE,etal.Tipsforconducting telesimulation-basedmedicaleducation. Cureus. 2021;13(1):e12479.

13.AustinA,RudolfF,FernandezJ,etal.COVID-19educational innovation:hybridin-personandvirtualsimulationforemergency medicinetrainees. AEMEducTrain. 2021;5(2):e10593.

14.HeffernanR,BrumptonK,RandlesD,etal.Acceptability,technological feasibilityandeducationalvalueofremotelyfacilitatedsimulation-based training:ascopingreview. MedEducOnline. 2021;26(1):1972506.

15.EmergencyMedicineResidents’ Association.EMRAMatch.Available at: https://webapps.emra.org/utils/spa/match#/search/map. AccessedJuly21,2021.

16.RickardsG,MageeC,ArtinoAR,Jr.Youcan’t ﬁxbyanalysiswhat you’vespoiledbydesign:developingsurveyinstrumentsandcollecting validityevidence. JGradMedEduc.2012;4(4):407–10.

17.GehlbachHandArtinoAR,Jr.Thesurveychecklist(manifesto). AcadMed.2018;93(3):360–66.

18.ArtinoAR,Jr.,DurningSJ,SklarDP.Guidelinesforreportingsurveybasedresearchsubmittedtoacademicmedicine. AcadMed 2018;93(3):337–40.

19.MessickS.Validity.In:LinnR.L,editor. EducationalMeasurement 3rd ed.NewYork,NY:Macmillan;1989:13–103.

20.DuszynskiTJ,FadelW,DixonBE,etal.Successivewaveanalysisto assessnonresponsebiasinastatewiderandomsampletesting studyforSARS-CoV-2. JPublicHealthManagPract. 2022;28(4):E685–91.

21.SharmaA,MinhDucNT,LuuLamThangT,etal.Aconsensus-based checklistforreportingofsurveystudies(CROSS). JGenInternMed. 2021;36(10):3179–87.

22.LevineM,SinghM,RestivoA,etal.Recommendationsforoptimizing virtualsimulation:atrialanderrorprocessfromtheCOVID-19 pandemic. JGradMedEduc.2022;14(1):18–21.

23.KirkpatrickJDandKirkpatrickWK. Kirkpatrick’sFourLevelsofTraining Evaluation.Alexandria,VA:ATDPress;2016.

24.MistryD,BrockCA,LindseyT.Thepresentandfutureofvirtual realityinmedicaleducation:anarrativereview. Cureus. 2023;15(12):e51124.

25.PallaviciniF,PepeA,ClericiM,etal.Virtualrealityapplicationsin medicineduringtheCOVID-19pandemic:systematicreview. JMIRSeriousGames. 2022Oct25;10(4):35000.

EDUCATIONAL SPECIAL ISSUE:BRIEF

PalliativeCareBootCampOffersSkillBuildingfor EmergencyMedicineResidents

JulieCooper,MD* JennaFredette,MD†

*UniversityofPennsylvania,DepartmentofEmergencyMedicine,Philadelphia,Pennsylvania † ChristianaCare,DepartmentofEmergencyMedicine,Newark,Delaware

SectionEditors:PaulLoganWeygandt,MDandBenjaminHoldenSchnapp,MD

Submissionhistory:SubmittedJune22,2023;RevisionreceivedMarch3,2024;AcceptedMay30,2024

ElectronicallypublishedSeptember6,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18381

BACKGROUND

Emergencymedicine(EM)residentsroutinelycarefor criticallyillpatientsinboththeemergencydepartment(ED) andintensivecareunits.1 Proﬁciencyinprimarypalliative careskillsisessentialforallemergencyclinicians.2,3 However,asigniﬁcantnumberofresidentslackexposureto formaleducationandtraininginpalliativecare.4,5 Moreover,educationandtraininginpalliativecare encompassesseveralAccreditationCouncilforGraduate MedicalEducation(ACGME)competenciesincluding systemnavigationforpatient-centeredcare,understanding thephysician’sroleinthehealthcaresystem,patient-and family-centeredcommunication,andinterprofessionaland teamcommunication.6

Currentcurriculaaddressingprimarypalliativecareskills inEMarenotablylimited.7–12 Historically,ourresidency experiencedinconsistenciesintheteachingofprimary palliativecareskills.Theyweresporadicallycoveredduring regularconferencesorlefttodeveloporganicallyovertime. Furthermore,postgraduateyear-2(PGY-2)residents,who primarilymanageseriouslyillpatients,foundthemselves engaginginchallengingserious-illnessconversationswith patientsandfamilieswithlittletonotraining.Recognizing theimperativeformorecomprehensiveeducation,we introducedafour-week,intensiveprimarypalliativecare curriculumspeciﬁcallytailoredforEMPGY-2residentsthat wasentitled “PalliativeCareBootcamp.”

OBJECTIVES

Theoverallobjectiveofthebootcampwastointroduce andstrengthenprimarypalliativecareskillsamongPGY-2 residentsatanindependentacademicmedicalcenter.Atthe endofthecurriculum,residentswouldbeableto1)deﬁnethe scopeofhospiceandpalliativemedicine;2)understandwhat primarypalliativecareskillsarefornon-specialtytrained physicians;3)recognizeEDpatientswithpalliativecare needs;4)implementahospiceevaluation;5)understandhow interdisciplinaryteamsareinvolvedinthecareofseriouslyill patients;and6)buildcommunicationskillsfordiscussing goalsofcare(GOC).

CURRICULARDESIGN

Thecurriculumandassessmentwereexemptfromthe institutionalreviewboard.UsingKern’ssix-stepapproachto curriculumdevelopment,wecreatedanintroductory primarypalliativecarecurriculum.AnEMfacultymember withaninterestinpalliativecareandresidencyleadership collaboratedtodevelopthecurriculum.Theresidency programendorsedthecurriculumasitalignedwitha curriculumredesigntoincludemorePGY-speciﬁceducation.

Thecurriculumwasinitiallydevelopedin2017.The interdisciplinarypalliativecareteamatthestudyinstitution servedascontentexperts.Theteamperformedabroad reviewoftheresidencycurriculumandprioritizedhigh-yield topicstailoredtothelocalcontext.Sessionswere scheduledduringweeklyconferenceandspannedfour consecutiveweeks.Thisscheduleallowedforanintensive experienceandallowedforrapidskilldevelopment.The curriculumisstrategicallydeliveredearlyinthePGY-2year toleverageresidents’ existingexperienceincaringfor seriouslyillpatientsandfacilitatemeaningfulreﬂection andinquiry.

Thecurriculumisstructuredintwophases(Table1).The ﬁrstphasespansthreeweeksandconsistsofthreetwo-hour sessions.Thesesessionsarededicatedtoprimarypalliative carefundamentalssuchasanintroductiontopalliativecare, prognosisandtrajectory,andnon-painsymptom management.SessionfacilitatorsincludedtheEMfaculty contentexpertaswellasmembersoftheinstitutional palliativecareteam,thedirectorofchaplaincywho specializedinfamilysupport,thedirectorofpalliativecare, andthepalliativecarefellow.Eachsessionencompasseda didacticsegment,interactivecase-baseddiscussionsusing scenariospreparedbyfacilitatorsorcontributedbyresidents, andopportunitiesforresidentquestions.

Inthe ﬁnalweek,residentsengagedinafour-hoursession inthesimulationcenter.ThissessionwasledbytheEM contentexpertwhoisatrainedfacilitatorwithVitalTalk,a nationalnon-proﬁtthatpromotesevidence-basededucation inserious-illnesscommunication.13 Thissessioninvolves usingastandardizedpatient.Residentsareassignedtoa

Table1. Thebreakdownofpalliativecarebootcampsessionsbyhourdetailingthetopic,learningobjectives,mappingtoACGME* competencies,andtheformatofthesession.

HourTopicObjectives

1Introtoprimarypalliative care inemergency medicine

Deﬁneprimarypalliativecareandidentify commonEDpresentationsofpatientswith unmetpalliativecareneeds.

Deﬁneadvancecareplanning,goalsofcare, codestatusandtreatmentlimitationsand describehowthesearecodiﬁedinlegaland medicaldocuments

InterpretaPOLST(PhysicianOrdersforLife SustainingTreatment)formanddescribeits useinacutecaresettings

2PrognosisandtrajectoryDescribefourcommontrajectoriesoflifelimitingillness

Deﬁneprognosisanddescribe3strategiesto assessprognosisinEDpatientswithserious illness

ACGME competenciesFormat/facilitator

Systemnavigationfor patientcenteredcare

Physicianrolein healthcaresystems

Lecture – EMfaculty contentexpert

Diagnosis,treatment, andclinicalreasoning

Case-basedlearning –EMcontentexpert

3Chaplainchat

4Non-painsymptom management

5Askaconsultant

6Introtohospice

7–10Seriousillness communicationworkshop (VitalTalk)

Describetheroleofthechaplaininthe interdisciplinarycareofseriouslyillpatients intheED

Chooseappropriate ﬁrst-andsecond-line treatmentforseriouslyillpatientsexperiencing nauseaandvomitingintheED

Chooseappropriate ﬁrst-andsecond-line treatmentforseriouslyillpatientsexperiencing dyspneaintheED

Chooseappropriate ﬁrst-andsecond-line treatmentforseriouslyillpatientsexperiencing constipationintheED

DescribetheroleoftheHPMclinicianinthe careofseriouslyillpatientsinthehospital

UnderstandtheroleofHPMconsultationin theemergencydepartment

Describethescopeofhospiceservicesand thesettingswhereitcantakeplace

Identifypatientswhomayqualifyforhospice andhowtogetthemevaluated

Providegoalconcordantcaretopatients enrolledinhospicewhopresenttotheED

Practiceskillsassociatedwithgoalsofcare conversationswithasimulatedpatient.

Systemnavigationfor patient-centeredcare

Interprofessionaland teamcommunication

Pharmacotherapy

Diagnosis,treatment, andclinicalreasoning

Case-based learning – chaplain

Case-basedlearning –hospitalpalliative carespecialist

Interprofessionaland teamcommunication

Systemnavigationfor patient-centeredcare

Physicianrolein healthcaresystems

Patient-andfamilycentered communication

Case-basedlearning –hospitalpalliative carespecialist

Lecture – community hospicemedicaldirector

Simulationand standardizedpatient skills-basedpractice –EMcontentexpert

*ACGME, AccreditationCouncilforGraduateMedicalEducation; ED,emergencydepartment; EM,emergencymedicine; HPM,hospiceand palliativemedicine.

smallgroupandtheyrole-playdeliveringseriousnewswith EM-basedscenarios.Thissessionbuildsskillsaround deliveringseriousnews.

Thecurriculumunderwentiterativeadjustmentsinformed byinformalfeedbackfrombothfacilitatorsandresidents.

Modiﬁcationsweremadebasedonfacilitatoravailability andinterest,resultingintheinclusionormodiﬁcationof topics,whilecertainsubjects,suchasopioidpain management,wereremovedduetoredundancyinother educationalsettings.

SURVEYDEVELOPMENT

Beforeimplementingthecurriculum,wecreatedabrief, pre-bootcampsurveytoassessresidents’ priorexposureand familiaritywithpalliativecare.Subsequently,twopostsurveyswereusedtogaugeresidents’ perceptionsregarding theachievementofsession-specificgoals.Wedevelopedthe firstsurveytoevaluatethe firstthreeweeksofthebootcamp. Theinitialdevelopmentcollectedallpotentialsurveyitems thatwererefinedthroughexpertconsultation.Thesurvey useda five-pointLikertscalerangingfrom1(strongly disagree)to5(stronglyagree).Thesurveyitemshadbeen pilottestedandrefinedinprecedingyearstoensurequestion clarity(Appendix1).

Asecondsurvey,whichwasusedforthesimulation-based session,promptedresidentstoratetheirself-assessed confidencesurroundingthespecificskillsonconducting GOCconversationscoveredinthesession(Appendix2).The surveyusesa five-pointLikertscalerangingfrom1(notvery confident)to5(veryconfident).

IMPACT/EFFECTIVENESS

Thecurriculumevaluationtookplaceduringthe2022 bootcamp.Eachsessionhadanaverageof8–10PGY-2 residents,ofatotal17potentialparticipants.Attendance variedfromweektoweekduetoexcusedabsences. Participationinbothpre-andpost-surveyswasvoluntary.Of theeligibleresidents,nine(52%)completedthepre-survey, revealingthatallbutoneresidenthadpriorexposuretoa palliativecarerotationduringmedicalschool,and7of9 respondents(77%)reportedpreviouscommunicationskills trainingduringtheirPGY-1year.

Post-interventionsurveyswerecollectedaftereach session,withcompletionratesrangingfrom25%(2/8 participants)to70%(7/10participants)persession.Notably, allrespondentsindicatedagreementorstrongagreement withtheachievementofeachsession’sobjectives.Forthe simulation-basedcommunicationsession,88%(8/9)reported increasedconfidenceoverall,88%(8/9)ofresidentsreported increasedconfidenceinrespondingtostrongemotions,and 100%(9/9)reportedenhancedconfidenceinelicitingpatient goalsandvalues.

TIPSFORSUCCESS/CHALLENGES/LESSONS LEARNED

Severalkeythemesemergedregardingthe implementationofabootcampcurriculuminprimary palliativecareforEMresidents.Onenotableadvantageof thiscurriculumisitslongitudinalformat,spanningfour consecutiveweekswithshortintervalsbetweensessions.This structureaffordsresidentstheopportunitytopracticenewly acquiredskillswhileactivelyworkinginclinicalsettings, fosteringcontinuousreﬂectionandreﬁnementoftheir abilities.Additionally,thecurriculumisadaptableand

enablesitsimplementationinprogramslackingEM palliativecare-trainedfaculty.Programscanuselocal resourcessuchasinstitutionalpalliativespecialists, interdisciplinarypalliativeteams,orseveralpublicly availableonlineresources.9,10,14

However,despiteitsstrengths,ourcurriculumfaces severalchallenges.Notably,residentsunabletoattend sessionsriskmissingvaluableeducationalopportunities,as thecurriculumisnotrepeatedduringtheacademicyear. Moreover,limitedopportunitiesforongoingskillacquisition andfeedbackoutsidescheduledsessionsmayhinder residents’ abilitytofullyintegratepalliativecareprinciples intotheirpractice.Furthermore,individualprogramsmay beunwillingtoinvest10hoursofcurriculumtothisspeciﬁc topicandskillset.Lastly,whiletherewasnocostforthe simulationtimeandstandardizedpatientsatthestudy institution,theremaybecostassociatedwiththisinother programsandthismustbeconsidered.

Furthermore,whileparticipantsexpressedsatisfactionwith thecurriculum,theoutcomesdatalacktherigornecessaryto deﬁnitivelyestablishitssuccess.Theimpactofthiscurriculum onlong-termknowledgeorclinicalbehaviorwithintheED remainsuncertain.Itwillbeimportanttoconductmore formalassessmentsofthecurriculumobjectivesandto evaluateitsapplicationintheclinicalsetting.

CONCLUSION

Astheroleofprimarypalliativecareinemergency medicinecontinuestoevolve,thereisagrowingneedto integratetheseessentialskillsandconceptsintoallEM residencies.Thebootcampformathasproventobea valuableeducationaltoolinourprogram,andits effectivenesswarrantsfurtherexplorationanddissemination withinthebroaderEMcommunity.

AddressforCorrespondence:JulieCooper,MD,Universityof Pennsylvania,DepartmentofEmergencyMedicine,3400Spruce St.,Philadelphia,PA19104.Email: julie.cooper@pennmedicine. upenn.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.SmithAK,McCarthyE,WeberE,etal.HalfofolderAmericansseenin emergencydepartmentinlastmonthoflife;mostadmittedtohospital, andmanydiethere. HealthAffairs. 2012;31(6):1277–85.

2.ShoenbergerJ,LambaS,GoettR,etal.Developmentofhospiceand palliativemedicineknowledgeandskillsforemergencymedicine residents:usingtheAccreditationCouncilforGraduateMedical Educationmilestoneframework. AEMEducTrain. 2018;2(2):130–45.

3.QuillTEandAbernethyAP.Generalistplusspecialistpalliative care creatingamoresustainablemodel. NEnglJMed. 2013;368(13):1173–5.

4.BackAL,FrommeEK,MeierDE.Trainingclinicianswithcommunication skillsneededtomatchmedicaltreatmentstopatientvalues. JAmerican GeriatricsSociety. 2019;67(S2):S435–41.

5.BaylisJ,HarrisDR,ChenC,etal.Palliativeandend-of-lifecare educationinCanadianemergencymedicineresidencyprograms:a nationalcross-sectionalsurvey. CJEM. 2019;21(2):219–25.

6.AccreditationCouncilforGraduateMedicalEducation.Emergency medicinemilestones.2021.Availableat: https://www.acgme.org/ specialties/emergency-medicine/milestones/ AccessedFebruary7,2024.

7.LoffredoAJ,ChanGK,WangDH,etal.UnitedStatesbestpractice guidelinesforprimarypalliativecareintheemergencydepartment. Ann EmergMed. 2021;78(5):658–69.

8.KrausCK,GreenbergMR,RayDE,etal.Palliativecareeducationin emergencymedicineresidencytraining:asurveyofprogramdirectors, associateprogramdirectors,andassistantprogramdirectors. JPain SymptomManage. 2016;51(5):898–906.

9.NorthwesternMedicine.EPEC:Educationinpalliativeandend-of-lifecare.Availableat: https://www.bioethics.northwestern.edu/programs/ epec/index.html.AccessedFebruary7,2024.

10.NguyenDandMateseT.Assessingtheeffectofahospiceandpalliative medicinecurriculumonemergencymedicineresidents’ knowledge. AmJHospPalliatCare. 2023;40(5):462–7.

11.StanichJ,SungaK,Loprinzi-BrauerC,etal.Teachingpalliativecareto emergencymedicineresidentsusinggamiﬁeddeliberatepracticebasedsimulation:palliativegamingsimulationstudy. JMIRMedEduc. 2023;9:e43710.

12.BeneschTD,MooreJE,BreyreAM,etal.Primarypalliativecare educationinemergencymedicineresidency:amixed-methodsanalysis ofayearlong,multimodalintervention. AEMEducTrain. 2022;6(6):e10823.

13.VitalTalk.VitalTalk.Availableat: https://www.vitaltalk.org/. AccessedFebruary21,2024.

14.CentertoAdvancePalliativeCare.CAPCclinicaltraining. Availableat: https://www.capc.org/clinical-training/ AccessedFebruary7,2024.

ORIGINAL RESEARCH

ImprovingPatientUnderstandingofEmergencyDepartment DischargeInstructions

SarahRussell,MD

NancyJacobson,MD

AshleyPavlic,MD,MA

MedicalCollegeofWisconsin,DepartmentofEmergencyMedicine,Milwaukee,Wisconsin

SectionEditor:MarkI.Langdorf,MD,MHPE

Submissionhistory:SubmittedNovember14,2023;RevisionreceivedMay30,2024;AcceptedJune12,2024

ElectronicallypublishedSeptember24,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18579

Introduction: Previousstudieshaveshownthatpatientsintheemergencydepartment(ED)are frequentlygivenincompletedischargeinstructionsthatarewrittenatleastfourgradelevelsabovethe recommendedsixth-gradereadinglevel,leadingtopoorunderstanding.Ouraimsinthisstudywereto implementstandardizeddischargeinstructionscontainingsixkeycomponentswrittenatamore appropriatereadinglevelforcommonemergencydepartment(ED)diagnosestoimprove patientunderstanding.

Methods: Weconductedthisstudyina20-bedEDatanurbanVeteran’sAdministrationhospital.Data wascollectedviain-personpatientandclinicianinterviews.Patientinterviewswereconductedafter patientsreceivedtheirdischargeinstructions.Wecomparedpatientresponsestoclinicianresponses andmarkedthemasincorrect,partiallycorrect,orcorrectwithascoreof0,0.5,or1,respectively.The maximumpossiblescoreforeachinterviewwassix.Sixkeycomponentsofdischargeinstructionswere askedabout:diagnosis;newmedications;at-homecare;durationofillness;reasonstoreturn;and follow-up.Therewere25patientsinthepre-interventiongroupand20intheinterventiongroupwiththe standardizedsetofinstructions.WeperformedaMann-WhitneyUtestonthetotalinterviewscoresinthe controlandinterventiongroupsandconductedasub-analysisontheindividualscoresforeachofthesix keycomponents.

Results: Thepatientsintheinterventiondemonstratedastatisticallysigniﬁcantincreaseinpatientcliniciancorrelationwhencomparedtothepatientsinthepre-interventiongroupoverall(P < 0.05),and twoofthesixkeycomponentsofthedischargeinstructionsindividuallyshowedstatisticallysigniﬁcant increaseinpatient-cliniciancorrelationwhenstandardizeddischargeinstructionswereused.

Conclusion: Patientswhoreceivedthestandardizeddischargeinstructionshadimproved understandingoftheirdischargeinstructions.Futureopportunitiesextendingoffthispilotstudyinclude expandingthenumberofdiagnosesforwhichstandardizedinstructionsareusedandinvestigating patient-centeredoutcomesrelatedtotheseinstructions.[WestJEmergMed.2024;25(6)917–920.]

INTRODUCTION

Severalstudieshaveanalyzedtheeffectivenessof dischargeinstructionsgiventoemergencydepartment(ED) patientsatthetimeofdischargeandhaveidentiﬁedareasfor improvement.1 Thesestudiesrecommendthatkey componentsofdischargeinstructionsincludediagnosis, expecteddurationofillness,at-homecare,return

precautions,andfollow-upplan.Nonetheless,manyED patientsdonotreceivedischargeinstructionsthatincludeall thesecomponents.2,3 Inadditiontobeingincomplete, dischargeinstructionsareoftendifﬁculttoread.4,5 Infact, dischargeinformationgiventotraumapatientsatone institutionwaswrittenatleastfourgradelevelshigheron averagethantheNationalInstitutesofHealth-recommended

sixthgradereadinglevel.Theynotedthatafterimproving readabilitybybreakingupcomplexsentences,usingsimple words,andusingbulletpointsandsubheadings,therewasa signiﬁcantdecreaseinpost-dischargereturnphonecallsand readmissions.5 Additionally,havingagoodunderstandingof one ’sdischargeinstructionscanhelppromoteoptimalhealth andrecoveryfollowinganEDvisit.Patientsmayalsohave fewerunnecessaryreturnvisitstotheEDiftheybetter understandtheirdischargeinstructions.6

Currently,dischargeinstructionsatthisurbanVeteran’ s Administration(VA)hospitalincludeasectionatthe beginningoftheinstructionswhereclinicianscanfreetext anyspeciﬁcinstructionstheyhaveforthepatient.This sectionmayalsobekeptblank.Thereisalsostandardized informationaboutthedischargediagnosis,whichisincluded inallinstructions.Inthispilotstudyweaimedtodetermine whetherimplementingdischargeinstructionsthatare standardizedatanappropriatereadingleveland includekeycomponentswouldimprovepatient understandingofdischargeinstructions(measuredby patient-cliniciancorrelation).

METHODS

Weconductedthispilotstudyata20-bedEDurbanVA hospital.Thisstudydidnotcollectanypersonalpatientdata andthuswasdeemedbytheVAinternalreviewboardofﬁce tobeinstitutionalreviewboard-exempt.Studyparticipants wereapproachedbynursingstaff,clinicians,orstudystaff andaskedwhethertheywouldbewillingtoparticipateina shortinterviewtohelpaqualityimprovementproject focusedondischargeinstructions.Ifthepatientagreed,they wereinterviewedbystudystaffregardingthekey componentsofdischargeinstructions.Theywereaskedto statetheirdiagnosis,what(ifany)newmedicationswere prescribed,whattheyneededtodoathometotakecareof theirillness,expecteddurationofillness,reasonstoreturnto theED,andwhotofollowupwith.Studystaffrecordedtheir answers.Patientswerepermittedtolookattheirdischarge instructionsatanytimeduringtheinterviewtohelpanswer thequestionsandwereremindedofthisatthestartofthe interview.Studystaffthenaskedtheclinician(physicianor advancedpracticepractioner[APP])thesamequestions.

Fortheinitialcontrolgroup,clinicianswerefreeto includewhatevertheywantedinthefree-textportionofthe dischargeinstructions.Thisgroupof25patientshadthe followingdischargediagnoses:edema;motorvehicle collision;concussion;strain;acutepsychosis;constipation; fracture;shingles;hyperglycemia;cysticacne;cervical radiculopathy;oralmucosallesions;conjunctivitis; sinusitis;pneumonia;earinfection;cellulitis;fatigue; diarrhea;chestpain;backpain;balanoposthitis;chronic obstructivepulmonarydisease(COPD);anddehydration. Theclinicianstreatingthisgroupincluded10physiciansand twoAPPs.Datawasagaincollectedbystudystaff(Russell)

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

PatientunderstandingofEDdischarge instructionsisimportantforpatientcare, outcomes,andexperience.

Whatwastheresearchquestion?

Doesimplementingstandardizeddischarge instructiontemplatesimprovepatient understandingattimeofdischarge?

Whatwasthemajor ﬁndingofthestudy?

Theinterventiongroupdemonstrateda statisticallysigni ﬁ cantincrease inunderstandingoftheir instructions(P < 0.05).

Howdoesthisimprovepopulationhealth?

GoodunderstandingofEDdischarge instructionsisvitaltopatienthealthand empowerspatientsbyallowingthemtobetter understandtheirdiseaseanditscourse.

intheformofin-personinterviewsandaddressedthesix keycomponents.

Asetofstandardizeddischargeinstructionswere developedfor12commonEDdiagnosesandeditedto containsixkeycomponents.Thesetemplateswerecreated withsubheadingsandbulletpointstomaketheinstructions easiertofollowandunderstand(AppendixA).Thedischarge diagnosesaddressedinthisgroupincludedmanyofthemost commonemergencydepartmentdiagnoses:abdominalpain; backpain;cellulitis;chestpain;congestiveheartfailure; COPD;concussion;fracture;headache;nofracture(sprain/ strain);ribfracture;andvertigo.Thesedischargeinstruction templateswerereviewedforaccuracyandcompletenessby threeboard-certi ﬁedemergencyphysicians,includingone studystaff,onedirectorofEDoperations,andone educationaldirector.

Aconveniencesampleofemergencyclinicians,including bothboard-certiﬁedphysiciansandphysicianassistants, voluntarilyparticipatedinthepost-standardized interventionphase.Volunteerclinicianshadthestandardized dischargeinstructionsuploadedintotheirdictationsoftware Dragon(NuanceCommunications,Inc,Burlington,MA) andusedthesestandardizedinstructionswhenstudystaff wasonsitetoconductinterviews.Thestudystaffthen collecteddataviain-personinterviewsfortheseclinicians andforthe20patientsforwhomthestandardizeddischarge instructionswereused.

Inbothgroups,patientresponseswerecomparedtotheir ownclinician’sresponsesandmarkedandcodedasincorrect (0),partiallycorrect(0.5),orcorrect(1)withamaximum totalscoreofsix.Resultswerescoredbyeachmemberofthe studyteamindependentlyaswellasbyathird,boardcertiﬁedemergencyphysicianswhowasthedirectorofED operations.WeperformedaMann-WhitneyUtestonthe totalinterviewscoresinthecontrolandinterventiongroups andconductedasub-analysisontheindividualscoresfor eachofthesixkeycomponents.

RESULTS

Demographics:Thetreatmentcliniciansforthepatientsin thebaselinegroupincluded10physiciansandtwoAPPs.The treatmentcliniciansinthepost-standardizedintervention groupincludedthreephysiciansandtwoAPPs.Notethat someclinicianswereinvolvedinbothgroups.

Patientsinthepre-standardizationgroupalreadyshowed highlevelsofunderstandinginthreeareas(above.75):their diagnosis;newmedications;andwhotofollowupwith.The patientsinthepost-standardizedgroupoveralldemonstrated astatisticallysigniﬁcantincreaseinpatient-clinician concordancewhencomparedtothepatientsinthebaseline group(P < 0.05)(Figure),andtwoofthethreelow understandingareas durationofillnessandreasonsto return hadstatisticallysigniﬁcantincreasesinpatientclinicianconcordanceinthebaselinevspoststandardizedgroup.

DISCUSSION

Thedatafromthispilotstudysuggeststhatimplementing dischargeinstructionsstandardizedtoincreasereadability andincludekeycomponentsimprovedpatient understandingcomparedtodischargeinstructionsenteredin viafreetextbytheclinician.Likeotherstudies,ourstudy demonstratedthatreasonstoreturnwereamongthemost poorlyunderstood.7 AsseenintheFigure,thereisclear improvementinthisareawiththeimplementationof

standardizedinstructions.Thisisessentialtopatientcarein theED.Transitionsofcarehavebeenidentiﬁedascritically importanttimesfortransferofinformation.8 Thisis especiallytruewhenpatientsaretransitioningfromhospitalbasedcareintheEDtohome.Indeed,patientunderstanding ofdischargeinstructionshasbeenshowntoimprovehealth outcomesincludingminimizingreturnvisits,increasing follow-up,andenablingimprovedat-homecompliancewith theirclinician’splanofcare.6

Further,institutionssuchastheCentersforMedicareand MedicaidServiceshaveidentiﬁedpatientunderstandingof dischargeinstructionsasakeydomainofpatientexperience, andpatientsareaskedhowwelltheywereabletounderstand thedischargeinstructionsprovidedduringtheirEDvisiton theEDConsumerAssessmentofHealthcareProvidersand SystemsSurvey.Onerecentstudyimplementedamnemonic “DCHOME” (dischargediagnosis,carerendered,health andlifestylemodiﬁcations,obstaclesafterdischarge, prescribedmedications,andexpectations)andformalized educationregardingitsimplementationamongresident physicians,whichdemonstratedsuccessinbothinclusionof thesecomponentsandpatientsatisfaction.9 Thisintervention includedseveralofthecomponentsweincludedinour standardizedwritteninstructions.

Havingagoodunderstandingofone’sdischarge instructionsisimportantformanyreasons,includingthat patientscanhaveoptimalhealthandrecoveryfollowingtheir EDvisit.Betterunderstandingofdischargeinstructionscan alsodecreaseunnecessaryreturnvisitstotheEDby empoweringpatientswiththeinformationtheyneedtomake appropriatefollow-upappointmentsandtobetter understandtheexpectedcourseoftheirillness,whichmay decreasetheunnecessarycostofanadditionalEDvisitfor thepatient.

LIMITATIONS

Onelimitationofthisstudyisthatinter-raterreliability wasnotassessedwithinthedatacollectionandstatistical analysis.Wedidnotcollectthisdataand,therefore,itis unclearhowcloselythedoctors’ ratingscorrelatedtoone another.Futureanalysisandinterventionswouldbenefit fromtwodoctorsratingtheunderstandingandthen performingkappastatisticstomeasurethelevelofagreement betweenthetwodoctors.Anadditionallimitationofthis studyisitssmallsamplesizes.Weusedsmallsamplesizesas thiswasapilotstudywiththegoalofassessingsignificant impactaswellasfeasibilityofimplementation.Asthispilot demonstratesstatisticalsignificanceandclearbeneficial impacttopatientunderstanding,wenowhaveafoundation forfutureexpansionandadditionalresearchwithinthisarea.

Basedonthispilotstudywerecognizeseveralfuture opportunities.Whilethisstudywasfocusedonstandardizing 12commondischargediagnoses,afutureworkcouldexpand thenumberofdiagnosesaswellasthenumberofclinicians.

Thereisanopportunitytoexaminepatient-centered outcomesincludingfollowingpatientsafterdischargeto assessknowledgeretention,returnEDvisits,andadherence withrecommendedfollow-up.Thispilotstudydemonstrates a ﬁrststepinbetterunderstandingthesepatientcentered outcomespotentiallyimpactedbydischargeinstructions. Further,nursingstaffweretheprimaryindividuals distributingthewrittendischargeinstructionstothepatients andexplainingthemone ﬁnaltimepriortodischarge.There iscurrentlywidelyvariablepracticeonhownursingstaff provideanddiscusstheseinstructionswiththepatients.This studydidnotaddressthisvariabilityasourgoalwasto evaluatehowchangingthesinglevariableofthewritten dischargeinstructionswouldaffectpatientunderstanding. Futureworkmayincludestandardizinghowcliniciansor nursingstaffprovidedischargeinstructionsasthishas alsobeenshowntoimpactpatientunderstanding andsatisfaction.9

CONCLUSION

Overall,thiscrucialpilotstudysuggeststhatstandardized dischargeinstructionssigniﬁcantlyimprovepatients’ understandingoftheirinstructionsoveralland,speciﬁcally, theexpecteddurationofillnessandreasonstoreturn.This interventioniseasytoimplement,costeffective,empowers patientstobetterunderstandtheirhealthcondition,impacts coreEDqualitymeasures,andshouldbefurtherstudied.

aspotentialsourcesofbias.Noauthorhasprofessionalor ﬁnancial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconﬂictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.CareyA,StarkweatherA,BaiA,etal.Beneciuk,emergencydepartment dischargeteachinginterventions:ascopingreview, JEmergNurs. 2024;50(3):444–62.

2.NavanandanN,SchmidtSK,CabreraN,etal.Thecaregiverperspective onunscheduled72-hourreturnvisitstopediatricacutecaresites:a focusondischargeprocesses. AcadPediatr. 2017;17(7):755–61.

3.SheikhH,BrezarA,DzwonekA,etal.Patientunderstandingof dischargeinstructionsintheemergencydepartment:dodifferent patientsneeddifferentapproaches? IntJEmergMed. 2018;11(1):5.

4.AlbertiTLandNanniniA.Patientcomprehensionofdischarge instructionsfromtheemergencydepartment:aliteraturereview. JAANP. 2012;25(4):186–94.

5.ChoudhryAJ,YounisM,Ray-ZackMD,etal.Enhancedreadabilityof dischargesummariesdecreasesprovidertelephonecallsandpatient readmissionsintheposthospitalsetting. Surgery. 2018;165(4):789–94.

6.DeSaiC,JanowiakK,SecheliB,etal.Empoweringpatients:simplifying dischargeinstructions. BMJOpenQuality. 2021;10(3):e001419.

7.EngelKG,BuckleyBA,ForthVE,etal.Patientunderstandingof emergencydepartmentdischargeinstructions:whereareknowledge deﬁcitsgreatest? AcadEmergMed. 2012;19(9):E1035–44.

AddressforCorrespondence:SarahRussell,MD,MedicalCollege ofWisconsin,DepartmentofEmergencyMedicine,Hubfor CollaborativeMedicine,8701WatertownPlankRd.,Milwaukee, WI53226.Email: srussell@mcw.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived

8.FriesenMA,WhiteSV,ByersJF.(2008).Handoffs:implicationsfor nurses.In:HughesRG(Ed.), PatientSafetyandQuality:AnEvidenceBasedHandbookforNurses.Rockville,MD:AgencyforHealthcare ResearchandQuality,2008.

9.DalleyMT,BacaMJ,RazaC,etal.Doesastandardizeddischarge communicationtoolimproveresidentperformanceandoverallpatient satisfaction? WestJEmergMed. 2020;22(1):52–9.

ORIGINAL RESEARCH

WhyDoPatientsOptfortheEmergencyDepartmentoverOther CareChoices?AMulti-HospitalAnalysis

CharlesW.Stube,MS*

AlexanderS.Ljungberg,DO*

JasonA.Borton,MD*

KunalChadha,MD†

KyleJ.Kelleran,PhD*

E.BrookeLerner,PhD*†°

SectionEditor:DavidThompson,MD

*UniversityatBuffalo,DepartmentofEmergencyMedicine,Buffalo,NewYork † UniversityatBuffalo,DepartmentofPediatrics,Buffalo,NewYork ° PosthumousAuthorship

Submissionhistory:SubmittedDecember21,2023;RevisionreceivedJuly5,2024;AcceptedAugust1,2024

ElectronicallypublishedSeptember25,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18647

Introduction: Thereareseveraloptionsforreceivingacutecarebesidesemergencydepartments(ED), suchasprimarycarephysician(PCP)ofﬁces,urgentcarecenters(UCC),andtelehealthservices.Itis unknownwhetherthesealternativemodesofcarehavedecreasedthenumberofEDvisitsforpatientsor whethertheyareconsideredbeforevisitingtheED.Acomprehensivestudyconsideringallpotential methodsofcareisneededtoaddresstheevolvinglandscapeofhealthcare.Ourgoalwastoidentifyany factorsorbarriersthatmayhaveinﬂuencedapatient’schoicetovisittheEDasopposedtoaUCC,PCP, anotherlocalED,orusetelehealthservices.

Methods: WesurveyedEDpatientsbetweenthreehospitalsitesinthegreaterBuffalo,NY,area.The surveyconsistedofquestionsregardingthepatients’ reasonsandrationaleforchoosingtheEDoverthe alternativecareoptions.Thestudyalsoinvolvedahealthrecordreviewofthepatients’ diagnoses,tests/ procedures,consults,and ﬁnaldispositionaftercompletionofthesurvey.

Results: Ofthe590patientsconsentedandsurveyed,152(25.7%)consideredseekingcareataUCC, 18(3.1%)consideredtelehealthservices,and146(24.7%)attemptedtocontacttheirPCP.Onthe recommendationoftheirPCP,patientspresentedtotheED110(20.7%)timesandonthe recommendationoftheclinicianattheUCC54(9.2%)times.Patients’ perceivedseriousnessoftheir conditionwasthemostcommonreasonfortheirselectedmodeoftransporttotheEDandreasonfor choosingtheEDasopposedtoalternativecaresites(PCP,UCC,telehealth).Basedoncriteriaforan avoidableEDvisit,83(14.1%)EDpatientsmetthesecriteria.

Conclusion: Individualsprioritizetheperceivedseverityoftheirconditionwhendecidingwheretoseek emergencycare.Whilesomeconsideredalternatives(PCP,UCC,telehealthservices),uncertainties abouttheirconditionandrecommendationsfromothercliniciansledmanytooptforEDcare.Our ﬁndings suggestapotentialgapinunderstandingtheseverityofsymptomsanddeterminingthemostsuitableplace toseekmedicalcarefortheseparticularconditions.[WestJEmergMed.2024;25(6)921–928.]

Keywords: emergencydepartment;urgentcarecenter;primarycarephysician;telehealth; hospitalutilization.

INTRODUCTION

Emergencydepartments(ED)havebecomeahavenfor patientsseekingurgentmedicalattention.Asrequiredby federallaw,EDscannotrefuseevaluationandemergency treatment,regardlessofthepatient’sabilitytopay.1 A1996 studyrevealedthat11.3%ofambulancetransportswere consideredunnecessary,highlightingapositivecorrelation betweenthesevisitstotheEDandlimitedtransportation options.2 Giventheincreasedavailabilityofalternative transportationchoicestodaycomparedwith1996,including publictransportationandride-sharingservices,this correlationmayhaveshifted.Thecurrentmedicallandscape alsoprovidesvariousalternativesformanagingemergent medicalconditions,includingseeingprimarycarephysicians (PCP),visitingurgentcarecenters(UCC),andusing telehealthservices.

Theriseofalternativemedicalcareoptionsraises questionsabouttheirimpactonreducingEDvisitsfor conditionstreatablethroughPCPs,UCCs,ortelehealth services.Recentstudieshaveexaminedwhypatientsoptfor theEDoverothermedicaltreatmentfacilities,citingfactors suchaslimitedaccesstoorconﬁdenceinprimarycare, perceivedurgency,convenience,recommendationsfrom otherphysicians,friends,orfamily,andthebeliefthattheir conditionnecessitatedresourcesprovidedbyhospital-based emergencycare. 3–11

Despitethe ﬁndingthat13.7–27.1%ofallEDvisitscould beevaluatedandtreatedatUCCsorretailclinicswithlower cost,patientsstillfrequentlychooseEDsfornonemergent care.7,12 Anothercommonoccurrenceamongpatients visitingtheEDwithloweracuityconditionsisunnecessary referralfromaPCPorUCC.13,14 Onestudyfoundthatthere weresigniﬁcantlymoreavoidablereferralsfromPCPs (13.9%)thanUCCs(7.9%).13 Ziteketalfound that35.9%ofthepatientsenrolledintheirstudywho transferredfromaUCCtotheEDwereconsideredan unnecessarytransfer.14

Limitedresearchhasexaminedpatientdecision-making whenchoosingbetweentheEDandUCC.A2018study highlightedpatients’ uncertaintyaboutwhatconstituted urgentcare,focusingonpsychologicalfactorsratherthan societalorphysicaldeterminants.15 Mukameletal(2019) addressedthesefactors,emphasizingout-of-pocketcostsand waittimesforseveralmedicalconditionsandcarechoices.8 Their ﬁndingsrevealedthatlowerout-of-pocketcostswere prioritizedoverwaittimeforconditionslowerinseverityor acuity,whereaswaittimegainedimportanceforconditions perceivedtobemoreurgent.8

Thecurrentstudyprovidesacomprehensiveassessmentof EDpatients’ choicesforcareandexploresthefactorsand obstaclesimpactingpatients’ choicesofaparticularEDover aPCP,UCC,telehealthservice,oranothernearbyED. Additionally,wewantedtoassesshowavoidable someoftheseEDvisitscouldbebyexaminingpatients’

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Traditionally,patientsoptfortheemergency department(ED)overothermedicaloptions duetomanyfactorsincludingaccess limitations,perceivedurgency,convenience, andrecommendationsfromothers.

Whatwastheresearchquestion?

Giventheirincreasedavailability,doED patientsconsiderusingalternativecare optionspriortoreportingtotheED?

Whatwasthemajor ﬁndingofthestudy?

AmongEDpatients,14.1%mettheavoidable visitcriteria,providinganopportunityto improveresourceallocation.

Howdoesthisimprovepopulationhealth?

Asasafetynetformedicineandsociety,EDs canbecomeoverburdened.Alternativecare optionsfornon-emergentcasesmayhelp alleviatetheloadonEDs,tofocusonthe sickestpatients.

perceptionsofPCPs’,UCCs’,andtelehealthservices’ abilitiestocareforthemedicalconditionsthatcaused themtoseekcareintheEDandtounderstandthe selectionpatternswithinthisgroup.Understandingthe location,size,clientele,andspecializedcareoftheEDsmay provideinsightastowhyindividualsoptforone medicalcareoptionoveranother.Athoroughexamination ofpatients’ choicescouldoffervaluableinsightsinto enhancingtheavailabilityandaccessibilityof variousmedicalcareoptionsforindividualswith urgentconditions.

METHODS StudyDesign

Thisstudyconsistedofamulti-hospitalsurveyand electronichealthchartreview.Researchassociates(RA) administeredthesurveytopatientsseekingcareatthree separatehospitalEDs.Surveyswereadministeredduring normalbusinesshourswhenmostothercareoptionswould beopenandavailable.Participantsprovidedwrittenconsent attheirbedsidepriortocompletingthesurvey.Thesurvey includedgeneraldemographicquestionsandseveral questionsregardingtheirdecisiontoseekcareinthespeciﬁc EDinwhichtheywereapproached,asopposedtousinga PCP,UCC,telehealthservice,oranotherlocalED.We

Table1. Hospitalsiteinformationandstatistics.17–20

HospitalsHospitaltypeBedsLocationSpecializationEDpatientsperyear

Site1County573UrbanFullservice,regionalLevelItraumacenter70,000

Site2Notforproﬁt484UrbanFullservice,regionalstrokeandSTEMIcenter64,000

Site3Notforproﬁt265SuburbanFullservice50,000 ED,emergencydepartment; STEMI,ST-ElevationMyocardialInfarction.

conductedasubsequenthealthchartreviewfordiagnoses, tests/procedures,consults,and ﬁnaldispositionforeach participatingpatient.Thesurveywasdevelopedfroma previousstudyconductedlocally,2 adaptedto ﬁtcurrent standardsandmedicalcareoptions,anditwasreviewedbya groupoflocalemergencyphysicians.13,16 Thisstudywas approvedbytheinstitutionalreviewboardattheUniversity atBuffalo.

Setting

Thesurveywasconductedatthreelocalhospitalsites locatedinasinglecounty.Twoareinthecenterofa metropolitanareaandoneinacitysuburb(Table1).Sites2 and3arepartofthesamehospitalsystem.TheEDsatall threesitesarestaffedbythesamephysiciangroup.The populationofthelocalcountyisabout950,000asof2021 andincludesthecitywithapopulationofabout277,000.

DataCollection

EachRAwastrainedbythestudycoordinatoratSites1 and2ontheproperproceduresforreviewingtheconsent form,administeringthesurvey,andcollectingthe ﬁnaldata outcomes.AllenrollmentsatSite3weredonesolelybythe studycoordinator.Enrollmentforthisstudybeganon January3,2023,andconcludedonMay1,2023.Data collectiontookplaceatallthreesitesbetween10 AM – 10 PM, Sunday-Saturday.Subjectswereincludediftheywereatleast 18yearsold,readandspokeEnglish,andhadthecapacityto provideconsenttoparticipate.TheRAsateachhospitalthen consultedwiththepatient’sclinicianstodeterminewhether thepatientwasabletogiveconsenttoparticipateinthe survey.TheRAsdidnotapproachpatientsiftheywere altered,toosicktoparticipate,mentallyincapable,nonEnglishspeaking,sleeping,potentiallyinfectious,receiving care, >89yearsold,orreportedbystaffasbeingtooagitated orupsettoparticipate.Additionally,prisonerswerenot consideredforthisstudyasthelocationoftheircareis arrangedwithouttheirinput.Ifpatientswereunableto beapproachedforthesereasons,theywererecorded asineligible.

Afterwrittenconsentwasreceived,thesurveywas administeredverbally,andeveryanswerwasrecordedonan iPadusingREDCap10.3.3(ResearchElectronicData CaptureVanderbiltUniversity,Nashville,TN)data managementplatformsoftwarehostedatUniversityat

Buffalo.Allquestionsfromthesurveywereaskedtothe patientsasopen-ended,buttheRAswhoaskedthequestion wouldcategorizetheanswerbasedonthesurveyoptions. TheRAsweretrainedonhowtocategorizeeachresponseby thestudycoordinator.Iftheresponsedidnot ﬁtanyofthe providedcategories,itwouldbelabeledas “other,” andthe RAwoulddescribetheansweronREDCapviaablank textbox.

Aftereachparticipatingpatientwasdischarged,admitted, ortransferredfromtheED,RAsrecordedtheirdischarge diagnosis,anytestsandproceduresdone,anyspecialists consulted,andthe finaldisposition.Weusedthisinformation todeterminewhetherthepatient’svisittotheEDcouldbe categorizedasavoidable.WedefinedanEDvisitas avoidableifthepatientdidnothaveahigh-acuitytriage categoryoflevel1(resuscitation)orlevel2(emergent),was notadmittedtothehospitalortransferred,hadnoadvanced imaging,hadnospecialistconsultationwhileintheED,and didnothaveadischargediagnosisofchestpainorsyncope. Wedefinedadvancedimagingasanyimagingotherthana radiograph(eg,computedtomography,magneticresonance imaging,orultrasound).Thesecriteriawerebasedon previousresearchcoupledwithaconsensusfromlocal emergencyphysiciansto fitregionalstandards.2,13,16

Analysis

Weanalyzedthedataobtainedfromthesurveysusing SPSSStatisticalSoftwarev27(IBMCorporation,Armonk, NY)afteritwasexportedfromREDCap.Descriptive statisticswereusedtoanalyzetheresponsestothesurveyand presentthedata.

RESULTS

Acrossthethreehospitals,52,246patientsreportedtothe EDduringthestudyperiod.Ofthe1,665peopleconsidered forparticipation,958(57%)wereapproachedand590 (35.4%)consentedtoparticipateinthesurvey,resultingina 1.1%studysampleofthetotalpatientpopulationduringthe collectionperiod(Table2).Mostparticipantswerefemale (60.6%)andWhite(75.8%)(Table3).

Themostcommonmethodsoftransportationreported werehavingafamilymemberorfrienddrivethem(43.9%), followedbyambulancetransport(28.3%)anddriving themselvestothehospital(19.7%).Ofthosepatientswhohad afamilymemberorfrienddrivethem,mostofthem

Table2. Studysamplerepresentationofemergencydepartment population.EmergencySeverityIndex.

Studypatients

Hospitals

Patients surveyed

TotalEDpatients duringstudyperiod

Average triageESI score ED patients seen Average triageESI score

Site11982.5218,0412.75

Site21972.7518,1222.85

Site31952.8816,0832.88

Total:5902.7152,2462.82

ED,emergencydepartment; ESI,emergencyseverityindex.

describedthereasonasbeingtoosicktodrivethemselves (67.2%).Ofthosepatientswhocameinanambulance, aroundhalfofthemexplainedthattheyfelttheyneeded immediatemedicalattention(48.5%).Additionally,mostof thosewhoarrivedbyambulancestatedtheyeithercalledthe ambulancethemselves(32.9%),orafamilymemberorfriend calledoneforthem(37.1%).

Ofthe590patients,530(89.8%)reportedhavingaPCP. Only146(27.5%)ofthose530attemptedtoreachouttotheir PCP,with127(24.0%)makingcontact.Amongthose127 patientswhosuccessfullycontactedtheirPCP,110(86.6%) statedthattheirphysicianadvisedvisitingtheED(Figure1). Ofthose110patients,75(68.2%)hadatriagecategoryof3 and30(27.3%)hadatriagecategoryof2(Table4).Among the152(25.7%)patientswhoconsideredvisitingaUCC,135 (88.8%)hadusedaUCCinthepast,148(97.4%)wereaware ofaUCCintheirarea,and140(92.1%)saidtheywould considerusingaUCCinthefuture.Whenaskedwhythey chosetheEDoveraUCC,54(35.5%)patientsreportedthey wenttoaUCC ﬁrst,buttheUCCclinicianrecommended theygototheED.Thesecondmostcommonanswerwasthe patientbelievedtheirconditionwastooseriousforaUCC, 33(22.4%)(Figure2).

Only18patients(3.1%)consideredusingtelehealth services.Ofthose18,12(66.7%)soughtcarethrougha telehealthservicevisitinthepast,16(88.9%)wouldconsider usingtelehealthservicesinthefuture,and10(55.6%)stated theybelievedtheyneededtheresourcesofahospital,whichis whytheychosetheEDoveratelehealthservicevisit.

ThemainreasonswhypatientschosetheirrespectiveEDs overotherlocalEDsincludedprioruseofhealthcareservices atthathospital(23.9%),livingneartheED(21.2%),andthe beliefthatthehospitalofferedthespecializedservicesthey needed(18.1%).Therewere190(32.2%)patientswitha triagecategoryof1or2.Atotalof254(43.1%)patientswere admittedtothehospital,325(55.1%)hadadvancedimaging, and150(25.4%)hadaspecialistconsultation.Atdischarge, 125(21.2%)patientshadadiagnosisofchestpainorsyncope.

Table3. Demographicsofconsentedparticipants.

Total(N = 590)

Gender

Male228(38.5%)

Female359(60.6%)

Other3(0.5%)

Prefernottoanswer0(0.0%)

Age

MeanYears51.15

SD17.82

Race

Black113(19.1%)

Asian/PaciﬁcIslander10(1.7%)

White449(75.8%)

NativeAmerican11(1.9%)

Other19(3.2%)

Prefernottoanswer2(0.3)

Hispanic/LatinX

Yes46(7.8%)

No544(91.9%)

Highestlevelofeducation

Nohighschool2(0.3%)

Somehighschool38(6.4%)

Highschoolgraduate153(25.8%)

Somecollege121(20.4%)

Associate’sdegree78(13.2%)

Bachelor’sdegree113(19.1%)

Postgraduatedegree78(13.2%)

Trade/technicaltraining7(1.2%)

Other0(0.0%)

Typeofhealthinsurance

Private374(63.4%)

Medicare112(19.0%)

Medicaid88(14.9%)

Uninsured13(2.2%)

Military3(0.5%)

Other0(0.0%)

Ofthe590patientssurveyed,only83(14.1%)patientsmet ourcriteriaforbeinganavoidablevisit.

DISCUSSION

Patients’ perceivedseriousnessoftheirconditionwasthe mostcommonreasonforseekingcareattheEDinsteadof alternativesourcesofcare(Figures1 and 2).Previousstudies suggestthatmanypeoplechoosetotakeanambulance becausesomeoneelsecalledtheambulanceforthemor

Descriptionofinvolvementofprimarycarephysician(PCP)indecisiontogototheED.

Reasonswhypatientschosetheemergencydepartment(ED)insteadofurgentcarecenter(UCC).

becauseoftheperceivedurgencyoruncertaintyabouttheir medicalconditions.21,22 Thisalignedwithour ﬁndings because,ofthe167patientswhopresentedtotheEDby ambulance,81(48.5%)statedthattheyneededimmediate medicalassistanceand29(17.4%)statedthattheyweretoo sickorintoomuchpaintodrivethemselves.Furthermore,of thepatientswhohadafamilymemberorfrienddrivethemto theED,thereasonofbeingtoosicktodrivethemselveswas farmorecommonthantheotherpossiblereasons(67.2%).

Ofthe590patientssurveyed,384(65.1%)didnotattempt tocontactaPCPpriortogoingtotheEDdespite530(89.8%) reportingthattheyhadaPCP.Previousresearchcites

patientsfavoringtheEDduetoperceivedurgency,limited accesstoPCPs,andtheconvenienceofreadilyavailabletests intheED.6,11,23,24 Anotherstudyfoundtheprimaryreason patientschosetheEDinsteadofaPCPwastheperceptionof speedandconvenience.However,this ﬁndingmaybe contradictorybecauseofprolongedEDwaittimesthatmay occurwithlessacuteconditions.25 SimilartoGorodetzeretal (2020),wefoundmorethandoublethenumberofreferrals fromPCPscomparedtoUCCs(110vs54,respectively).13

MostpatientswhodidnotcalltheirPCP’sofﬁcestatedthat theybelievedtheirconditionwastoourgent,whichis comparabletopreviousstudies.5,11,24,26

Figure1.

Figure2.

Table4. Triagecategorycomparedtorecommendationfromprimarycarephysician(N = 127).

TriageCategory

1:Resuscitation2:Emergent3:Urgent4:Lessurgent5:Nonurgent

DoctorrecommendedgoingtoED(n = 110)2(1.8%)30(27.3%)75(68.2%)3(2.7%)0(0%)

DoctordidNOTrecommendgoingtoED(n = 17)0(0%)5(29.4%)11(64.7%)1(5.9%)0(0%)

PCP,primarycarephysician; ED,emergencydepartment.

Additionally,forthosepatientswhowenttotheED againsttheadviceoftheirPCP,56.3%statedthatthey believedtheirconditionwastooseriousorthattheythought theyneededtheresourcesofahospitalfortheircondition (Figure1),whichalignswithpreviousresearch.11 Notably,of thepatientswhowenttotheEDdespitenotbeingadvisedto dosobytheirPCP,29.4%hadatriageacuityof2and64.7% hadatriageacuityof3(Table4),indicatingthatsome patients’ self-referralmayhavebeenmoreadvantageous thaniftheyhadnotchosentovisittheED.Althoughthis mayidentifyanareaforimprovementforpatients’ PCPs,itis difﬁculttoaccuratelyinterpretsituationsbecausethisstudy didnotrecordwhetherpatientscalledorphysicallyvisited theirPCPs’ ofﬁcesortowhompatientsmayhavespoken tothere.

OfthepatientswhoconsideredvisitingaUCC,mostof themreportedusingaUCCinthepast,knewofaUCCin theirarea,andsaidtheywouldconsiderusingaUCCinthe future.Thisinformationcontraststhe ﬁndingsofPopeetal, suggestingthatpeopleintheUnitedStatesmighthavemore ofageneralawarenessofwhataUCCisandtheservicesthey mayprovide.15 Inthisstudy,patientswhostatedthatthey didnotconsiderUCCswerenotaskedwhy.Addingthis questiontothesurveymayprovideabetterindicationofthe psychological,societal,andphysicaldeterminantsastowhy patientschoosetheEDoverUCCssuchascosts,waittimes, andlackofunderstandingofUCCservices,asprevious literaturesuggests.8,15

Toourknowledge,thisisoneofthe ﬁrststudiesto investigatepatientchoicesbetweenEDandtelehealth services.Althoughithasbeenshownthattelehealthservice usedecreasedEDvolumesduringtheearlydaysofthe COVID-19pandemicin2020,thecurrentstudywas conductedJanuary–May2023anddidnotreceivemany responses(18,3.1%)pertainingtotheconsiderationof telehealthservicesinthispopulation.9 Tenparticipants (55.6%)ofthe18whoconsideredtelehealthstatedtheychose theEDovertelehealthservicesbecauseoftheperceivedneed forhospitalresources,similartothereasoningbehind choosingtheEDoverPCPofﬁcesorUCCs.Furthermore, whenthissurveywasconducted,notallhealthinsurances policiescoveredtelehealthservices,potentiallylimiting theirimpact.

Understandingthesigniﬁcantdecision-makingbehavior andfutureconsiderationsforthesepatientsischallenging,

giventhatonly14.1%ofthepatientsmetourcriteriafor avoidablevisits.Additionally,severalstudieshaveattempted tolabela “non-urgent” EDvisitinthepast,eachwith differentcriteria,samplesize,studydesign,andresults.8,12–14 Thiscomplicatesdistinguishingbetweenthosewhotruly requiredEDcareandthosewhomightnothaveneededit, creatingaconvolutedandambiguousprocess.

LIMITATIONS

Forprivacy,surveyswereconductedafterpatientswere assignedandmovedtoanEDroom.Patientswhowere treatedinnon-privateareassuchasfasttrack,hallwaybed, orwaitingroom-adjacentareaswerenotincluded;thesemay representagroupwithagreaterratioofavoidablevisits. Additionally,RAswereunableinterviewpatientswho presentedtotheEDandleftbeforereceivingtreatment.

TherequirementforRAstoreviewaconsentformand obtainasignaturefromthepatientmayhaveresultedin reluctanceorhesitationforparticipationandsubsequent declinationtoparticipatefrom109patientsformultiple reasons.First,intheirreviewoftheconsentform,RAswere requiredtoexplainthatthestudyteamwouldbeobtaining basicinformationfromthepatient’shealthrecordafterthe patientcompletedthesurvey,whichmayhavebeen perceivedasapotentialbreachofconﬁdentiality.Next, reviewingtheconsentdocumenttookapproximatelyfour minutes,whichcouldhavebeenenoughtimeforthe prospectivesubjectstoloseinterest,potentiallyfeeltooillto participate,orforacliniciantointerveneduringthe enrollmentprocess.Lastly,theregulatoryrequirementof obtainingwrittenconsentmayhavedecreasedthepotential numberofpatientsthatcouldhavebeenenrolledinthisstudy andmayhaveintroducedbiasintoour ﬁndings.

DiscussingandansweringquestionsabouttheirEDvisits maybeanemotionalorsensitivetopicforpatients.Although theRAsweretrainedtoaskthequestionsinanonjudgmentalandwelcomingtone,somepatientsmayhave beendisinclinedtoprovidehonestorcompleteresponses. Additionally,goingtotheEDforsomemaybeconsidereda traumaticexperience,regardlessoftriageacuity, whichmayhavereducedwillingnesstoparticipate.Finally, patientswhoweretoosick,intoxicated,orincapacitated werenotapproachedtoparticipateduetotheir condition.Thesepatientswerepresumedtobean

unavoidableEDvisit;thus,omittingtheirdatamay impactedtheresults.

Futurestudiesshouldfocusonincludingruralhospitals comparedtosuburbanandurbanhospitals.Inclusionofa ruralsettingmaycontributemoredatasurroundingpatients’ useoftelehealthservices.Additionally,includingthose patientsseeninfasttrackorotherloweracuityareas wouldprovidemoreinformationonavoidableEDvisits. Previousstudiesalsofoundthatthereisahighnumberof avoidablepediatricpatientvisitstotheED.14 Incorporating pediatricpatientsandthedecision-makingoftheir accompanyingadult(s)infutureworkcouldalso shedlightonhowpeopledecidewheretogofortheir emergencycare.

CONCLUSION

Perour findings,individualsprimarilyrelyontheir perceptionoftheseverityoftheirconditionwhenmaking decisionsaboutseekingemergencycare.Whileseveral patientscontemplatedalternativeoptionssuchasscheduling avisittoaPCP’soffice,visitingaUCC,oraccessing healthcarethroughtelemedicineservices,theuncertainty surroundingtheirmedicalcondition,recommendationsfrom otherhealthcareprofessionals,andtheperceivedqualityof caresignificantlyinfluencedtheirchoiceindirectingthemto theED.Non-emergentpatientsreporttotheEDformany reasonsincludingadiscrepancyinbothunderstandingthe severityofsymptomsanddeterminingthemostsuitable placetoseekmedicalcare.

AddressforCorrespondence:AlexanderS.Ljungberg,DO, UniversityatBuffalo,DepartmentofEmergencyMedicine, 77GoodellSt.Ste.340,Buffalo,NY14203. Email: ajs36@buffalo.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.AmericanCollegeofEmergencyPhysicians.UnderstandingEMTALA. Availableat: https://www.acep.org/life-as-a-physician/ethics–legal/ emtala/emtala-fact-sheet.AccessedMarch15,2023.

2.BillittierAJ,MoscatiR,JanickeD,etal.Amultisitesurveyoffactors contributingtomedicallyunnecessaryambulancetransports. Acad EmergMed. 1996;3(11):1046–50.

3.CosterJE,TurnerJK,BradburyD,etal.Whydopeoplechoose emergencyandurgentcareservices?Arapidreviewutilizinga systematicliteraturesearchandnarrativesynthesis. AcadEmergMed. 2017;24(9):1137–49.

4.AllenL,CummingsJR,HockenberryJM.Theimpactofurgentcare centersonnonemergentemergencydepartmentvisits. HealthServ Res. 2021;56(4):721–30.

5.AlmulhimN,AlmulhimF,AlGharashA,etal.Preferenceforvisiting emergencydepartmentoverprimaryhealthcarecenteramong populationinsaudiarabia. Cureus. 2021;13(12):1–9.

6.KorczakV,YakubuK,AngellB,etal.Understandingpatientpreferences foremergencycareforlowertriageacuitypresentationsduringgphours: aqualitativestudyinAustralia. BMCHealthServRes. 2022;22(1):1442–52.

7.MontalbanoA,RodeanJ,KangasJ,etal.Urgentcareandemergency departmentvisitsinthepediatricMedicaidpopulation. Pediatrics. 2016;137(4):1–6.

8.MukamelDB,LaddH,AminA,etal.Patients’ preferencesovercaresettings forminorillnessesandinjuries. HealthServRes. 2019;54(4):827–38.

9.RenoEM,LiBH,EutermoserM,etal.Temporalassociationsbetween emergencydepartmentandtelehealthvolumesduringtheCOVID-19 pandemic:atime-seriesanalysisfrom2academicmedicalcenters. Am JEmergMed. 2022;54(1):238–41.

10.TapiaAD,HowardJT,BeboNL,etal.Aretrospectivereviewof emergencydepartmentvisitsthatmaybeappropriateformanagement innon-emergencysettings. MilMed. 2022;187(9–10):1153–9.

11.VogelJA,RisingKL,JonesJ,etal.Reasonspatientschoosethe emergencydepartmentoverprimarycare:aqualitativemetasynthesis. J GenInternMed. 2019;34(11):2610–9.

12.WeinickRM,BurnsRM,MehrotraA.Howmanyemergencydepartment visitscouldbemanagedaturgentcarecentersandretailclinics? Health Aff. 2010;29(9):1630–6.

13.GorodetzerR,AlpertEA,OrrZ,etal.Lessonslearnedfroman evaluationofreferralstotheemergencydepartment. IsrJHealthPolicy Res. 2020;9(18):1–8.

14.ZitekT,TanoneI,RamosA,etal.Mosttransfersfromurgentcare centerstoemergencydepartmentsaredischargedandmanyare unnecessary. JEmergMed. 2018;54(6):882–8.

15.PopeC,McKennaG,TurnbullJ,etal.Navigatingandmakingsenseof urgentandemergencycareprocessesandprovision. HealthExpect. 2019;22(3):435–43.

16.Uscher-PinesL,PinesJ,KellermannA,etal.Decidingtovisitthe emergencydepartmentfornon-urgentconditions:asystematicreview oftheliterature. AmJManagCare. 2013;19(1):47–59.

17.BuffaloGeneralMedicalCenter/GatesVascularInstitute.Emergency departmentcare&treatment.Availableat: https://www.buffalogeneral. org/care/Emergency-Department.AccessedMarch15,2023.

18.BuffaloGeneralMedicalCenter/GatesVascularInstitute.StrokeCare CenteratGatesVascularInstitute|care&treatment.Availableat: https://www.buffalogeneral.org/care/Stroke-Care-Centerat-Gates-Vascular-Institute.AccessedMarch15,2023.

19.ErieCountyMedicalCenter.Emergencydepartment-trauma& emergencycare.Availableat: https://www.ecmc.edu/ health-services-and-doctors/trauma-and-emergency-care/ emergency-department/.AccessedMarch15,2023.

20.MillardFillmoreSuburbanHospital.Emergencydepartmentcare& treatment.Availableat: https://www.millardﬁllmoresuburban.org/care/ Emergency-Department.AccessedMarch15,2023.

21.BookerMJ,PurdyS,BarnesR,etal.Ambulanceusefor ‘primarycare’ problems:anethnographicstudyofseekingandprovidinghelpinauk ambulanceservice. BMJOpen. 2019;9(10):e033037.

22.YarrisLM,MorenoR,SchmidtTA,etal.Reasonswhypatientschoose anambulanceandwillingnesstoconsideralternatives. AcadEmerg Med. 2006;13(4):401–5.

23.O’CathainA,SimpsonR,PhillipsM,etal.Tendencytocallan ambulanceorattendanemergencydepartmentforminorornon-urgent problems:avignette-basedpopulationsurveyinbritain. EmergMedJ. 2022;39(6):436–42.

24.RocovichCandPatelT.Emergencydepartmentvisits:whyadults choosetheemergencyroomoveraprimarycarephysicianvisitduring regularofﬁcehours? WorldJEmergMed. 2012;3(2):91–7.

25.LinD,PatrickJ,LabeauF.Estimatingthewaitingtimeofmulti-priority emergencypatientswithdownstreamblocking. HealthCareManagSci. 2014;17(1):88–99.

26.BakerDW,StevensCD,BrookRH.Determinantsofemergency departmentusebyambulatorypatientsatanurbanpublichospital. AnnEmergMed. 1995;25(3):311–6.

ORIGINAL RESEARCH

EmergencyDepartmentPatientSatisfactionScoresAreLowerfor PatientsWhoArriveDuringtheNightShift

TonyZitek,MD*†

LukeWeber,MD*

TatianaNunez,MD*

LuisPuron,DO*

AdamRoitman,BS†

ClaudiaCorbea,BS†

DanaSherman,DNP*†

MichaelShalaby,MD*†

FraydaKresch,MD*

DavidA.Farcy,MD*†

SectionEditor: LauraWalker,MD

*MountSinaiMedicalCenter,DepartmentofEmergencyMedicine,MiamiBeach,Florida † FloridaInternationalUniversity,HerbertWertheimCollegeofMedicine,Miami,Florida

Submissionhistory:SubmittedMarch11,2024;RevisionreceivedAugust29,2024;AcceptedAugust30,2024

ElectronicallypublishedOctober2,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.20326

Background: Increasingly,patientsatisfactionscoresarebeingusedtoassessemergencyphysicians. Wesoughttodeterminewhetherthepatientsatisfactionscorescollectedbyourhospitalsystemare lowerforpatientswhoaretreatedintheemergencydepartment(ED)onnightshiftsascomparedtothose treatedondayshifts.

Methods: Weperformedacross-sectionalanalysisofpatientsatisfactionscoresfromthreeEDsin Florida.WeobtainedsatisfactiondatafromNRCHealth(thecompanythatprovidesoursurveys)usinga randomsampleof1,000completedsurveysfrompatientstreatedin2022;wealsoperformedmanual chartreviewtoobtainclinicaldata.Thesatisfactionsurveysaskedpatientshowlikelytheywouldbeto recommendthefacility(from0–10).Patientswhoprovidedascoreof9or10wereconsidered “promoters.” Forourprimaryanalysis,wecomparedthepercentageofpromotersforthedayshift encounters(7 AM to7 PM)tothenightshiftencounters(7 PM to7 AM).Wealsoperformedamultivariable logisticregressionanalysisusingseveraldemographicandclinicalvariablestofurtherassessthe associationbetweennightshiftarrivalandsatisfactionscores.

Results: Ofthe1,000surveysanalyzed,66.3%ofpatientsarrivedduringthedayshift,and33.7% arrivedduringthenightshift.Ofthosewhoarrivedduringthedayshift,525(79.2%)werepromoters comparedto228(67.7%)ofthosewhoarrivedduringthenightshift,adifferenceof11.5%(95% conﬁdenceinterval[CI]5.7–17.4%), P < 0.001.Onmultivariableanalysis,nightshiftarrivalwas associatedwithalowerchanceofapatientbeingapromoter,withadjustedoddsratio0.60(95%CI 0.43–0.84), P = 0.003.

Conclusion: PatientswhopresentedtotheEDduringthenightshiftwerelesslikelytobe promotersthanpatientswhoarrivedduringthedayshift.Assessmentsofpatientsatisfaction datashouldaccountfortimeofvisitandotherfacility-relatedandoperationalcharacteristics.

[WestJEmergMed.2024;25(6)929–937.]

INTRODUCTION

Withemergencydepartments(ED)open24/7,most emergencyphysiciansworksomenightshifts. Unfortunately,priordatahasshownthatnightshiftworkis associatedwithincreasedriskofavarietyofmedical conditions1–5 andmotorvehiclecollisionsafterthoseshifts.6 Additionally,emergencyphysiciansworkingatnightmay havetodealwithreducedsupportstaff,tiredpatients,and feweravailableconsultants.Moreover,multiplepriorstudies havedemonstratedthatwhileonnightshift,cognitive performancedeclines.7,8

Despitetheuniquechallengesofnightshifts,emergency physiciansaregenerallyheldtothesamestandardsonnight shiftsastheyareondayshifts,andonewaytheyarenow assessedisbypatientsatisfactionscores.Indeed,patient satisfactionhasbecomeanincreasinglyimportantpartof healthcareinlargepartbecauseoftheincentivesinitiatedby theAffordableCareActin20109;now,bothinstitutionand physicianpaymentaresometimesbasedonpatient satisfactionscores.10

Priorstudieshaveshownthatcertainfactorsincluding shorterEDlengthofstay(LOS),11–13 olderpatientage,14 and goodcommunication15 areassociatedwithbetterED patientsatisfactionscores.Twopriorstudieshave investigatedtherelationshipbetweentreatmentduringnight shiftsandpatientsatisfactionscores.14,16 Onefoundno statisticallysigniﬁcantassociation,14 whileanotherfound thatphysicianswhoworkedfewernightshiftshadhigher patientsatisfactionscores.16 Giventheconﬂictingevidence todateandtheincreasingemphasisonpatientsatisfaction, wefeltthatadditionalstudywaswarrantedtoassessthe relationshipbetweennightshiftworkintheEDand patientsatisfaction.

Ourprimaryobjectiveinthisstudywastodetermine whetherpatientswhoarecaredforduringnightshifts providelowerpatientsatisfactionscoresthanthosecaredfor duringdayshifts,usingthereal-worldsatisfactiondata. Secondarily,wesoughttodeterminewhetherother demographicandclinicalcharacteristicsareassociatedwith EDpatientsatisfactionscores.

METHODS

StudyDesignandSetting

Weperformedacross-sectionalanalysisofEDpatient satisfactionscoresfrompatientswhopresentedtoasingle hospitalsystemintheStateofFloridainthesoutheastern UnitedStatesfromJanuary1–December31,2022. Speciﬁcally,weperformedasecondaryanalysisofa previouslycollecteddatasetofsatisfactionscores,andwe performedachartreviewtosupplementthatdata.We followedtheStrengtheningtheReportingofObservational StudiesinEpidemiology(STROBE)guidelines.Thestudywas approvedbytheMountSinaiMedicalCenterInstitutional ReviewBoard.Thisstudyreceivednoexternalfunding.

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Physiciansarejudgedbasedonpatient satisfactionscores.Priordatafoundthat certainpatientandfacilitycharacteristicsare associatedwithsatisfactionscores.

Whatwastheresearchquestion?

DopatientswhopresenttotheEDatnight providelowersatisfactionscoresthanpatients whopresentduringthedaytime?

Whatwasthemajor ﬁndingofthestudy?

Ofdayshiftpatients,79.2%were “promoters” vs 67.7%ofnightshiftpatients(difference11.5% [95%CI5.7–17.4%]),P < 0.001.

Howdoesthisimprovepopulationhealth?

Thisdatahelpsusbetterinterpretpatient satisfactiondata,whichmayhelpimproveour abilitytoprovidepatient-centeredcare.

Ourhospitalsystemhasatertiarycare,community teachinghospitallocatedinMiamiBeach,Florida,aswellas afreestandingEDlocatedinHialeah,Florida,(freestanding ED#1)andafreestandingEDlocatedinAventura,Florida, (freestandingED#2).Themainhospital’sEDhad56,005 visitsduring2022,whilefreestandingED#1had37,932 visitsandfreestandingED#2had19,635visits.Emergency medicineresidentsworkshiftsonlyatthemainhospital’ s ED.Advancedpracticepractitioners(APP)workshiftsatall threefacilities.Shifttimesareshownin Table1.In2022, threeemergencyphysiciansonlyworkedonlynightshifts, andonephysicianworkedonlydayshifts.Someattending physiciansandAPPsonlyworkedatonefacility;others workedattwoorallthree.

SelectionofParticipants

In2022,NRCHealth(Lincoln,NE)administeredour patientsatisfactionsurveysandtrackedsatisfactiondata. Surveysweresentbybothtextmessageandemailtoall patientswholefttheED.Allpatientswhocompletedthe NRCHealthEDpatientsatisfactionsurveyin2022were eligibleforinclusioninthisstudy.Admittedpatientswerenot sentsurveysandwereexcludedfromanalysis.

Measurements

NRCHealthkeepsadatabasewiththeresponsesfrom satisfactionsurveysanddemographicinformationaboutthe

Table1. Emergencydepartmentstaffshifttimesin2022.

patientswhocompletethesurveys.Inourhospitalsystem, currently,individualphysician-levelpatientsatisfaction scoresaretrackedandassessedusingthesedata,but compensationisnotdependentuponthem.Wegenerateda reportfromNRCHealth’sdataforallpatientswho completedasatisfactionsurveyduring2022andthenuseda randomnumbergeneratortocreateasampleof1,000patient encountersforanalysis.Foreachofthesepatientencounters, twomedicalstudentstransferredpatientresponsesand availabledemographicinformationintoaspreadsheetin MicrosoftExcelv16.79.1(MicrosoftCorp,Redmond,WA). Inparticular,NRCHealthprovideduswiththefollowing dataforeachencounter:thedateofthevisit;thefacility;the nameofthephysicianorAPP;methodofpatientresponse (emailortextmessage);andthepatient’ sage,sex,race, medicalrecordnumber,address,maritalstatus,and preferredlanguage.Thetwomedicalstudentswho abstractedthesedatapointshadnoroleintheabstractionof theotherdatadiscussedbelow.

Next,wecreatedaseparatespreadsheetwithadditional clinicalinformationforeachofthe1.000patientencounters usingourelectronichealthrecordsystem(EHR)(Epic SystemsCorporation,Madison,WI).Sixabstractors(three emergencymedicineresidents,twoemergencyattendings, andonenursepractitioner)performedmanualchartreview todeterminethepatient’sethnicity,modeofarrivaltothe ED,timesofarrivalanddeparture,EmergencySeverity Index(ESI)score,17 disposition,clinicianwhodischargedthe patient(andtheirsupervisingattending,ifapplicable),and whetherornoteachofthefollowingwasperformedduring thepatientencounter:residentparticipation;sign-out;blood test;advancedimaging;in-personconsultantevaluation, consultationbyphone(only);opioidpainmedicine administration;andprescriptionprovided.

Ingeneral,manualchartreviewfollowedthemethods suggestedbyKajietal.18 Theabstractorswhoperformed

manualchartreviewwereblindedfromthesatisfactiondata. Noneoftheabstractorsorinvestigatorshavebeena nocturnist.Theabstractors filledin15columnsinthe spreadsheetwiththedatapointsabove.Theyweretrainedon properdataabstractionbytheprincipalinvestigator(TZ), andtheyfollowedadatadictionarythatexplicitlydefinedthe variablesandexplainedwhereto findthemintheEHR.The datadictionaryisincludedasan appendix,whichprovides detaileddefinitionsofallvariables.Thedefinitionsofafew importantvariablesarealsodefinedhereasfollows: Weconsideredpatientstohavearrivedduringtheday shiftiftheyarrivedintheEDbetween7 AM–7 PM andtothe nightshiftiftheyarrivedbetween7 PM–7 AM.Wechosethese definitionsbecausemanyphysicianandnursingshiftsfollow thesetimeschedulesinoursystem(Table1).Wealsodivided patientsintothetimeofyeartheycametotheEDby standardquarters.

Thetypeofclinician(physicianorAPP)whoevaluated thepatientprimarilywasdeterminedbasedonthenameof theclinicianonthesurveyasperNRCHealthdata.For example,apatientwasconsideredtohavebeenseen primarilybyanAPPiftheAPPwasthepersonlistedonthe satisfactionsurvey.Asmentionedabove,wealsomanually recordedthenameoftheclinician(andtheirsupervising attending)whodischargedthepatientforeachpatient encounter.Forpatientswhowerenotsignedout,the dischargingclinician(orsupervisingattending)wasfully consistentwiththelistednameonthesurveys.However,for patientswhoweresignedout,sometimestheinitialclinician whotreatedthepatientwaslistedonthesurveyand sometimesasubsequentonewas.Sinceadministratorsassess thesatisfactiondatabasedonthenameoftheclinicianonthe surveys,weusedthenameoftheclinicianonthesurveyasthe primarytreatingclinician.

PatientswholefttheEDbeforebeingevaluatedbya physicianorAPPcouldstillbeincludedinthestudyifthey

completedasatisfactionsurveybutwereconsideredtohave notbeenseenbyaphysicianorAPP.Allsixabstractors obtainedthedataforagroupofthesame50patientstoallow foranassessmentoftheinter-raterreliability.Wecalculated thefree-marginalkappaforthetwovariablesthatwe consideredtobethemostdifﬁculttoabstract:sign-out,and in-personevaluationbyconsultant.

Aftercompletingdatacollection,theprincipal investigator(TZ)mergedthespreadsheetswiththe satisfactiondataandtheclinicaldata,andthedatawas analyzedasdescribedbelow.

Outcomes

In2022,ouradministrationconsideredthemost importantquestiononthesatisfactionsurveystobe: “How likelyisitthatyouwouldrecommend[facilityname]toa friendorcolleague?” (from0–10).Apatientwhoprovideda scoreof9or10wasconsideredtobea “promoter”;ascoreof 7or8wasconsidered “passive”;andascoreof0–6was consideredtobea “detractor.” Thepercentageofpromoters minusthepercentageofdetractorsisdeemedthe “net promoterscore,” whichisusedtomeasureoverall satisfactioninhealthcareaswellasinotherbusinesses.19,20 Ourprimaryoutcomewasthepercentageofcompleted patientsatisfactionsurveysthatqualiﬁedaspromoters. Secondarily,wedeterminedthenetpromoterscoreandthe adjustedoddsratiosforbeingapromoterforseveral demographicandclinicalvariables.

Analysis

BasedonapreliminaryanalysisofNRCHealthdata,we anticipatedthattherewouldbeapproximatelytwiceasmany completedsurveysfrompatientswhoarrivedduringtheday shiftvsthenight.Additionally,weknewthatapproximately 75%ofourcompletedsurveysin2022qualiﬁedaspromoters. Basedongestalt,wehypothesizedthatthepercentageof promotersfromthedayshiftwouldbeeightpointshigher thanthenightshift.Totestourhypothesiswithanalphaof 0.05andpower0.8,werequiredresponsesfrom957patients. Weroundedthisupto1,000andchosethatasour samplesize.

Forourprimaryanalysis,wecomparedthepercentageof promotersforpatientencountersinwhichthepatientarrived duringthedayshiftcomparedtonightshift.Wemadethe unadjustedcomparisonourprimaryanalysissincethatis howpatientsatisfactionscoresarebeingusedtoassess emergencyphysiciansinourhospitalsystem.Wealso comparedthedemographicandclinicalcharacteristicsofthe patientencountersforthetwogroups.Wedetermined normalitywiththeShapiro-Wilktest.Fornormal distributions,wecomparedthemeansofgroupsusing t -tests. Fornon-normaldistributions,wecomparedmediansusing themediantest.WeusedtheFisherexacttesttocompare categoricalvariables.

Secondarily,giventhatapatientwhoarrivesduringthe endofadayshiftmightbemostlytreatedbythenightstaff (orviceversa),wealsoanalyzedpatientsbasedonthetimeof EDdeparture.Inotherwords,weconsiderednightshift patientstobethosewhodepartedbetween7 PM – 7 AM. Lastly,tofurtherisolatethenight-timehourswhenpeople aregenerallysleeping,wedividedpatientsbyarrivaltime intothreeeight-hourepochs:6 AM – 2 PM (day),2 PM – 10 PM (swing),and10 PM – 6 AM (night).

Lastly,weperformedamultivariablelogisticregression analysiswith “promoter” (yesorno)asthedependent variable.Basedonpriordata,11–14,16,21–23 weincludedthe followingvariablesinourmodel:EDLOS(continuous);ED site(categorical);elderly(age > 65)(binary);pediatric (binary);race(Whiteornot);ethnicity(HispanicornonHispanic),healthinsurancetype(noinsurance,commercial, orgovernment/other);non-Englishspeaking(binary); EmergencySeverityIndex(ESI)(2or3vs4or5);and advancedimagingperformed(binary).Basedoninvestigator hypothesis,wealsochosetoincludethefollowingas covariates:quarteroftheyear(1,2,3,or4);married(binary); seenbyaresident(binary);seenprimarilybyanAPP (binary);arrivalbyambulance(binary);andbloodtest performed(binary).Wealsohypothesizedthatconsultant evaluationswouldbeassociatedwithbettersatisfaction

113,572ED visits during 2022:

56,005visits to the main hospitalED

19,635visits to Freestanding ED #1

37,932visits to Freestanding ED #2

87,885patient encounters in which patient departed from the hospital.

Excluded 25,687admissions.

1,381 patients not sent surveys from January 1-4.

86,504patient encounters in which patient was sent a satisfaction survey.

80,129surveys not completed.

6,375(7.4%) patientencounters for which satisfaction surveywas completed.

1,000 patient surveys randomly selected. 663 arrived between 7 AM and 7 PM (day shift).

arrived between 7 PM and 7 AM (night shift).

Figure. Flowofpatientencounters.

Table2. Characteristicsofpatientvisitsforpatientswhoarrivedduringthedayshift(7 AM to7 PM)ascomparedtothosewhoarrivedduringthe nightshift(7 PM to7 AM).

Day-shiftarrival (n = 663)

Night-shift arrival(n = 337)

Patientdemographics

Meanage(SD)49.3(22.2)43.0(23.4)6.3(3.2to9.3) <0.001*

Pediatric(< 18years),n(%)60(9.0)46(13.7)4.7(0.3to8.9)0.03*

Elderly(> 65years),n(%)203(30.6)78(23.2)7.4(1.8to13.2)0.01*

Male,n(%)273(41.2)147(43.6)2.4( 4.0to8.9)0.46

RaceWhite,n(%)530(79.9)248(73.6)6.3(0.7to12.0)0.02

RaceBlack,n(%)53(8.0)31(9.2)1.2( 2.5to4.9)0.52

Hispanic,n(%)365(55.1)205(60.8)5.8( 0.7to12.2)0.08

Commercialhealthinsurance,n(%)391(59.0)203(60.2)1.2( 5.2to7.7)0.70

Nohealthinsurance,n(%)53(8.0)48(14.2)6.3(2.0to10.5)0.002*

Married,n(%)267(40.8)99(29.6)11.1(5.0to17.3) <0.001*

Non-Englishspeaking,n(%)228(34.4)114(33.8)0.6( 5.7to6.8)0.86

Fromoutofstate,n(%)41(6.2)23(6.8)0.6( 2.6to3.9)0.70

Timeofyearofvisit

Quarter1,n(%)155(23.4)82(24.3)1.0( 4.7to6.6)0.74

Quarter2,n(%)186(28.1)87(25.8)2.2( 3.6to8.0)0.75

Quarter3,n(%)157(23.7)83(4.6)1.0( 4.7to6.6)0.74 Quarter4,n(%)165(24.9)85(25.2)0.3( 5.4to6.0)0.91

Clinicianandfacilitycharacteristics

Mainhospital388(58.5)204(60.5)2.0( 4.4to8.4)0.54

FreestandingED#1160(24.1)67(19.9)4.3( 1.1to9.6)0.13 FreestandingED#2115(17.4)66(19.6)2.2( 2.9to7.4)0.38

Residentparticipated,n(%)217(32.7)179(53.1)20.4(14.0to26.8) <0.001* Advancedpracticepractitioner,n(%)171(25.8)36(10.7)15.1(10.4to19.8) <0.001* Sign-out,n(%)24(3.6)30(8.9)5.3(1.9to8.6) <0.001*

Clinicalcharacteristics

ESIa 2,n(%)20(3.0)13(3.9)0.9( 1.6to3.3)0.48

ESI3,n(%)401(60.5)229(68.0)7.5(1.3to13.7)0.02*

ESI4,n(%)239(36.1)93(27.6)8.5(2.4to14.5)0.007*

ESI5,n(%)3(0.5)2(0.6)0.1( 0.8to1.1)0.77

Arrivedbyambulance,n(%)24(3.6)22(6.5)2.9(0.0to5.9)0.04*

Bloodtestperformed,n(%)287(43.3)126(37.4)5.9( 0.5to12.3)0.07 Advancedimagingperformed,n(%)228(34.4)90(26.7)7.7(1.7to13.6)0.01*

In-personconsultantevaluation,n(%)20(3.0)8(2.4)0.7( 1.4to2.7)0.56

Phone(only)consultation,n(%)23(3.5)2(0.6)2.9(1.3to4.5)0.006*

Opioidpainmedicinegiven,n(%)72(10.9)36(10.7)0.2( 3.9to4.2)0.93

Givenprescription,n(%)374(56.4)170(50.4)6.0( 0.6to12.5)0.07

Medianlengthofstay(IQR),min194(123–259)184(123–265)10( 7to28)0.28

AMA,eloped,orLBTb,n(%)18(2.7)17(5.0)2.3( 0.3to5.0)0.06

aEmergencySeverityIndex.TherewerenopatientswithanESIof1. bAllotherpatientsweredischargedexceptfortwowhoweretransferredtootherhospitals. *Indicatesastatisticallysigniﬁcantdifferencebetweengroups.

AMA,againstmedicaladvice; CI,conﬁdenceinterval; ED,emergencydepartment; ESI,EmergencySeverityIndex; IQR,interquartilerange; LBT,leftbeforetreatment.

Volume25,No.6:November2024WesternJournal

scores,buttheseoccurredtoorarelyinourdatasettobe includedintheregressionanalysis.WeusedtheHosmerLemeshowstatistictoassessgoodnessof ﬁtofthe regressionmodel.

DatawasaggregatedinExcelandanalyzedinRStudio v2023.03.0(RStudioPBC,Boston,MA).Usingtwo-sided hypothesistests,weconsidered P < 0.05tobe statisticallysigniﬁcant.

MissingData

Inafewcases,raceandethnicitywerenotrecorded.This washandledasfollows:Patientswhoweredocumentedas havingWhiteorCaucasianracewereconsideredtobe “White.” PatientswhoweredocumentedasBlackorAfrican Americanwereconsideredtobe “Black.” Patients documentedasAsian,AmericanIndian,multiracial,other, orforwhomracewasnotdocumentedwereconsidered neitherWhitenorBlack.Similarly,ifapatientwas documentedas “Hispanic,” theirethnicitywasconsideredto be “Hispanic.” Iftheyweredocumentedtobenon-Hispanic oriftheirethnicitywasnotdocumented,theywere considered “non-Hispanic.”

RESULTS

Overall

Asshowninthe Figure,6,375satisfactionsurveyswere completedin2022,andwerandomlyselected1,000for analysis.Ofthese,824patientsrespondedbytextmessageand 176respondedbyemail.Ourdataincludedsurveysevaluating 44differentattendingphysiciansand18APPs.Therewereno missingdatapoints,exceptfor fivepatientsforwhomnorace andethnicitywererecorded.Inter-raterreliabilityforthetwo assessedvariableswasalmostperfectwithfree-marginal kappa0.98(95%confidenceinterval[CI]0.94–1.0)forsignoutand0.98(95%CI0.9–1.0)forin-personevaluationbya consultant.Overall,75.3%ofpatientencountersqualifiedas promoters,andnetpromoterscorewas57.8.

NightvsDayShifts

Intotal,663(66.3%)patientsarrivedduringthedayshift, and337(33.7%)arrivedduringthenightshift. Table2 shows

acomparisonofcharacteristicsofthesetwogroups.Notably, thegroupswerenotbalancedonseveralcharacteristics includingage,race,maritalstatus,insurance,typeof cliniciansinvolved,ESI,andadvancedimagingperformed. Regardingtheprimaryoutcome,525(79.2%)ofthosewho arrivedduringthedayshiftwerepromoterscomparedto228 (67.7%)ofthosewhoarrivedduringthenightshift,an absolutedifferencebetweengroupsof11.5%(95%CI 5.7–17.4%), P < 0.001.Thenetpromoterscorefortheday shiftwas64.9and44.0forthenightshift.Datastratiﬁedby facilityareshownin Table3.

Whenredeﬁningdayshiftbydeparturetime,therewere 492day-shiftpatientsand508night-shiftpatients.Ofthose, 396(80.5%)and357(70.3%)werepromotersforthedayand nightshift,respectively,adifferenceof10.2%(95%CI 4.9–15.6%), P < 0.001.

Whenanalyzingthedatabyeight-hourepochs,307 (80.2%)of383patientswhoarrivedbetween6 AM–2 PM were promoters.Meanwhile,339(75.5%)of449patientswho arrivedbetween2 PM–10 PM werepromoters,and107(63.7%) of168whoarrivedbetween10 PM–6 AM werepromoters. Combiningtheeight-hourdayandswingshiftstogether, 77.6%ofsurveyswerepromoters,whichis13.9% (95%CI6.2–21.8%)higherthantheeight-hournightshift group, P < 0.001.

Nineteencompletedsurveyscamefrompatientsseenby oneofourthreenocturnists.Ofthose,12(63.2%)were promoters.Additionally,theonephysicianwhoonlyworked dayshiftshad24completedsurveys,ofwhich21(87.5%) werepromoters.Excludingthecombined43encountersfrom thatphysicianandthethreenocturnistsmadeitsuchthat 78.8%ofpatientswhoarrivedbetween7 AM–7 PM were promotersand68.2%ofthepatientswhoarrivedbetween7 PM–7 AM werepromoters,adifferenceof10.6%(95%CI 4.6–16.6%), P =< 0.001.

MultivariableRegressionAnalysis

Onmultivariableanalysis,arrivalduringthenightshift hadastatisticallysigniﬁcantassociationwithalower chancethatthepatientwouldbeapromoter,withadjusted oddsratio0.60(95%CI0.43–0.84), P = 0.003.Other

Table3. Thepercentageofcompletedsatisfactionsurveysconsideredpromotersoverallandateachofthethreeemergencydepartments, comparingpatientswhoarrivedondayshiftvsnightshift.

Overall(N = 1,000)525(79.2)228(67.7)11.5(5.7to17.4) <

Mainhospital(n = 592)304(78.4)137(67.2)11.2(3.6to18.8)*0.003*

FreestandingED#1(n = 227)160(85.0)47(70.2)14.8(2.6to27.1)*0.01*

FreestandingED#2(n = 181)85(73.9)44(66.7)7.2( 6.7to21.2)0.30

*Indicatesastatisticallysigniﬁcantdifference. CI,conﬁdenceinterval; ED,emergencydepartment.

Table4. Theadjustedoddsratiosofvariousdemographicandclinicalvariablesandtheirassociationwithbeinga “promoter” (apatientwho giveshighratingstoaphysicianonpatientsatisfactionsurveys).

CharacteristicAdjustedoddsratioforbeingapromoter(95%CI)

Demographics

Elderly(age > 65)(n = 281)2.62(1.72–4.08)*

Pediatric(age < 18)(n = 106)0.80(0.48–1.34)

Whiterace(n = 778)

0.91(0.62–1.31)

Hispanicethnicity(n = 570)0.86(0.59–1.26)

HealthInsurance

Commercial(n = 594)

0.78(0.53–1.14)

Governmentorother Reference

Noinsurance(n = 101)

Married(n = 366)

0.83(0.47–1.46)

0.98(0.70–1.40)

Non-Englishspeaking(n = 341) 1.82(1.18–2.82)*

Timeofyearofvisit

Quarter1(n = 237)

0.93(0.60–1.43)

Quarter2(n = 273) 0.84(0.55–1.28)

Quarter3(n = 240) 1.29(0.82–2.03)

Quarter4(n = 250) Reference

Facility

MainED(n = 592) Reference

FreestandingED#1(n = 227)0.91(0.38–1.08)

FreestandingED#2(n = 181)0.65(0.51–1.60)

Cliniciancharacteristics

Seenbyaresident(n = 397)0.81(0.51–1.27)

APPprimarily(n = 207) 0.78(0.51–1.19)

Sign-out(n = 54) 1.08(0.54–2.28)

Clinicalcharacteristics

Emergencyseverityindex

ESIlevel2or3(n = 663) Reference

ESIlevel4or5(n = 337)0.97(0.67–1.41)

Arrivedbyambulance(n = 46)0.54(0.28–1.09)

Bloodtestperformed(n = 413)1.11(0.76–1.64)

AdvancedImagingperformed(n = 318)1.38(0.92–2.09)

Opioidpainmedicinegiven(n = 108)1.01(0.60–1.73)

Prescriptiongiven(n = 543)1.15(0.84–1.59)

EDlengthofstay(for1-hincrease)0.94(0.84–1.04)

Arrivalduringnightshift(n = 337)0.60(0.43–0.84)*

*Indicatesastatisticallysigniﬁcantassociation.

APP,advancedpracticepractitioner; CI,conﬁdenceinterval; ED,emergencydepartment; ESI,EmergencySeverityIndex.

thannight-shiftarrival,noothervariableswereassociated withareducedchanceofbeingapromoter.Ontheother hand,elderlypatients(age > 65)andnon-Englishspeaking patientshadpositiveassociationswithbeinga promoter(Table4).

DISCUSSION

Inthiscross-sectionalstudy,wefoundthatEDpatients whoarriveordepartduringnightshiftarelesslikelytobe promotersascomparedtoday-shiftpatients.Notably,the patientpopulationthatcompletedthesatisfactionsurveys

andarrivedduringanightshiftwassubstantiallydifferent thanthosepatientswhoarrivedduringdayshift.Considering thisandotherintuitivechallengesofnightshifts,unadjusted comparisonsofphysicianswhoworkdifferentratiosofday andnightshiftsonanynumberofmetricsarelikelytobe compromised.However,inourstudyevenafteradjustingfor severaldifferencesbetweendayandnightshifts,westill foundanassociationbetweennight-shiftarrivalandlower patientsatisfactionscores.

Priordataonthissubjecthasbeenmixed.Onepriorstudy evaluatedtherelationshipbetweennightshiftsandED patientsatisfactionscoresusingdatafrom2009–2013froma singleEDanddidnotshowasigniﬁcantassociation.14 On theotherhand,apriorlargestudyusingdatafrom42 facilitiesfrom2012–2015foundthatphysicianswhoworked fewernightshiftshadhigherpatientsatisfactionscores.16

Bothofthesestudiesattemptedtoassessalargenumberof physician,facility,andoperationalfactorsthatmightaffect patientsatisfactionscores.Arelativestrengthofourstudy wasthatitwasamoretargetedandgranularassessment speciﬁcallyofnight-vsday-shiftpatientsatisfactionscores.

Ourstudywasnotdesignedtospecificallyassessthe associationsofothervariableswithpatientsatisfaction scores,butwewillbrieflyreviewthesecondary findings.In thisregard,ourresultswerelargelyconsistentwithprevious work,includingour findingsthatelderlypatientsandnonEnglishspeakingpatientsaremorelikelytoprovidehigh satisfactionscores.14,22 Priordatahasbeenmixedwith regardtotheassociationofLOSandpatient satisfaction.11–13,16,24,25 Wefailedto findanassociation betweenEDLOSandpatientsatisfaction(Table4),whichis consistentwithpreviousworkthathasreportedthat perceivedLOSismoreimportantthanactualLOS.24,25

Overall,althoughempathyandcommunicationare importantcontributorstopatientsatisfaction15,22,24 thatan emergencycliniciancanmostlycontrol,therearemany factorsthattheycannot.Ourdataandpreviousdemonstrate thatnightshiftwork,thepatientpopulation,14 andthe facility16,23 allinﬂuencepatientsatisfactionscores. ConsideringalsothattheresponserateforEDsatisfaction surveysissolow(<10%inoursystemandsimilarinmany others26)andthatonlydischargedpatientsaresentsurveys, werecommendagainsttheuseofpatientsatisfactionscores todeterminepaymentforemergencyclinicians.

LIMITATIONS

Thisstudyhadseverallimitations.First,ourdatacomes fromasinglehospitalsystemthathasfairlyhighpatient satisfactionscores;so,ourresultsmaynotbeapplicableto otherEDs.Additionally,giventheretrospectiveand observationalnatureofthestudy,therecouldhavebeen someunmeasuredconfoundersthatcouldexplainthe differencesinpatientsatisfactionbetweenthedayandnight shifts.Namely,whilephysiciansusuallyworkbothdayand

nightshifts,nursingstaffandsupportstaffmoretypically workonlydaysoronlynights.Therefore,differencesin staffingmightexplainthedifferencesinsatisfactionscores. Moreover,priorstudieshavedemonstratedthat communicationisanimportantfactorinEDpatient satisfactionscores,22,27 butgiventhedesignofthisstudy,it wasnotpossibletoassessthequalityofcommunication. Next,ourdatadidnothavethegranularitytoadjustfor patientvolumeforeachshift,whichcouldhaveimpacted patientsatisfaction.However,giventhatthemedianED LOSwassimilarintheday-andnight-shiftgroups,wedoubt thatdifferencesinpatientvolumewouldexplainthelower satisfactionscoresbynight-shiftpatients.Lastly,thelow responseratetopatientsatisfactionsurveysisalimitationin thatsurveyresponsesarelikelysubstantiallyinfluencedby selectionbias,butwedonotconsiderthisalimitationspecific toourstudybecauseourgoalwastocomparethereal-world patientsatisfactionscoresfromday-vsnight-shiftpatients (withcurrentsurveytechniques).Ourresultsthusprovidea comparisonofthedatathatisactuallybeingusedtoassess clinicians’ performanceonpatientsatisfaction.

CONCLUSION

Inthiscross-sectionalstudy,night-shiftarrivaltotheED wasassociatedwithastatisticallysignificantlowerchance thatthepatientwouldbeapromoteronsatisfactionsurveys. Giventhis findingandpreviousdatasuggestingthatother issuesbeyondthephysician’scontrolheavilyinfluence satisfactionscores,facilityfactors,patientcharacteristics, andoperationalfactors(includingthetimeoftheED visit)shouldbeconsideredwhenassessingpatient satisfactionscores.

AddressforCorrespondence:TonyZitek,MD,MountSinaiMedical Center,DepartmentofEmergencyMedicine,4300AltonRd.,Miami Beach,FL33140.Email: zitek10@gmail.com

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.VetterC,DashtiHS,LaneJM,etal.Nightshiftwork,geneticrisk,and type2diabetesintheUKbiobank. DiabetesCare. 2018;41(4):762–9.

2.VetterC,DevoreEE,WegrzynLR,etal.Associationbetweenrotating nightshiftworkandriskofcoronaryheartdiseaseamongwomen. JAMA. 2016;315(16):1726–34.

3.MaidstoneRJ,TurnerJ,VetterC,etal.Nightshiftworkisassociated withanincreasedriskofasthma. Thorax. 2021;76(1):53–60.

4.Fagundo-RiveraJ,G´omez-SalgadoJ,García-IglesiasJJ,etal. Relationshipbetweennightshiftsandriskofbreastcanceramong nurses:asystematicreview. Medicina(Kaunas). 2020;56(12):680.

5.BrownDL,FeskanichD,SanchezBN,etal.Rotatingnightshiftworkand theriskofischemicstroke. AmJEpidemiol. 2009;169(11):1370–7.

6.SteeleMT,MaOJ,WatsonWA,etal.Theoccupationalriskofmotor vehiclecollisionsforemergencymedicineresidents. AcadEmergMed. 1999;6(10):1050–3.

7.DulaDJ,DulaNL,HamrickC,etal.Theeffectofworkingserialnight shiftsonthecognitivefunctioningofemergencyphysicians. AnnEmerg Med. 2001;38(2):152–5.

8.Smith-CogginsR,RosekindMR,HurdS.Relationshipofdayversus nightsleeptophysicianperformanceandmood. AnnEmergMed. 1994;24(5):928–34.

9.HealthCare.gov.ReadtheAffordableCareAct.Availableat: https:// www.healthcare.gov/where-can-i-read-the-affordable-care-act/ AccessedAugust16,2024.

10.FinkelsteinJ,LiftonJ,CaponeC.Redesigningphysiciancompensation andimprovingEDperformance. HealthcFinancManage. 2011:65(6).114–7.

11.ParkerBTandMarcoC.Emergencydepartmentlengthofstay: accuracyofpatientestimates. WestJEmergMed. 2014;15(2):170–5.

12.ChangAM,LinA,FuR,etal.Associationsofemergencydepartment lengthofstaywithpubliclyreportedquality-of-caremeasures. Acad EmergMed. 2017;24(2):246–50.

13.WalshMandKnottJC.Satisfactionwiththeemergencydepartment environmentdecreaseswithlengthofstay. EmergMedJ. 2010;27(11):821–8.

14.HandelDA,FrenchLK,NicholJ,etal.Associationsbetweenpatient andemergencydepartmentoperationalcharacteristicsandpatient satisfactionscoresinanadultpopulation. AnnEmergMed. 2014;64(6):604–8.

15.SonisJD,AaronsonEL,LeeRY,etal.Emergencydepartmentpatient experience. JPatientExp. 2018;5(2):101–6.

16.PinesJM,PennintiP,AlfarajS,etal.Measurementunderthe microscope:highvariabilityandlimitedconstructvalidityinemergency departmentpatient-experiencescores. AnnEmergMed. 2018;71(5):545–54.e6.

17.WuerzRC,TraversD,GilboyN,etal.Implementationandreﬁnementof theemergencyseverityindex. AcadEmergMed. 2001;8(2):170–6.

18.KajiA,SchrigerD,GreenS.Lookingthroughtheretrospectoscope: reducingbiasinemergencymedicinechartreviewstudies. AnnEmerg Med. 2014;64(3):292–8.

19.ReichheldFF.Theonenumberyouneedtogrow. HarvBusRev. 2003;81(124):46–54.

20.LiyakasaK.Customerexperienceiscriticalinnetpromoter benchmarks.2012.Availableat: http://www.destinationcrm.com/ Articles/Columns-Departments/Insight/Customer-Experience-IsCritical-in-Net-Promoter-Benchmarks-82569.aspx AccessedNovember29,2023.

21.BoudreauxED,FriedmanJ,ChanskyME,etal.Emergencydepartment patientsatisfaction:examiningtheroleofacuity. AcadEmergMed. 2004;11(2):162–8.

22.SunBC,AdamsJ,OravEJ,etal.Determinantsofpatientsatisfaction andwillingnesstoreturnwithemergencycare. AnnEmergMed. 2000;35(5):426–34.

23.BendeskyBS,HunterK,KirchhoffMA,etal.Samephysician,different location,differentpatientsatisfactionscores. AnnEmergMed. 2016;68(5):531–5.

24.NicholJR,FuR,FrenchK,etal.Associationbetweenpatientand emergencydepartmentoperationalcharacteristicsandpatient satisfactionscoresinapediatricpopulation. PediatrEmergCare. 2016;32(3):139–41.

25.BoudreauxEDandO’HeaL.Patientsatisfactionintheemergency department:areviewoftheliteratureandimplicationsforpractice. JEmergMed. 2004;26(1):13–26.

26.JaquisWp,DesRochersLR,EnguidanosER,etal.Patientsatisfaction. 2011.Availableat: https://www.acep.org/patientsatisfaction/.Accessed September24,2017.

27.TomaG,TrinerW,McNuttLA.Patientsatisfactionasafunctionof emergencydepartmentprevisitexpectations. AnnEmergMed. 2009;54(3):360–7.e6.

ORIGINAL RESEARCH

ComparisonofEmergencyDepartmentDispositionTimesinAdult LevelIandLevelIITraumaCenters

SierraLane,BS*

JeffryNahmias,MD,MHPE*

MichaelLekawa,MD*

JohnChristianFox,MD†

CarrieChandwani,MD†

ShahramLot ﬁpour,MD,MPH† AregGrigorian,MD*

SectionEditor: GaryJohnson,MD

*UniversityofCalifornia,Irvine,DepartmentofSurgery,DivisionofTrauma, BurnsandSurgicalCriticalCare,Orange,California † UniversityofCalifornia,Irvine,DepartmentofEmergencyMedicine, Orange,California

Submissionhistory:SubmittedMarch25,2024;RevisionreceivedSeptember6,2024;AcceptedSeptember10,2024

ElectronicallypublishedOctober22,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.20523

Introduction: Theef ﬁcientutilizationofresourcesisacrucialaspectofhealthcare,particularlyinboth LevelIandLevelIIAmericanCollegeofSurgeons(ACS)-veriﬁedtraumacenters.Theeffectofresource allocationonemergencydepartmentlengthofstay(ED-LOS)oftraumapatientshasremainedunderinvestigated.AsEDcrowdinghasbecomemoreprevalent,especiallyatquaternarycarecenters,an evaluationofthepotentialdisparitiesinED-LOSbetweenLevelIandLevelIItraumacentersis warranted.WehypothesizedalongerED-LOSatLevelIcenterscomparedtoLevelIIcenters.

Methods: Wequeriedthe2017–2021TraumaQualityImprovementProcess(TQIP)databasefor traumapatients ≥18yearsofagepresentingtoeitheraLevel-Ior-IIcenter.TheTQIPdeﬁnesED-LOSas thetimefromarrivaluntilthetimeanEDdisposition(admissionordischarge)orderiswritten.We excludedtransferredpatientsandthosewithmissingdataregardingACStraumacenterveriﬁcation level.Weperformedbivariateanalyses,aswellassubgroupanalysesbasedonlocationofdisposition.

Results: Of2,225,067traumapatients,59.3%(1,318,497)receivedtreatmentatLevelIcenters.No signiﬁcantdifferenceswerefoundinInjurySeverityScoresbetweenpatientsadmittedtotheoperating roomornon-intensivecareunit(ICU)locations,ordischargedhomefromLevel-Iand-IIcenters(all P < 0.05).TheED-LOSfortraumapatientswaslongeratLevel-Icentersforallpatientcategories:overall (198vs145minutes[min], P < 0.001),dischargedhome(286vs160min, P < 0.001),non-ICU admissions(234vs164min, P < 0.001),andthoserequiringsurgery(126vs101min, P < 0.001).

Conclusion: Evenwhentreatingpatientswithsimilarinjuryseverity,traumapatientsatLevelItrauma centershadlongerED-LOScomparedtoLevelIIcenters,irrespectiveofthepatients’ ﬁnaldisposition (surgery,non-ICUadmission,ordischarge).TooptimizeresourceutilizationandalleviateEDsaturation, furtherresearchmustdelveintotheunderlyingcausesofthesediscrepanciestoidentifybestpractices andsolutions.[WestJEmergMed.2024;25(6)938–945.]

INTRODUCTION

Traumacontinuestoposeasigniﬁcantpublichealth challengethatplacessubstantialdemandsonhealthcare systems.Since2010,traumahasconsistentlybeentheleading causeofdeathforyoungadults.1 TheCoalitionforNational

TraumaResearchreportsthattraumaaccountsfor approximately41millionemergencydepartment(ED)visits eachyearaswellastwomillionhospitaladmissions annually.2 Inthiscontext,thelengthofstay(LOS)intheED actsasakeymetric,reﬂectingtheefﬁciencyandeffectiveness

ofpatientcare.ProlongedED-LOSisassociatedwith adverseclinicaloutcomes,includingincreasedriskof hospital-acquiredinfections,delaysintheadministrationof criticalmedications,andincreasedmortality,which highlightstheimportanceofrapidandwell-coordinated emergencycare. 3–10 Existingliteraturehighlightsdisparities inED-LOSacrossvariousmedicalcenters;however,thereis asigniﬁcantlackofdatafocusingontraumacenters.11

TraumacentersaredesignatedbytheAmericanCollegeof Surgeons(ACS)basedonpatientvolume,stafﬁng,resources, injuryprevention,andeducation.12 Thistieredstructurehas enabledashiftfromtraditional,hospital-centricmodelstoa moreintegrated,regionalizedsystemoftraumacare.13 Despiteexistingstudieshighlightingthecomplexitiesof traumacasesandtheimpactofprolongedED-LOS,there remainsasubstantialgapinresearchconcerninghow resourceallocationaffectsED-LOSfortraumapatients, particularlybetweenvariouslevelsoftraumacenters.14–18 Theselevelsmaydifferintermsofresourcesandcapabilities intheED,withLevelItraumacenters(L1TC)typically handlingmorecomplexcasesandhavingmore comprehensiveresourcescomparedtoLevelIItrauma centers(L2TC).

TheimportanceofinvestigatingED-LOSdifferences betweenL1TCsandL2TCshasbecomemorepronouncedin thewakeoftheCOVID-19pandemic.Thepandemic affectedtraumamechanismsandoutcomesincluding exacerbatedEDcrowding,alongstandingissuein healthcare,andposeduniquechallengestotraumacare, particularlyinhigherleveltraumacenters,whichoftenserve asquaternarycarefacilities.19–24 Crowdingleadstodelaysin careandabidirectionalimpactonbothtraumaandnontraumapatients.TheinﬂuxoftraumapatientstotheED reallocatesstaffandresourcesfromotherpatients undergoingsimultaneousevaluationandtreatment, increasingtheirED-LOS.25,26

InthisstudyweaimedtoanalyzealargeUnitedStates traumadatabasetocompareED-LOSbetweenadulttrauma patientsatL1TCsandL2TCs.Wehypothesizedanincreased ED-LOSatL1TCscomparedtoL2TCs.Thisresearchmay helpimprovepatientexperienceandqualityofhealthcareas EDcrowdingcontinuestoimpacthospitalsnationwide.

METHODS

Thisstudywasdeemedexemptfrominstitutionalboard review,andawaiverofconsentwasgrantedforuseofadeidentifiednationaldatabase.Weperformedaretrospective analysisoftheTraumaQualityImprovementProgram (TQIP)databasefrom2017–2021.Patients ≥18yearsofage presentingtoeitheranACS-verifiedL1TCorL2TCwere included.Weexcludedallpatientstransferredfromanother facilityaswellasthosewithmissingdataregardingACS traumacenter-verificationlevel.Ourprimaryfocuswasto accuratelyassessED-LOSfortraumapatients.Including

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

ProlongedEDlengthofstay(LOS)islinked toadverseclinicaloutcomesandhighlights theimportanceofrapidandwell-coordinated emergencycare.

Whatwastheresearchquestion?

HowdoesED-LOSfortraumapatientsdiffer betweenLevelIandLevelIItraumacenters?

Whatwasthemajor ﬁndingofthestudy?

TheED-LOSfortraumapatientsatLevelI centerswaslongeroverall(198vs145 minutes,P < 0.001)comparedto LevelIIcenters.

Howdoesthisimprovepopulationhealth?

Thisresearchenhancesourunderstandingof patientexperienceandqualityofhealthcare byaddressingEDcrowding,alongstanding issuenationwide.

transferpatientswouldhaveintroducedconfoundingfactors thatcouldhavesigniﬁcantlyskewedouranalysis.Trauma transferpatientsmayhavealreadyundergoneextensive evaluationsandimagingattheinitialhospital,whichcan artiﬁciallyshortentheirED-LOSatthereceivinghospital. Additionally,someofthesepatientsmayhavebeenpreacceptedbythetraumateam,resultinginamoreexpedited admissionprocesscomparedtonon-transferpatients. Therefore,includingtransferpatientswouldnotprovidean accuraterepresentationofED-LOSfortraumapatients.We comparedtwogroups:adulttraumapatientstreatedat L1TCsvsL2TCs.Thisincludedacomparisonofallpatients regardlessoflevelofcare.

Wecollectedpatientdemographicvariablesincludingage andprehospitalcomorbiditiessuchasdiabetesmellitus, hypertension,anticoagulanttherapy,mentalorpersonality disorder,smokingstatus,houselessness,andsubstanceuse. TheinjuryproﬁleincludedtheInjurySeverityScore(ISS), andtheAbbreviatedInjuryScale(AIS)ofthehead, abdomen,andthorax.Wealsocollectedvitalsonarrival includinghypotension(systolicbloodpressure ≤ 90 millimetersofmercury),tachycardia(heartrate > 120beats perminute),andtachypnea(respiratoryrate > 22breathsper minute).TheprimaryoutcomemeasuredwasED-LOS. Additionally,wecollectedpatientdispositionfromtheED, includingadmissiontothegeneralhospital ﬂoor,intensive careunit(ICU),operatingroom(OR),ordischargetohome.

Wealsoanalyzedinpatientcomplications,suchasacute kidneyinjury,cardiacarrest,unplannedintubation, ventilator-acquiredpneumonia,anddeepveinthrombosis. Wecontrastedpatientcharacteristics,injuryproﬁles, complications,anddispositionsbetweenadultpatients treatedatL1TCsandL2TCs.

WeperformedbivariateanalysesusingaMann-Whitney Utesttocomparecontinuousvariablesandchi-squareto comparecategoricalvariables.Wereportcategoricaldataas percentagesandcontinuousdataasmedianswith interquartilerange(IQR)orasmeanswithstandard deviation.Amultivariablelogisticregressionanalysiswas alsoperformedtodeterminetheassociatedriskofmortality andcomplications.Eachmodelincludedknownriskfactors formortalityandinhospitalcomplicationsfortrauma patientsincludingage,vitalsonadmission,mechanism,ISS, andthepresenceoftraumaticbraininjury.27–30 These covariatesweredeterminedbyco-authorconsensusand reviewoftheliterature.All P -valuesweretwo-sidedwitha statisticalsigniﬁcancelevelof <0.05.Weperformedall analyseswithSPSSStatisticsforWindowsv29(IBMCorp,

Armonk,NY).TheStrengtheningtheReportingof ObservationalStudiesinEpidemiology(STROBE)checklist wasusedtoensureadherencewithestablishedguidelinesfor reportingobservationalstudies.31

RESULTS

Demographics,Characteristics,andInjuriesofPatientsat Level-IvsII-TraumaCenters

Of2,225,067patients,59.3%(1,318,497)received treatmentataL1TCand40.7%(906,570)ataL2TC. PatientsatL1TCsweregenerallyyounger(median50vs 58years, P < 0.001)thanatL2TCs.TheL1TCandL2TC patientshadasimilarmedianISSof9.However,patientsat L2TCshadhigherratesofthefollowingprehospital comorbidities:anticoagulanttherapy(12.1%vs8.8%, P < 0.001);diabetesmellitus(15.2%vs13.5%, P < 0.001),and hypertension(37.2%vs31.9%, P < 0.001).PatientsatL1TCs weremoreoftenhouseless(1.6%vs1.2%, P < 0.001);more oftenunderwentbloodtransfusionscomparedtoL2TCs (5.9%vs3.6%, P < 0.001),andhadhigherratesofsubstance usedisorder(9.0%vs6.3%, P < 0.001)(Table1).Patients

Table1. Demographics,comorbidities,andvitalsignsofadulttraumapatientstreatedatlevelIvslevelIItraumacenters.

Age,year,median(IQR)50(32,68)58(39,77)

Comorbidities,n(%)

Alcoholusedisorder92,090(7.1%)54,266(6.1%)

Houselessness*4,834(1.6%)2,383(1.2%)

Congestiveheartfailure48,149(3.7%)40,712(4.5%)

Currentsmoker306,022(23.5%)168,447(18.8%) <

Chronicrenalfailure20,982(1.6%)15,992(1.8%) <0.001

Cerebrovascularaccident32,748(2.5%)25,637(2.9%) <0.001

Diabetesmellitus175,825(13.5%)135,779(15.2%) <0.001

Hypertension416,425(31.9%)334,025(37.2%) <0.001

Chronicobstructivepulmonarydisease76,547(5.9%)65,353(7.3%) <0.001

Cirrhosis15,868(1.2%)9,301(1.0%) <0.001

Dementia53,592(4.1%)53,164(5.9%) <0.001

Anticoagulanttherapy114,428(8.8%)108,082(12.1%) <0.001

Anginapectoris1,872(0.1%)2,137(0.2%) <0.001

Myocardialinfarction7,609(0.6%)5,833(0.7%) <0.001

Peripheralarterialdisease10,938(0.8%)9,362(1.0%) <0.001

Substanceusedisorder117,680(9.0%)56,160(6.3%) <0.001

Vitalsonadmission,n(%)

Hypotension(SBP < 90)59,051(4.6%)29,024(3.3%) <0.001

Tachycardia(HR > 120)99,388(7.7%)55,816(6.3%) <0.001

Tachypnea(RR > 22)223,774(17.6%)136,745(15.5%) <

Bloodtransfusion,n(%)78,273(5.9%)32,230(3.6%) <0.001

*Onlyincludes2021data. IQR;interquartilerange; HR,heartrate; RR,respiratoryrate; SBP,systolicbloodpressure.

treatedatL1TCsalsohadincreasedratesofhigh-grade injuries(AIS > 3)tothehead(14.4%vs14.2%, P < 0.002), abdomen(4.8%vs.3.5%, P < 0.001),andthorax(15.8%vs. 13.1%, P < 0.001)(Table2).

ED-LOSofL1TCandL2TC

PatientsatL1TCswereadmittedathigherratestothe ICU(19.3%vs17.6%, P < 0.001)anddirectlytotheOR (13.7%vs10.6%, P < 0.001),whilepatientsatL2TCswere admittedathigherratestothegeneralhospital ﬂoor/ward (57.1%vs55.4%, P < 0.001)anddischargedhome(9.8%vs 7.9%, P < 0.001).TheL1TCpatientshadincreasedmedian ED-LOSforalldispositionswhencomparedtoL2TC patients:overall(198vs145minutes[min], P < 0.001); dischargedhome(286vs160min P < 0.001);non-ICU admissions(234vs164min, P < 0.001),ICUadmissions(123 vs108min, P < 0.001),anddirecttransporttotheOR(126vs 101min, P < 0.001)(Table3).

OtherMeasuredOutcomesofLevelI-and II-TraumaCenters

WhencomparedwithL2TCs,theoccurrenceofan inhospitalcomplicationwashigheratL1TCs(5.8%vs4.4%, P < 0.001).Thisincludedincreasedratesofunplanned intubation(1.1%vs0.8%, P < 0.001),ventilator-acquired pneumonia(0.5%vs0.3%, P < 0.001),anddeepvein thrombosis(0.7%vs0.5%, P < 0.001)atL1TCs.Increased ratesofunplannedICUadmissions(1.6%vs1.3%, P < 0.001)andunplannedreturnstotheOR(0.7%vs0.5%, P < 0.001)alsooccurredmorecommonlyatL1TCs(Table4).

Afteradjustingforconfounders,L1TCpatientscontinuedto exhibitahigherassociatedriskofcomplications(oddsratio [OR]1.22,conﬁdenceinterval[CI]1.20–1.24, P < 0.001). ComparedwithL2TCpatients,L1TCpatientsexhibiteda higherrateofmortality(4.8%vs3.8%, P < 0.001) (Table4).However,thistrenddidnotpersistwhen controllingforknownriskfactorsofmortality(OR0.99, CI0.97–1.01, P = 0.09).

DISCUSSION

Thiscomprehensive ﬁve-yearretrospectivenational analysisrevealedthatdespitecomparableinjuryburdens, patientstreatedatL1TCsexperiencedalongerassociated ED-LOSacrossalldispositioncategories,alongwitha higherrateandassociatedriskofcomplications,compared tothoseatL2TCs.Interestingly,theassociatedriskof mortalityremainedsimilarbetweenthetwolevelsoftrauma centerdesignations.

Emergencydepartmentcrowdingremainsaprominent issue,representingapervasivechallengeassociatedwith delayedtreatment,reducedpatientsatisfaction,and increasedmortality.32,33 Thissituationoccurswhenthe demandforemergencycaresurpassestheavailableresources intheED,hospital,orboth.34 Despitevariationsin deﬁnitionsandmeasurementsofcrowdingamonghospitals, itsrepercussionswillalmostalwaysresultinalongerEDLOS.35 Whiteetal’sstudy,focusingondischargedpatients, corroboratesthiscorrelationbyrevealinga10%increasein ED-LOSforpatientsseenduringperiodsofEDcrowding.36 Crowdingoftenleadstoabottleneckeffectinpatient ﬂow,

Characteristic,n(%)LevelI(n

Bluntmechanism,n(%)1,107,121(84.0%)810,732(89.4%)

AISgrade > 3,n(%)

Head189,248(14.4%)128,778(14.2%)

Abdomen63,011(4.8%)31,659(3.5%)

Thorax208,098(15.8%)118,421(13.1%)

Injury,n(%)

Brain221,032(16.8%)134,573(14.8%) <0.001

Liver43,616(3.3%)18,070(2.0%) <0.001

Smallintestine17,861(1.4%)6,904(0.8%) <0.001

Colon16,162(1.2%)5,979(0.7%) <0.001

Rectum2,072(0.2%)675(0.1%) <0.001

Kidney20,043(1.5%)9,370(1.0%) <0.001

Spleen36,477(2.8%)17,827(2.0%) <0.001

Pancreas4,756(0.4%)1,687(0.2%) <0.001

Stomach5,138(0.4%)1,723(0.2%) <0.001

AIS,AbbreviatedInjuryScale; IQR,interquartilerange; ISS,InjurySeverityScale.

Table2. InjuriesforadulttraumapatientstreatedatlevelIvslevelIItraumacenter.

Table3. DispositionofadulttraumapatientstreatedatalevelIvslevelIItraumacenter.

DispositionfromED,n(%)

Admitto ﬂoor731,039(55.4%)517,613(57.1%) <0.001

AdmittoICU254,892(19.3%)159,987(17.6%) <0.001

DirecttoOR180,479(13.7%)95,952(10.6%) <0.001

Dischargedhome103,779(7.9%)88,399(9.8%) <0.001

EDLOS,minutes,median(IQR)

Allpatients198(233)145(138) <0.001

Admitto ﬂoor234(230)164(140) <0.001

AdmittoICU123(163)108(108) <0.001

DirecttoOR126(196)101(117) <0.001

Dischargedhome286(283)160(139) <0.001

ED,emergencydepartment; ICU,intensivecareunit; IQR,interquartilerange; LOS,lengthofstay; OR,operatingroom.

wherepatientsawaitingadmissionoccupyEDbeds,thus limitingtheavailabilityfornewpatients.Thisscenariois furtherexacerbatedduringpeaktimesorpublichealthcrises, liketheCOVID-19pandemic,whereaninﬂuxofpatientscan overwhelmEDresources.Prolongedwaittimescanleadto patientdiscomfortanddissatisfaction.37 Pinesetalobserved thatextendedED-LOSwaslinkedtoadiminished probabilityofpatientsrecommendingthehospitaltoothers, coupledwithalowerperceptionofeffectiveteamwork amonghospitalstaff.38

Thedynamicsofresourceutilizationattraumacenters requiresfurtherinvestigationtouncovertheunderlying reasonsfortheobservedprolongedED-LOSatL1TCs.For instance,thesecentersarewidelyacknowledgedfor managinghighpatientvolumesandcateringtomore complexcasesthatmighthinderthe ﬂowofpatientsthrough theED.39 Thisissupportedbyourstudydemonstratingthat L1TCsmoreoftentreatpatientswithsevereinjuriestothe head,abdomen,andthorax,comparedtoL2TCs.Thismay necessitatemorecomprehensivediagnosticevaluations, specializedinterventions,and/orcoordinationamong varioussurgicalspecialties,allofwhichcontributetolonger LOSintheED.Additionally,L1TCsoftenhostresidency programsandfrequentlyinvolveresidentsandhousestaffin patientcare,afeaturelesscommonlyfoundatL2TCs.40 The involvementoftraineesmaycontributetoalongerED-LOS, asresidentsandhousestaffmayneedtoconsultwith attendingphysiciansandsupervisorstodiscusstreatment plans,whichmayleadtoincreaseddeliberationanddecisionmakingtime.41–43 Understandingtheimpactofeducational programsonED-LOSisnecessaryforoptimizingresource allocationandenhancingefﬁciencyoftraumacaredelivery withindifferentlevelsofresourcecenters.

IncreasedED-LOSmayresultinworsenedclinical outcomes.Wedidnot ﬁndahigherriskofmortalityfor patientstreatedatL1TCs;however,wedid ﬁndahigher associatedriskofinhospitalcomplications.Thispattern suggestssuboptimalutilizationoravailabilityofimportant resources,potentiallyleadingtothedecompensationof patients.Insupportofthishypothesis,wefoundthatpatients atL1TCshadhigherratesofunplannedintubation,ICU admission,andreturntotheOR.WhiletheTQIPdatabaseis notgranularenoughtodeterminewhetherthese complicationsweretheresultofincreasedED-LOS,itdoes highlighttheneedforenhancedmanagementstrategiesto ensurethatpatientsreceivetimelyandeffectivecare, particularlyinhigh-acuitysettingswherethemarginforerror isminimal.

EfﬁcientlyaddressingthechallengeofED-LOSinvolvesa multifacetedapproach,integratingbothstrategiccapacity managementandinnovativepatientcarepractices.Key strategiesincludeoptimizinginpatientbeduse,expanding EDcapacitythroughadditionalbedsorstafﬁng,andearly

PatientsatL1TCswerealsomorefrequentlyimpactedby socialdeterminantsofhealthincludinghouselessnessand substanceusedisorder.OurstudyrevealedthatL1TCsmore oftencaredforhouselesspatientsandthosesufferingfrom substanceusedisorder.Unhousedpatientstendtoexperience longerED-LOSsincedispositionplanningandarrangements provetobemorecomplicatedforpatientslackingstable housingwhiletheirmedicalneedsarebeingaddressed.44–46 Moreover,houselesspatientsfacemarkedlyhigheroddsof hospitaladmissioncomparedtotheirhousedcounterparts,a disparitylikelyinﬂuencedbyclinicians’ concernsoverthe risksandsafetyofdischargingindividualsbacktothe streets.47,48 Thepronouncedpresenceofsocialdeterminants ofhealthamongpatientsatL1TCshighlightsthecomplex interplaybetweenhealthcaredeliveryandsocietalissues, emphasizingtheneedforfurtherinvestigationinto healthcaredisparities.

Table4. OutcomesforadulttraumapatientstreatedatlevelIvslevelIItraumacenters.

Characteristic,n(%)LevelI(n

Anycomplication76,217(5.8%)39,600(4.4%) <

Cardiacarrest12,623(1.0%)6,782(0.7%) <

Catheter-associatedUTI2,329(0.2%)1,073(0.1%) <0.001

DeepSSI1,863(0.1%)646(0.1%) <0.001

OrganspaceSSI1,432(0.1%)392(<0.1%) <0.001

SuperﬁcialSSI1,360(0.1%)635(0.1%) <0.001

Deepveinthrombosis8,802(0.7%)4,647(0.5%) <0.001

Pulmonaryembolism4,890(0.4%)2,016(0.2%) <0.001

Unplannedintubation13,927(1.1%)7,400(0.8%) <0.001

Acutekidneyinjury7,296(0.6%)3,971(0.4%) <0.001

Pressureulcer6,036(0.5%)2,948(0.3%) <0.001

Acuterespiratorydistresssyndrome3,884(0.3%)1,822(0.2%) <0.001

UnplannedreturntoOR8,784(0.7%)4,126(0.5%) <0.001

Sepsis4,107(0.3%)1,968(0.2%) <0.001

Stroke3,581(0.3%)1,931(0.2%) <0.001

UnplannedICUadmission21,417(1.6%)11,540(1.3%) <0.001

Ventilator-associatedPNA7,027(0.5%)2,654(0.3%) <0.001

Mortalityrate,n(%)63,347(4.8%)34,067(3.8%) <0.001

ICU,intensivecareunit; OR,operatingroom; PNA,pneumonia; SSI,surgicalsiteinfection; UTI,urinarytractinfection.

physicianassessmentstoexpeditedecision-making,thereby reducingEDcrowdingandprolongedED-LOS.49 Proven interventionssuchasthefast-trackprocess,whichnotably reducedED-LOSforlow-acuitypatientsby25%,andrevised triageapproachesinL1TCshavedemonstratedsuccessin expeditingcareandreducingED-LOS.50 Anotherexample ofeffectivetriage-systemredesigninvolvesestablishing specializedunitsspeciﬁcallyforlessseverecases,alongwith theinclusionofadvancedpracticepractitioners.This approachhassuccessfullyledtoareductioninED-LOSby morethan30minutes.51

LIMITATIONS

Thisstudyislimitedbypotentialreportingandselection biases,codingerrors,andmissingdatainherentindatabase studies.Wedidnotconsiderawidearrayofexternalfactors thatcouldinfluenceED-LOS,suchas fluctuationsinED volumepercenter,disproportionateincreasesincenters approved,andvariationsinhospitalandEDoccupancy. Additionally,ourstudycontainsconstraintsinidentifying specifictreatmentlocationswithinthehospital,whetherthat beadedicatedtraumaareaorthegeneralED.Afurther limitationisourinabilitytocontrolforcompetingLOS factors.Specifically,wewereunabletoaccountforfactors includingresidentstaffing,consultmanagement,andthe differingpracticepatternsforemergentandnon-emergent carebetweentheEDandotherhospitalsettings.

Geographicaldifferencesbetweentraumacenterswere notconsidered,whichmightimpactED-LOSdueto variationsinregionalhealthcarepolicies,patient demographics,andresourceavailability.TheTQIPdatabase doesnotprovidegranulardetailsonspecificinterventions anddecision-makingprocessesintheED,whichcouldaffect LOS.Furthermore,wedidnotincludepatient socioeconomicfactorsintheanalysis,whichcouldhave impactedED-LOSandpatientoutcomes.Finally,aswithall databasestudies,wecannotconcludeanydefinitivecausality statementregardingtraumacenterlevelandED-LOS. Despitetheselimitations,our findingscontribute significantlytothediscourseonED-LOS,layinga foundationforfutureresearchaimedatoptimizingresource allocationandimprovingtraumacaredeliveryinL1TCs andL2TCs.

CONCLUSION

Thiscomprehensiveanalysishighlightsasigniﬁcant observeddisparityinEDlengthofstaybetweenLevelIand LevelIItraumacenters.LevelItraumacentersconsistently reportedlongerassociatedED-LOSacrossvariouspatient dispositions,aswellasahigherriskofcomplications,despite treatingsimilarlyinjuredpatients.Factorsleadingtothese ﬁndingscouldrangefromoperationalprotocolsandresource managementtopatientcasecomplexityandinstitutional policies.DuetolimitationsoftheTraumaQuality

ImprovementProgramdatabase,wewereunabletoattribute theobserveddifferencesinED-LOStoanysinglefactor,as theassociationsobservedinourstudyarebasedondatafrom alargenationaldatabase,whichenhancesthe generalizabilityofour findingsacrossdiversesettings.This broadscopereducestheinfluenceofregionalpoliciesand allowsourresultstobeapplicableonawiderscale.However, addressingtheseunderlyingcausesisessentialnotonly forenhancingtheefficiencyofpatient flowthroughthe hospitalbutalsoforimprovingtheoverallqualityofcare providedtotraumapatients.Toeffectivelytacklethisissue, furtherprospectiveresearchisneededtodelveintothe specificsofwhythesediscrepanciesexist.Thisincludes examininghospitaloperationalstrategies,patient flow processes,staffingmodels,andtheuseoftechnologyin patientmanagement.

AddressforCorrespondence:AregGrigorian,MD,Universityof California,Irvine,DepartmentofSurgery,DivisionofTrauma,Burns andSurgicalCriticalCare,333TheCityBlvd.West,Suite1600, Orange,CA.Email: agrigori@uci.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.TheAmericanAssociationfortheSurgeryofTrauma.Traumafactsand links.2020.Availableat: https://www.aast.org/resources/trauma-facts AccessedFebruary24,2024.

2.HickersonB.MissionoftheCoalitionforNationalTraumaResearch. Availableat: https://www.nattrauma.org/who-we-are/mission/ AccessedDecember26,2023.

3.LauqueD,KhalemskyA,BoudiZ,etal.Length-of-stayintheemergency departmentandin-hospitalmortality:asystematicreviewandmetaanalysis. JClinMed. 2022;12(1):32.

4.MoweryNT,DoughertyS,HildrethA,etal.Emergencydepartment lengthofstayisanindependentpredictorofhospitalmortalityintrauma activationpatients. JTraumaAcuteCareSurg. 2011;70(6):1317–25.

5.PinesJM,LocalioAR,HollanderJE,etal.Theimpactofemergency departmentcrowdingmeasuresontimetoantibioticsforpatientswith community-acquiredpneumonia. AnnEmergMed. 2007;50(5):510–6.

6.CarrBG,KayeAJ,WiebeDJ,etal.Emergencydepartmentlengthof stay:amajorriskfactorforpneumoniainintubatedblunttrauma patients. JTraumaAcuteCareSurg. 2007;63(1):9–12.

7.SchullMJ,VermeulenM,SlaughterG,etal.Emergencydepartment crowdingandthrombolysisdelaysinacutemyocardialinfarction. Ann EmergMed. 2004;44(6):577–85.

8.SinghN,RobinsonRD,DuaneTM,etal.RoleofEDcrowdingrelativeto traumaqualitycareinalevel1traumacenter. AmJEmergMed. 2019;37(4):579–84.

9.ChalﬁnDB,TrzeciakS,LikourezosA,etal.Impactofdelayedtransferof criticallyillpatientsfromtheemergencydepartmenttotheintensivecare unit. CritCareMed. 2007;35(6):1477–83.

10.SingerAJ,ThodeHC,ViccelioP,etal.Theassociationbetweenlength ofemergencydepartmentboardingandmortality. AcadEmergMed. 2011;18(12):1324–9.

11.ReznekM,MichaelS,HarbertsonC,etal.Clinicaloperationsof academicversusnon-academicemergencydepartments:adescriptive comparisonoftwolargeemergencydepartmentoperationssurveys. BMCEmergMed. 2019;19(1):72.

12.AmericanCollegeofSurgeons.Resourcesforoptimalcareofthe injured.2022.Availableat: https://www.facs.org/media/1qumyf4b/ 2022_vrc_injured-patient-standardsmanual_ﬁnal.pdf AccessedDecember26,2023.

13.GlanceLG,OslerTM,MukamelD,etal.Impactoftraumacenter designationonoutcomes:isthereadifferencebetweenlevelIand levelIItraumacenters? JAmCollSurg. 2012;215(3):372–8.

14.KrasnoffCC,GrigorianA,ChinT,etal.Pediatricburn-traumapatients haveincreasedlengthofstaycomparedtotrauma-onlypatients:a propensitymatchedanalysis. Burns. 2021;47(1):78–83.

15.DuanD,DengH,ChenY,etal.Associatedpredictorsofprolonged lengthofstayinpatientssurvivingextensiveburns:alargemulticenter retrospectivestudy. Burns. 2024;50(2):413–23.

16.MejaddamAY,ElmerJ,SiderisAC,etal.Prolongedemergency departmentlengthofstayisnotassociatedwithworseoutcomesin traumaticbraininjury. JEmergMed. 2013;45(3):384–91.

17.ElmerJ,PallinDJ,LiuS,etal.Prolongedemergencydepartmentlength ofstayisnotassociatedwithworseneurologicoutcomesinpatientswith intracerebralhemorrhage. NeurocritCare. 2012;17(3):334–42.

18.BurgessL,Ray-BarruelG,KynochK.Associationbetweenemergency departmentlengthofstayandpatientoutcomes:asystematicreview. ResNursHealth. 2022;45(1):59–93.

19.LuceroA,SokolK,HyunJ,etal.Worseningofemergencydepartment lengthofstayduringtheCOVID-19pandemic. JAmCollEmerg PhysiciansOpen. 2021;2(3):e12489.

20.ClarkDEandRyanLM.Concurrentpredictionofhospitalmortality andlengthofstayfromriskfactorsonadmission. HealthServRes. 2002;37(3):631–45.

21.YeatesEO,GrigorianA,BarriosC,etal.Changesintraumatic mechanismsofinjuryinSouthernCaliforniarelatedtoCOVID-19: penetratingtraumaasasecondpandemic. JTraumaAcuteCareSurg. 2021;90(4):714–21.

22.YeatesEO,GrigorianA,SchellenbergM,etal.COVID-19intrauma:a propensity-matchedanalysisofCOVIDandnon-COVIDtrauma patients. EurJTraumaEmergSurg. 2021;47(5):1335–42.

23.YeatesEO,GrigorianA,SchellenbergM,etal.Decreasedhospital lengthofstayandintensivecareunitadmissionsfornon-COVID blunttraumapatientsduringtheCOVID-19pandemic. AmJSurg. 2022;224(1):90–5.

24.OskvarekJJ,ZocchiM,BlackBS,etal.Emergencydepartmentvolume, severity,andcrowdingsincetheonsetofthecoronavirusdisease2019 pandemic. AnnEmergMed. 2023;82(6):650–60.

25.SmithDC,ChapitalA,UperesaBMB,etal.Traumaactivationsandtheir effectsonnon-traumapatients. JEmergMed. 2011;41(1):90–4.

26.AryaR,DossantosF,Ohman-StricklandP,etal.Impactoftrauma activationontheEDlengthofstayfornontraumaticpatients. AmJEmergMed. 2012;30(2):311–6.

27.AlbitarO,BallouzeR,OoiJP,etal.Riskfactorsformortalityamong COVID-19patients. DiabetesResClinPract. 2020;166:108293.

28.BattleC,CarterK,NeweyL,etal.Riskfactorsthatpredictmortalityin patientswithbluntchestwalltrauma:anupdatedsystematicreviewand meta-analysis. EmergMedJ. 2023;40(5):369–78.

29.ClementND,HallAJ,MakaramNS,etal.IMPACT-restart:theinﬂuence ofCOVID-19onpostoperativemortalityandriskfactorsassociatedwith SARS-CoV-2infectionafterorthopaedicandtraumasurgery. BoneJoint J. 2020;102(12):1774–81.

30.YadavK,LampronJ,NadjR,etal.Predictorsofmortalityamongolder majortraumapatients. CJEM. 2023;25(11):865–72.

31.vonElmE,AltmanDG,EggerM,etal.Thestrengtheningthe reportingofobservationalstudiesinepidemiology(STROBE) statement:guidelinesforreportingobservationalstudies. BMJ. 2007;335(7624):806–8.

32.ChangAM,LinA,FuR,etal.Associationsofemergencydepartment lengthofstaywithpubliclyreportedquality-of-caremeasures. AcadEmergMed. 2017;24(2):246–50.

33.JohnsonKDandWinkelmanC.Theeffectofemergencydepartment crowdingonpatientoutcomes:aliteraturereview. AdvEmergNursJ. 2011;33(1):39–54.

34.AmericanCollegeofEmergencyPhysicians.Crowdingpolicy statements.2019.Availableat: https://www.acep.org/siteassets/newpdfs/policy-statements/crowding.pdf.AccessedDecember31,2023.

35.McCarthyML,DingR,PinesJM,etal.Comparisonofmethodsfor measuringcrowdinganditseffectsonlengthofstayintheemergency department. AcadEmergMed. 2011;18(12):1269–77.

36.WhiteBA,BiddingerPD,ChangY,etal.Boardinginpatientsinthe emergencydepartmentincreasesdischargedpatientlengthofstay. JEmergMed. 2013;44(1):230–5.

37.LizanaFG,BotaDP,DeCubberM,etal.Long-termoutcomein ICUpatients:whataboutqualityoflife? IntensiveCareMed. 2003;29(8):1286–93.

38.PinesJM,IyerS,DisbotM,etal.Effectofemergencydepartment crowdingonpatientsatisfactionforadmittedpatients. AcadEmergMed. 2008;15(9):825–31.

39.TrzeciakSandRiversEP.Emergencydepartmentovercrowdinginthe UnitedStates:anemergingthreattopatientsafetyandpublichealth. EmergMedJ. 2003;20(5):402–5.

40.MadayagRM,SercyE,BergGM,etal.Aretraumaresearchprogramsin academicandnon-academiccentersmeasuredbyequalstandards?A surveyof137levelItraumacentersintheUnitedStates. PatientSaf Surg. 2021;15(1):34.

41.DeLaneyM,ZimmermanKD,StroutTD,etal.Theeffectofmedical studentsandresidentsonmeasuresofefﬁciencyandtimelinessinan academicmedicalcenteremergencydepartment. AcadMed. 2013;88(11):1723–31.

42.JamesC,HarperM,JohnstonP,etal.Effectoftraineesonlengthof stayinthepediatricemergencydepartment. AcadEmergMed. 2009;16(9):859–65.

43.LammersRL,RoigerM,RiceL,etal.Theeffectofanewemergency medicineresidencyprogramonpatientlengthofstayina communityhospitalemergencydepartment. AcadEmergMed. 2003;10(7):725–30.

44.KramerCB,GibranNS,HeimbachDM,etal.Assaultandsubstance abusecharacterizeburninjuriesinhomelesspatients. JBurnCareRes. 2008;29(3):461–7.

45.Kowal-VernAandLatenserBA.Demographicsofthehomelessinan urbanburnunit. JBurnCareRes. 2007;28(1):105–10.

46.FerradaP,AnandRJ,AboutanosM.Theuninsured,thehomeless,and theundocumentedimmigranttraumapatient.Revealinghealth-care disparityatalevel1traumacenter. AmSurg. 2016;82(1):E1–2.

47.SilverCM,ThomasAC,ReddyS,etal.Injurypatternsandhospital admissionaftertraumaamongpeopleexperiencinghomelessness. JAMANetwOpen. 2023;6(6):e2320862.

48.TrinhT,ElferganiA,BannM.Qualitativeanalysisofdispositiondecision makingforpatientsreferredforadmissionfromtheemergency departmentwithoutdeﬁnitemedicalacuity. BMJOpen. 2021;11(7):e046598.

49.MorleyC,UnwinM,PetersonGM,etal.Emergencydepartment crowding:asystematicreviewofcauses,consequencesandsolutions. PLoSOne. 2018;13(8):e0203316.

50.MercerMP,SinghMK,KanzariaHK.Reducingemergencydepartment lengthofstay. JAMA. 2019;321(14):1402–3.

51.GardnerRM,FriedmanNA,CarlsonM,etal.Impactofrevisedtriageto improvethroughputinanEDwithlimitedtraditionalfasttrackpopulation. AmJEmergMed. 2018;36(1):124–7.

IyesattaM.Emeli,MD,MPH PatrickG.Meloy,MD

SectionEditor:CortlynBrown,MD

COMMENTARY

WeighingIn

EmoryUniversitySchoolofMedicine,DepartmentofEmergencyMedicine, Atlanta,Georgia

Submissionhistory:SubmittedJanuary10,2024;RevisionreceivedMay13,2024;AcceptedAugust15,2024

ElectronicallypublishedOctober4,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18690

“Whatisyourpatient’sbiggestproblem?”

Thissimplequestionwasposedtwodecadesagobyan attendingphysiciantoamedicalstudentrotatinginafamily medicineclinic.Thepatienthadthecommontrioofdiabetes, hypertension,andhyperlipidemia,andthestudentwent throughthislistofconditions.Thestudent’sattending, however,remainedunconvinced. “Obesity,” theattending ﬁnallysaid. “Obesityisthispatient’sbiggestproblem.” The patientwasoverweight,yes,butinthatcomfortablewaythat weoverlookinourfamilymembers,inourfriends,andin ourselves.Thatwasthe ﬁrsttimeinthatmedicalstudent’ s trainingthatanyclinicianhadputweightonparwith,and indeedabove,otherestablishedchronicmedicalconditions. Itwasalsothelasttime.

Thisvignettehighlightsourlackofroutine acknowledgmentofobesityasadisease.

Obesityisadisease,1,2 justasdiabetes,hypertension,and hyperlipidemiaarediseases.Obesityisalsoapublichealth crisis,bothlocallyandglobally.3 IntheUnitedStates,upto oneinthreeadultsandoneinsixchildrenareaffected,and theincidenceandprevalencehasbeenincreasingsteadily overthepast20years.3 Theincreaseinmorbidityand mortalityduetoobesitycannotbeunderstated:obeseadults havelowerlifeexpectancy,lowerqualityoflife,and substantialincreasesinhealthcarecosts.3 Additionally, obesityincreasestheriskofbeingdiagnosedwithother lifelongdiseasessuchasdiabetes,hypertension,coronary arterydisease,myocardialinfarction,andstroke.4 In addition,thecombinationofobesityandothercomorbid conditions,eg,smoking,canincreasetheriskofdeathbyup to11timesthatofanon-obese,non-smoker.5 Itisthrough theselensesthatpublichealthofﬁcialshaverecognized obesityasanalarmingdiseaseprocessthatthreatensthe livesofbothyoungandoldandisincreasinglyprevalentin UShealthcare.

Asemergencyphysicians,wedonot,forthemostpart, havelong-termrelationshipswithourpatients.Weexpect primarycarephysiciansandthosewhoprovideongoingcare tomakeittheirbusinesstocounselonweightloss. Unfortunately,thedatasaysotherwise.AccordingtotheUS CentersforDiseaseControlandPrevention(CDC),evenin

physicianvisitsmadespeciﬁcallyforobesity,weight-related healtheducationwasonlyofferedhalfofthetime.6

Emergencydepartments(ED)arethesafetynetsoftheUS healthcaresystem,caringforacute,unscheduledpatients regardlessofabilitytopay.WhilemanyEDvisitsare subsequentlydeemedtobenon-emergent,theystillrequirea timelyevaluationandinvolvesome,albeitnotas comprehensive,preventativecareinterventions.Wetellour patientstoquitsmoking,practicesafesex,takeseizure precautions,weartheirseatbeltsandbicyclehelmets,andget theirCOVID-19shots.Chartsarelitteredwith “dotphrases” suchasthefollowing: “Patientiscounseledfortobacco cessation” ; “WilladmitforTIArule-outandinitiatestatin therapy” ; “Patientistreatedempiricallyforurethritisandis counseledonsafesexualpractices” ;or “Patientisgiven informationonlow-saltdietforcongestiveheartfailure.” However,adviceonproperdietandexerciseisuncommon;in ourclinicalexperience,wetendtoaddressobesityonlywhen itismorbidandwhenitdisruptsourintendedcareplan;for example,whenthepatientistoooverweightforastretcher,a scanner,oraprocedure.Shouldwebetalkingaboutweight intheED?Iftheanswerisyes,thenwhatisthebestwayto havesuchdiscussions?

Thequestionofthemedicalcommunity’sapproachto obesityisimportantfortworeasons.The ﬁrstiswhatthe COVID-19pandemictaughtusaboutobesity.TheCOVID19pandemicbroughtmanyofthefailuresinUShealthcare intosharpfocus,andtheobesityepidemicwasoneofthem. Ourcollectivepersonalexperienceofclinicalmedicine duringthepandemicmadeitclear:COVID-19andobesity makecatastrophicbedfellows.Ofthe2.5millionCOVID-19 deathsreportedbyFebruary2021,2.2millionoccurredin countrieswheremorethanhalfthepopulationhadabody massindexover25.7 Additionally,theCDCreportedthat obesitytripledtherateofhospitalizationforCOVID-19.8 Thesecondreasonthistopicistimelyisbecauseofthe increasingmovetonormalizeobesity,9 amovethatmakesit lesslikelyforphysiciansandpatientsaliketorecognize obesityforwhatitis:adisease.Thistypeofnormalization threatenstomakeouropeningvignettethe continuedstandard.

Ofcourse,theissueofobesityiscomplex.Obesity’ scauses aremultifactorial,andanytreatmentmust,assuch,bemultileveled.10 Inpreventativehealth,thesocialecologymodel (SEM) aframeworkforhealthpromotionthatusesa multitieredapproachtoaddresstheinterplaybetween multiplefactorsthatinfluenceagivenproblem wouldbe employedtotackleobesityatthepopulationlevel.TheSEM postulatesthatconditionssuchasobesityareshapedat severallevels:theindividual,interpersonal,organizational, community,andpublicpolicylevels.11 TheSEMpivotsaway fromanisolatedattentivenessonindividualbehaviortoa moreencompassingunderstandingoftiersofinfluence.The SEMpromotesinterventionsonthesocialdeterminantsof health,theforcesandsystemsthatshapethewell-beingofa population.Thismeansnotonlysupportingindividual effortstowardhealthydietandregularphysicalactivitybut alsochangingthecontextinwhichbehaviorsariseandare sustained.Thus,publichealthofficialscanincludetransfat eliminationinprocessedfoods,communityinfrastructure thatpromotesgreaterphysicalactivity,improvingaccessto healthyfoods,etc,aspotentialinterventions.

Givenallthis,onemightdecidethatobesityistoocomplex todealwithintheED.Nopatienttoourknowledgehasever signedintoanEDwithachiefcomplaintof “Iam overweight”;thatistosay,apatient’sweightisneverthe reasontheychoosetocometotheED.So,whytrytodiscuss whatmanyconsiderasensitivematterandriskjeopardizing one ’srapportwithapatient?WebelievetheED,while certainlynottheend-all-be-all,isstillanimportanttierof inﬂuence;so,suchdiscussionsdohaveaplacethere.

Howthencanwetalktoourpatientsaboutobesityinthe ED?Thereareguidelinesandpracticalapproachespublished inthemedicalliteraturethatprovideaframeworkinthe primarycaresettingforengagingwithpatientsabouttheir weight.12,13 TheCanadianObesityNetworkdevelopedthe “5As” ofobesitycounseling;ask,assess,advise,agreeand assist.14 GiventhebrevityoftheEDencounter,wesuggest theEDapproachshouldbeanabbreviatedversionofthe 5As,perhapsbettertermedthe3As.Itwouldinvolve ﬁrst asking permissiontodiscussthepatient’shealthrisk,onerisk beingthattheyarewellaboveidealbodyweight.Sensitivity andnon-judgmentallanguageareparamount.Ifthepatient isopentoaconversation,onecouldgoonto advise themon howtheirobesitycanhurtthem.Finally,itwouldbe importantto assist thepatientbyprovidingresourcesfor outpatientfollow-upandmanagement.Thisisnotunlike whatwealreadydowhenweprovidebriefcounselingto patientsforotherhealthconcerns.Thedifferenceisthat otherhealthconcernsarerecognizedbybothpatientsand cliniciansasaddressableissues,whilethereisapervasivelack ofrecognitionofobesityasadisease.

Butobesityisadisease.Itbehoovesusnow,morethan ever,toaddressitassuch.

AddressforCorrespondence:IyesattaM.Emeli,MD,MPH,Emory UniversitySchoolofMedicine,DepartmentofEmergencyMedicine, 531AsburyCir.,AnnexBuildingSuiteN340,Atlanta,GA30322. Email: imassaq@emory.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.Obesity:Preventingandmanagingtheglobalepidemic.Reportofa WHOconsultation. WorldHealthOrganTechRepSer. 2000;894:1–253.

2.AmericanMedicalAssociationHouseofDelegates.Resolution420. Recognitionofobesityasadisease.2013.Availableat: https://www. ama-assn.org/house-delegates/annual-meeting/proceedings-2013annual-meeting-house-delegates.AccessedNovember27,2023.

3.TiwariAandBalasundaramP.Publichealthconsiderationsregarding obesity.In: StatPearls[Internet].TreasureIsland,FL:StatPearls Publishing,2023.Availableat: https://www.ncbi.nlm.nih.gov/books/ NBK572122/. AccessedNovember27,2023.

4.Pi-SunyerX.Themedicalrisksofobesity. PostgradMed. 2009;121(6):21–33.

5.FreedmanDM,SigurdsonAJ,RajaramanP,etal.Themortality riskofsmokingandobesitycombined. AmJPrevMed. 2006;31(5):355–62.

6.TalwalkarAandMcCartyF.Characteristicsofphysicianofﬁcevisitsfor obesitybyadultsaged20andover:UnitedStates,2012. NCHSData Brief. 2016;(237):1–8.

7.WiseJ.Covid-19:Highestdeathratesseenincountrieswithmost overweightpopulations. BMJ. 2021;372:n623.

8.KompaniyetsL,GoodmanAB,BelayB,etal.Bodymassindexandrisk forCOVID-19– relatedhospitalization,intensivecareunitadmission, invasivemechanicalventilation,anddeath UnitedStates, March–December2020.(2021). MMWRMorbMortalWklyRep. 2021;70(10):355–61.

9.HarleyZ,KelloggV,MandelT,etal. AnEcologicalApproachtoObesity andEatingDisorders.Montreal,Canada:Pressbooks,2020.

10.AtkinsonRL.Etiologiesofobesity.In:GoldsteinDJ(Ed.), The ManagementofEatingDisordersandObesity (105–18).2nded. Totowa,NJ:HumanaPress.2015.

11.GlanzKandBishopD.Theroleofbehavioralsciencetheoryin developmentandimplementationofpublichealthinterventions. AnnRevPublicHealth. 2010;31:399–418.

12.VallisM,Piccinini-VallisH,SharmaAM,etal.Clinicalreview:modiﬁed5 as:minimalinterventionforobesitycounselinginprimarycare. CanFam Physician. 2013;59(1):27–31.

13.McGowanBM.Apracticalguidetoengagingindividualswithobesity. ObesFacts. 2016;9(3):182–92.

14.Rueda-ClausenCF,BenterudE,BondT,etal.Effectof implementingthe5Asofobesitymanagementframework onprovider-patientinteractionsinprimarycare. ClinObes. 2014;4(1):39–44.

MichaelKaduce,MPS*

AntonioFernandez,PhD†

ScottBourn,PhD†

DustinCalhoun,MD‡ JeffersonWilliams,MD§

MalloryDeLuca,BS,NRP§ HeidiAbraham,MD∥

KevinUhl¶

BrianBregenzer,LP∥

BaxterLarmon,PhD# RemleP.Crowe,PhD†

AlisonTreichel,MPH† J.BrentMyers,MD†

SectionEditor:JoshuaGaither,MD

ORIGINAL RESEARCH

EMSProviders:APilotStudy

*FalckHealthInstitute,Orange,California

† ESO,Austin,Texas

‡ UniversityofCincinnati,Cincinnati,Ohio

§ WakeCountyEmergencyMedicalServices,WakeCounty,NorthCarolina

∥ Austin-TravisCountyEmergencyMedicalServices,Austin,Texas

¶ CincinnatiFireDepartment,Cincinnati,Ohio

# UniversityofCaliforniaLosAngeles,DavidGeffenSchoolofMedicine, LosAngeles,California

Submissionhistory:SubmittedMay9,2024;RevisionreceivedAugust9,2024;AcceptedAugust12,2024

ElectronicallypublishedOctober1,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.21175

Background: Ourprimaryobjectiveevaluatedtheperceptionofemergencymedicalservice(EMS) providers’ reviewofautomatedhospitaloutcomedata.Secondarily,weassessedparticipationin outcomereviewasameansofmicrolearningtoobtaincontinuingeducation(CE).

Methods: FromOctober–December2023,threehigh-volumeEMSsystemsparticipatedinathree-part interventionwithresultsevaluatedusingamixed-methodsapproach.First,EMSproviders(emergency medicaltechniciansandparamedics)wereinvited,viatheirelectronichealthrecord(EHR),tocompletea presurveyevaluatingtheirperceptionsofreviewingoutcomes.Then,EMSproviderswerenotiﬁedabout theopportunitytoearnCEviaamicrolearningintervention,offeringCommissiononAccreditationfor Pre-HospitalContinuingEducation(CAPCE)-approvedCEhoursforcompletionofoutcomereviewsand associatedlearningmodules.Finally,EMSproviderswereinvitedtocompleteapost-surveymirroring thepre-survey.Qualitativeanalysesidentiﬁedthemesamongopen-endedresponses.Quantitative analysesexaminedperceptionsbetweenpre-andpost-surveys.

Results: Of843providerscontacted,217respondedtothepre-survey(25.7%).Themostendorsed rationaleforreviewingoutcomesincludedimprovingclinicalknowledge(95%),improvingpatientcare (94%),andknowingwhethercaremadeadifference(93%).Nearlyall(91%)reportedbeingmorelikelyto reviewoutcomesifCEwereawarded.Amongthe67whocompletedtheopen-endeditems,thethree dominantthemesincludedenhancepersonalconﬁdenceandcompetence(43%);acquirepersonal knowledge(39%);andoperations(21%).Of211providerswhoparticipatedintheintervention,56(27%) wereawardedCE.Atotalof152providersrespondedtothepost-survey,andthepercentagewho agreedthatreviewingoutcomesimprovesjobsatisfactionrosefrom89%to95%betweenpre-andpostsurveys(P = 0.05).

Conclusion: EMSproviderssupportedthepersonalandprofessionaldevelopmentandpatientcare improvementofreviewingpatients’ outcomeswithassociatedCE.Furtherstudyiswarrantedtoevaluatethe generalizabilityofthese ﬁndingsandthebestuserexperience.[WestJEmergMed.2024;25(6)949–957.]

BACKGROUND

Emergencymedicalservices(EMS)providersprovidethe majorityofprehospitalmedicalcareintheUnitedStatesand serveasacrucialcomponentofthenation’shealthcare deliverysystem.ForEMSproviderstomaintaintheirlicense, everystatehasuniquecontinuingeducation(CE) requirements,typicallyinthefashionofrequiredhours,for licensurerenewal.TheNationalRegistryofEmergency MedicalTechniciansalsomaintainsanhourandtopic requirementforrecertiﬁcation.Thishourmodelhasbeen usedtoensurethecontinualcompetencyofEMSproviders,1 asithasbeenhistoricallyacceptedthatcompetencycanbe maintainedthroughCEhourrequirementsdespitealackof empiricalevidence.2

Incontinuingmedicaleducation(CME),abroader conceptofcontinuedprofessionaldevelopment(CPD),is emerging.Continuedprofessionaldevelopmentincludes educationfocusedonproblemidentificationandsolution development,allowingthehealthcareprofessionaltotailor thelearningprocesstotheirindividualneeds.3 Theprocessof CPDrecognizesaone-size-fits-allapproach,whichisnot specificenoughforeachlearner.TheInstituteofMedicine recommendsaCPDsystemthatincludespatient-centered care,interprofessionalteamwork,qualityimprovement application,andclinicaloutcomedatautilizationfor individual,team,andinstitutionalassessment.2,4 Feedback hasalsobeendemonstratedtoimprovesystemperformance andpatientoutcomes.5 Verypublicvoices,includingthe EMSAgenda2050,haveincludedcallsforEMSsystemsand providerstoreceivefeedback,includingpatientoutcomes,in realtime,asameansforcontinuousqualityimprovement, thusmovingtowardaCPDapproachtocompetency.6 The NationalAssociationofEMSPhysicianshasalsocalledfor continualmonitoringofairwayperformancedataanditsuse inthecontinuedcredentialingprocessandquality managementactivitieswithlarge-scalebidirectional informationsharedbetweenEMSandreceivingfacilitiesin theirpositionpaperonairwaymanagementandtraining.7

DespitethisdesiredtransitiontoCPD,providingpatientspecificoutcomestoEMSprovidershaslongbeena challenge.BidirectionaldatasharingbetweenEMSand hospitalshasraisedconcernsaboutpatientprivacyand technologicalchallenges8,9 Fortunately,thistrendis improving,basedontheprovisionofoutcomeelementsasa partoftheNationalEmergencyMedicalServices InformationSystemdatasetandtheclarificationthatsuch data-sharingisconsistentwiththeHealthInsurance PortabilityandAccountabilityActguidelines.10–12 When surveyed,EMSprovidersreportedadesireforpatientspecificoutcomes.Theyevenreportedusinginformal networksorgoingaroundthesystemtoobtainpatient outcomeinformation,assistingthemtodevelopclinical skills.7 Thislackofinsight,specificallycenteredaround

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

EMSprovidersreceivelimitedformalclinical feedbackormicrolearningcontinuing education(CE)followingthetreatmentand transferoftheirpatients.

Whatwastheresearchquestion?

WhataretheperceptionsofEMSproviders ’ reviewofautomatedhospitaloutcomedata andassociatedCEcredit?

Whatwasthemajor ﬁndingofthestudy?

FollowingoutcomereviewandCE opportunities,surveysdemonstrated jobsatisfactionrosefrom89%to 95%(P = 0.05).

Howdoesthisimprovepopulationhealth?

EMSreviewofpatientoutcomesimprovesjob satisfactionandclinicalknowledge,thus providingameansforcontinuedcompetency ofahighlytrainedEMSworkforce.

patientoutcomesandEMSproviderdiagnosisaccuracy,has beenreportedtoimpactprovidermentalhealth.13

TheEMSproviderisinterestedinincludingpatient outcomedataasameansofprofessionaldevelopmentsimilar totheircolleaguesinmedicine.14,15 Physicians’ continued medicaleducationhasincludedelectronichealthrecords (EHR)toassessthequalityofcareandhasbeenusedto suggestareasforimprovementthroughtheuseofCME.16 Similarly,medicaldatabaseshavefoundwaystohelp physiciansreceiveCMEwithroutineclinicalquestionsand problem-solvingintheirdailypractice.17 Graduatemedical educationalsoenvisionsasysteminwhichpatienthealth recordsandoutcomedatacanbeincorporatedintothe curriculum.Thisframeworkincludesmentorsorinstructors usingtheoutcomedatatoassess,supervise,andteach, creatingamature,professionalcommunitywhereeveryone receivesandprovidesfeedback.18 TheAccreditation CouncilforGraduateMedicalEducationsimilarlyrequires programstoconnectresident-physicianeducationto patientoutcomes.19

Usingpatientoutcomedataasfeedbackcontinuestobe calledforand,onasmallscale,hasbeendemonstratedtobe aneffectivepartofCPDprogramsforEMSproviders.In fact,theEHRservesasavaluableresourceforCPDin providingpatient-speciﬁcoutcomes.20 WhenevaluatingCE

inpediatricemergencies,briefandfrequentCEprogramsare recommendedasameanstoproviderepetitionwith immediatefeedbackanderrorcorrection.21 Whilefeedback followingacallisoutsidetheproposedtheoretical frameworkforclinicaljudgmentinEMS,ithasbeennotedas importantinthedevelopmentofEMSprovidersandallows forimprovementinperformance.22 Post-resuscitation providerfeedbackforpatientswhohavesufferedcardiac arrestandheartattackshasledtoimprovementsintimeand treatment.23,24 Sammuelandcolleaguesconductedascoping reviewoftheeffectsofCPDonhealthcareprofessionals’ performanceandpatientoutcomesandwerealsoableto demonstratechangesinproviders’ behaviorand patientoutcomes.25

DespitecallsfromleadingnationalEMSorganizations andotherhealthcareprofessionstoincorporatepatient outcomesintoCE,littleisknownaboutEMSproviders’ perceptionsofautomatedpatientoutcomedatanoritsuseas microlearningtoobtaincontinuingeducation.Ourprimary objectiveinthisstudywastoevaluateEMSproviders’ perceptionsoftheutilityofautomatedhospitaloutcomedata forprofessionaldevelopment.Secondarily,weevaluated theirparticipationinoutcomereviewasameansof microlearningtoobtainCE.

METHODS

StudyDesignandPopulation

Inthismixed-methodsstudyweuedquantitativeand qualitativemethodstounderstandEMSprovider perceptionsregardingtheuseofanautomatedsystem enablingthemtoobtainCommissiononAccreditationfor Pre-HospitalContinuingEducation(CAPCE)-approvedCE byreviewingpatientoutcomesandcompletingassociated learningmodules.TheCAPCEisanaccreditingbody chargedwiththereviewandaccreditationofEMSCE.26 The studywasconductedfromOctober–December2023inthree high-volumeurbanEMSsystems.TheEMSproviderswere certiﬁedattheemergencymedicaltechnicianthroughthe paramediclevelandwerecontinuouslyprovided100%of theirrequiredCEhoursthroughtheiremployer.

Atthebeginningofthestudy,EMSprovidersineach systemreceivedapromptinvitingthemtoanonymously participateinthispilotstudywhentheyloggedintotheir EHR(ESO,AustinTX).Eachparticipatingagencyusesa systemthatallowsEMSproviderstoautomaticallyreceive outcomedatafromthehospital,speciﬁctopatientsthey encounteredintheprehospitalsetting.AllCEactivitieswere completedinasingleonlinelearningplatform.

Forthosewhoagreedtoparticipateinthe ﬁrstphaseofthe pilotstudy,participantswereaskedtovoluntarilycompletea pre-survey.Theywerealsoinformedthattheywouldbe eligibletoreceiveCEhours,approvedbyCAPCE,upon completionofoutcomereviews,associatedvideos,and outcomereviewassessments.Aprompttoreviewoutcomes

aspartofthisstudywasreceivedbyparticipantswhenthey loggedintotheEHRandnavigatedtotheoutcomereview section.TheCEphasewasopenforatleasttwoweeks.This wasfollowedbyanotherinvitationintheEHRtocomplete thepost-surveythatmirroredthepre-survey.Participants wererequiredtospeciﬁcallyoptintoeachsurveyandtheCE phaseofthestudy.

CAPCE-ApprovedContinuingEducation

DuringtheCEphase,participantswereprovidedalink fromtheoutcomereviewpagethatdirectedthemtologinto thelearningplatform.Twoprerequisitevideosprovided backgroundinformationanddescribedthetypesofoutcome informationavailable:InternationalClassiﬁcationof DiseasesdiagnosiscodesandCentersforMedicareand MedicaidServicesdisposition(eg,dischargetohome, dischargetoskillnursing,transfer,death,etc).Upon completionoftheseprerequisites,participantswereableto reviewtheirpatient-speciﬁcoutcomesandansweraseriesof questionsregardingeachreview.

DataAnalysis

Weusedquantitativeanalysestoexamineperceptions betweenpre-andpost-surveysusingchi-squaretestsor Fisherexacttests,asappropriate.Likert-scaleresponse options(agreevsdisagree)anddemographics(Level: paramedicvsother;Role:patientcareprovidervsother; yearsofexperience:0–4years,5–10years,11–20years, ≥21years;previousfrequencyofoutcomesreview:very frequently/frequentlyvsoccasionally/rarely/never)were collapsed,asneededandwhereappropriate,duetocellsize. ParticipationintheCAPCECEphaseofthestudywasalso quantiﬁed.Dataisreportedaspercentagesandfrequencies. P -valuesofstatisticaltestsarealsoincluded.Weperformed quantitativeanalysisusingSTATAMPversion18.0 (StataCorpLLC;CollegeStation,TX).

Qualitativeanalysesidentiﬁedthemesamongopen-ended responses.Weusedconventionalcontentanalysisas describedbyHsiehandShannon(2005).Aninductive approachwasusedtoextractmeaningandthemesfrom responses.Allcodesweregeneratedfromthecontentdirectly withoutapriorithemes.Oncecodeswereidentiﬁed,weused adeductiveapproachtodeterminethefrequencyand distributionofthemesaswellasanyrelationshipsthat existedbetweenthemesandrespondentcharacteristics. QualitativeanalysiswasassistedusingtheQDA MinerLite3.0(ProvalisResearch;Montreal,Quebec) softwarepackage.

ThisstudywasdeemedexemptbytheUniversityof CaliforniaLosAngelesInstitutionalReviewBoard.

RESULTS

Duringthestudyperiod,843EMSprovidersfromthe participatingagenciesloggedintotheEHRandwere

thereforeeligibletoparticipate.Atotalof217(25.7%)EMS providersanonymouslyrespondedtothepre-surveywhileon duty,and152(18.0%)anonymouslyrespondedtothepostsurveywhileonduty.Demographicsweresimilaramong thosethatrespondedtothepre-andpost-survey(Table1).

Overall,responsesweresimilarwhencomparingthose whorespondedtothepre-surveyandthosewhorespondedto thepost-survey(Table2).Notably,89%inthepre-surveyvs 95%inthepost-surveyindicatedagreementthatoutcomes reviewenhancedjobsatisfaction(P = 0.05).When evaluatingonlythosewithaparamediccertification,wesaw astatisticallysignificantincrease(P = 0.03)inthepercentage ofthoseindicatingtheyreviewoutcomestoimproveclinical knowledgewhencomparingpre-surveyrespondentstopostsurveyrespondents(94%vs99%,respectively).Wealso notedthatpre-andpost-surveydifferenceswithrespectto improvementsinjobsatisfactionremainedsignificantwhen onlyevaluatingparamedics.Nostatisticallysignificant differencewasnotedamongpre-andpost-surveyrespondents ofthosewithothercertificationlevels(TableS1).When

examiningyearsofEMSexperienceorroleintheEMS system,responsesbetweenpre-andpost-surveyrespondents weresimilarregardlessofhowmanyyearstheindividualhad workedinEMS(TableS2).

Thereweretwosignificantdifferenceswhencomparing thosewhorespondedtothepre-surveyandthosewho respondedtothepost-surveywhenstratifiedbyhistorical frequencyofoutcomesreview.Amongthosewhoindicated thattheyreviewpatientoutcomesfrequentlyorvery frequently,wesawasignificantlyhigherpercentageofpostsurveyrespondentsindicatingthattheyreviewedpatient outcomestoobtainclosure(90%vs98%, P = 0.01, respectively)andthatreviewingpatientoutcomesimproves jobsatisfaction(92%vs99%, P = 0.02,respectively) (TableS3).

CAPCE-ApprovedContinuingEducation

Atotalof211individualsfromthethreeparticipating EMSsystemsoptedintotheCAPCECEphaseofthispilot study.Amongthose,63%(133/211)successfullyloggedinto

%(n)Pre-survey(N = 217)Post-survey(N = 152)

Certi ﬁcationlevel

Emergencymedicaltechnician(EMT)14.8%(32)11.2%(17)

Advancedemergencymedicaltechnician(AEMT)2.3%(5)3.3%(5)

Paramedic77.9%(169)82.9%(126)

Other1.8%(4)0.0%(0)

None3.2%(7)2.6%(4)

Rolewithintheorganization

Patientcareprofessional61.2%(131)71.3%(107)

First-linesupervisor20.6%(44)18.0%(27)

Administrator/manager4.7%(10)1.3%(2)

Preceptor3.7%(8)3.3%(5)

Educator0.9%(2)0.7%(1)

Other8.9%(19)5.3%(8)

Yearsofexperience(4categories)

0–4years18.1%(39)18.4%(28)

5–10years28.2%(61)28.9%(44)

11–20years31.5%(68)28.9%(44)

≥21years22.2%(48)23.7%(36)

Frequencyofpatientoutcomesreview

Veryfrequently(morethanonceperweek)38.8%(83)36.2%(55) Frequently(aboutonceperweek)32.2%(69)33.6%(51) Occasionally(aboutonceortwicepermonth)19.2%(41)25.7%(39)

Rarely(aboutonceortwiceperyear)3.3%(7)2.0%(3) Veryrarely(lessthanonceperyear)2.8%(6)2.6%(4)

Never(Ihaveneverviewedanoutcome)3.7%(8)0.0%(0)

Table1. Participantcharacteristics.

Table2. Pre-surveyandpost-surveyoverallresponses.

Pre-surveyPost-survey

%(n)DisagreeAgreeDisagreeAgree P-value

IreviewoutcomestoimprovethecareIprovidetopatients.6.0%(12)94.0%(188)4.1%(6)95.9%(141)0.43

Ireviewoutcomestoimprovemyclinicalknowledge.5.5%(11)94.5%(188)2.1%(3)97.9%(140)0.17

Ireviewoutcomestoknowwhethermycaremadeadifference.7.0%(14)93.0%(186)4.1%(6)95.9%(140)0.26

IreviewoutcomestoknowwhetherIprovidedtherightcare.6.9%(14)93.1%(188)2.8%(4)97.2%(141)0.09

Ireviewoutcomestoobtainclosureonpatientencounters.14.0%(28)86.0%(172)8.9%(13)91.1%(133)0.15

Reviewinghospitaloutcomedatathroughthepatientoutcomes featurehelpsimprovemyjobsatisfaction. 11.3%(24)88.7%(188)5.3%(8)94.7%(142)0.05

Reviewinghospitaloutcomedatathroughthepatientoutcomes featurehelpsimprovemyclinicalknowledge. 6.8%(14)93.2%(192)6.9%(10)93.1%(135)0.97

IfIwereprovided15minutesofapprovedcontinuingeducation creditforeachpatientoutcomeIreviewed,Iwouldbemorelikely toreviewmypatientoutcomesinthepatientoutcomesfeature.

theESOlearningplatform.Ofthosewhosuccessfullylogged in,74%(98/133)completedtheirrequiredproﬁle.The prerequisiteswerecompletedby70%(64/98)ofthosewitha completedproﬁle,and88%(56/64)ofthosecompletedat leastoneCEactivity.AmongthosewhocompletedanyCE, 88%(35/64)completedmorethanoneCEactivityandless than10%(4/64)completedallavailableCEactivities (Figure1).Atotalof287outcomeswerereviewed,and responsesfollowingtheoutcomereviewactivityarelisted in Table3.

QualitativeAnalysis

Therewere72pre-surveyparticipantswhoansweredthe free-textquestionthatasked, “Pleasedescribeanyother reasonsyoureviewoutcomes.” Free-textresponseswere uploadedintoQDADataMinerforanalysis.Allcodingwas performedbyasingleresearcher(SB)andreviewedbyAF andMK.Mostinitialcodesrelatedtoreasonsrespondents reviewedtheoutcomesorbenefitsgainedbydoingso.The mostcommoncodeidentifiedwas improveunderstandingof patientsI ’ veseen, whichwasidentified14times(13%ofall codes).Thesecondmostidentifiedcodewasaf fi rming(that mydiagnosisandtreatmentwerecorrect), identifiedeight times(7%ofallcodes).Weexcludedfromfurtheranalysis10 respondentswhoseonlytextcontentwascodedas “donot reviewoutcomes” or “wishlist” (acodeforentriesthat describedfeaturestheywantedtoseeinthefuture).

Wegroupedthecodesintothemesorganizedbycommon attributes.Theseinitialcodeswereorganizedbybeneficiary ofpatientoutcomereviewCE:therespondent;thepatient encounter;andtheEMSoperation(FigureS1).Notably, unknown, agroupcreatedforcodesthatdidn’tseemto fitany oftheidentifiedthemes,comprised17%ofallcodes.Asa result,weperformedasecondthematicanalysisinanattempt tointegrate “unknown” codesandidentifyhigherlevel

meaningsbehindthecodes.Theresultingthemesrepresented broaderpersonal,operational,andsystem-relatedbeneﬁtsto reviewingpatientoutcomes.Theserevisedthemesandtheir associatedcodeswerereviewedbyAFandMKwithminor revisionsrecommendedandintegrated.

Opted in to CAPCE CE Phase (211)

Successfully Logged into ESO ODL (133)

Created Profile with CAPCE required information (98)

Completed all Prerequisites (64)

Did not log into ESO ODL (78) Incomplete Profile (35) Incomplete Prerequisites (34) CE activities incomplete (8)

Completed at least 1 outcome review and earned CE (56)

Figure1. CAPCECEoptionalparticipation. CAPCE,CommissiononAccreditationforPre-Hospital ContinuingEducation; CE,continuingeducation.

Table3. Continuingeducationactivityreviewquestions andresponses.

Allcontinuingeducationreviews

Goalsofyouroutcomereview(selectallthatapply):

Obtaincontinuingeducationcreditfor licensurerenewal.

Increasemyknowledgeregardingtheclinical conditionofthepatient.

Determinewhethermyassessmentand treatmentswereappropriate.

Determinewhethermy ﬁeldimpressionaligns withthehospitaldiagnosis.

Basedonthisreview:

22%(224)

25%(247)

26%(258)

27%(267)

Iwouldprovidethesametreatment.84%(242)

Ineedmoreinformation.6%(17)

Iwouldmodifymytreatment.10%(28)

Indicatehowstronglyyouagreeordisagreewith thefollowing:Myuseofoutcomesreinforcesor increasesmyclinicalknowledge.

4 – stronglyagree52%(149)

3 – agree46%(132)

2 – disagree0%(1)

1 – stronglydisagree2%(5)

Indicatehowstronglyyouagreeordisagreewith thefollowing:Myuseofoutcomesimprovesthe qualityofcareIprovidetomypatients.

4 – stronglyagree50%(142)

3 – agree49%(139)

2 – disagree1%(3)

1 – stronglydisagree1%(2)

Indicatehowstronglyyouagreeordisagreewith thefollowing:IfIwereprovidedadditionalonline continuingeducationopportunitiesrelatedtothis patientoutcome,Iwouldcompletethetraining.

4 – stronglyagree44%(127)

3 – agree55%(156)

2 – disagree0%(1)

1 – stronglydisagree1%(2)

Therewereshiftsintheperceivedvalueofpatientoutcome reviewbetweenthepre-andpost-surveyperiodsandacross individualswithdifferentbackgroundsandexperience (Figure2).Thoserespondentswhoindicatedtheydidnot providedirectpatientcareplacedthemostvalueon operationsandsystemwidequality.Fortheremainderof respondents,pre-surveyresultsfoundthatthepredominant valueidentifiedwasrelatedto acquiringpersonalknowledge, with enhancingpersonalcon fi denceandcompetence, and improvingpersonalclinicalpractice following nearly10% lower.Duringthepost-survey, acquiringpersonal knowledge felltothefourthposition,with enhancing personalcon fi denceandcompetence thepredominant theme,followedby improvingpersonalclinicalpractice and operations. Inthepost-surveypopulation, acquiring personalknowledge remainedvaluedbyEMTsand individualswith0–4yearsofexperiencebutwasnotvaluedat allinthosewith ≥21yearsofexperience.

DISCUSSION

EnhancePersonalConﬁdence/CompetenceandImprove PatientCare

Whensurveyedanonymously,EMSprovidersfavored viewingtheirpatientoutcomes.Qualitativeanalysisofpresurveyresultsrevealedthatthemajorityofcertified personnelfelttheprimarybenefitwouldbeincreasesin personalknowledge;thisattitudewasparticularlyprevalent forEMTsandindividualswhohadlimitedexperience.In post-surveyresults,providersfoundthatreviewingpatient outcomeswasmorerelevanttogainingconfidenceand competenceratherthanknowledgeacquisition.McGuire andcolleaguesevaluatedfeedbackrequestsreceivedfrom EMSprovidersandfoundthemostcommonrequestwasfor the finaldiagnosisandoutcome/disposition.28 This enhancementiscommonlyreflectedinEMSproviderswho assessandcarefortheirpatientsbutarelimitedtoonlythe beginningofthepatient’sexperience,notthe finaloutcome. Providingthemwiththeoutcomeoftheirpatientallowsthem toeitherincreaseconfidenceintheirassessmentand diagnosisorparticipateinanopportunityforcontinued learning.This findingrepresentstheworkforce’sdesireforan outcome-centriccontinuedcompetencemodelover traditionalformsofCE.

The ﬁnalthemesandcodegroupingscanbefoundin FigureS2.Webelievethesecodesandthemesaccurately reﬂectsurveyresponses,andweusedthemforanalysisofthe post-survey,free-textquestionusingthesamemethod.

Table4 providesanoverviewofthethemes,dominantcodes, andkeyquotesforpre-andpost-surveygroups.Inaddition, wecomparedthematicfrequencyreportsbetweentheentire surveygroupandvariousdemographicgroupsby respondentlevel,experience,andfrequencyofoutcome review.Theseresultscanbefoundin TableS1–S4.

Providersalsoperceivedimprovedpatientcarebyviewing theirhospitaloutcomedata,whichpreviousworkhas demonstratedisbeneficialtothelearningprocessandcan improvepatientoutcomes.Thequalitativereviewofpostsurveyresultsrevealedanincreaseintheemphasison enhancingpersonalconfidenceandcompetenceand improvingpersonalclinicalpractice,suggestingthat providersseethepotentialforthiseducationtoimprovetheir clinicalpracticeratherthanjustprovidingknowledge.Postresuscitationfeedbackhasalsobeendemonstratedto improvethequalityofAdvancedLifeSupport,specificallyin

Table4. Overviewofthemes,codes,andkeyquotesfromthepre-andpost-survey.

ThemeDescriptionDominantcodespre-surveyDominantcodespost-survey

Acquirepersonal knowledge

Enhancepersonal conﬁdenceand competence

Codesthatreﬂectlearningaboutclinical presentations,medicalknowledge,and hospitalrelatedtopatientsseen.

• Improvemyunderstandingof patientsI’veseen

• Furtherclinicalknowledge

• Improvemyunderstandingof patientsI’veseen

Keyquote: Tofollowuponpatientencountersandunderstandthecarethepatientlaterreceived

CodesthatafﬁrmthatEMSdiagnosis andcarewereappropriate,reﬂectonthe impactofEMScareonpatientoutcome, andcontributetoselfQA/QI.

• Afﬁrming

• SelfQA/QI

• Afﬁrming

Improvepersonal clinicalpractice

Keyquote: Ireviewchartstoreviewmy ﬁndingsandQA/QImyselfforfuturepatients

Codesthatreﬂectimprovementin personalclinicalpracticeincluding informingfuturediagnosisandtreatment ofrecurringpatients.

• Futurebetterment

• Informfuturedifferential diagnosis

• Informfuturevisitsto samepatient

• Futurebettermentto improvecare

Keyquote: Helpsprioritizefollow-upsandwhatresourcesshouldbebroughttobearforourclients

OperationsCodesreﬂectingimpactonoperational tasksincludingdocumentation,followupsservicesandtraining.

• It’smyjob

• Researchbillinginquiry

SupportsystemwideQI

• Training/reinforcement fortrainees

• Informfollow-up postdischarge

Keyquote: Ireviewoutcomedatawith employees ithelpsuslearntogether,buildingtheeducationalsafetyamong ourgroup

Codesreﬂectingcollaboration,cultureof learningandeducation,andorganization wideQA/QI.

• Organization-widecompares crewmedicaldecisionmaking, protocolutilizationtopatient diagnosisandoutcome

• SystemQA/QI

• Createacultureoflearning andeducation

Keyquote: Ithelpscreateacultureinwhichlearningandeducationarethefocus,noterrorsormissteps

EMS,emergencymedicalservices; QA,qualityassurance; QI,qualityimprovement.

Figure2. Comparisonofthemespre-topost-survey.

survivaluntilhospitaldischargeandfavorableneurological outcomesandiswhyreal-life,post-resuscitationfeedbackis recommended.23 Timetotreatmentinheartattackvictims hasalsobeenshowntodecreasefollowingthe implementationofdatafeedbacktoEMSproviders,evenin asystemthatwasalreadyachievinginternationally establishedgoals.24 Sammuelandcolleaguesconducteda scopingreviewoftheeffectsofCPDonhealthprofessionals’ performanceandpatientoutcomesanddemonstrated changesinproviders’ behaviorandpatientoutcomes.25 SurveyresultspublishedbyPollardandBlackfoundthat afterreceivingpatientoutcomedata,mostEMSproviders reﬂectedonthecallanddidfurtherreading.29

EMSProviderJobSatisfaction

Studyparticipantsfeltthatlearningtheirpatient-speci fic hospitaloutcomesimprovedjobsatisfaction.Similarresults havedemonstratedthatwhenasked,EMSproviders anticipatepatientoutcomefeedbackbenefitingtheirwellbeingandworkengagement.14 Providingpatientoutcomes helpsbringEMSprovidersfurtherintothehealthcare continuum,anddoingsoreducesthecausesofburnoutfor them.30 Thisisanimportant findingasagenciesfacea workforceshortageandtheUSseesfewerandfewerEMTs andparamedics.31

Microlearning

TheCAPCE-approvedCEwaspresentedtoEMS providersintheformofmicrolearning.Microlearning consistsofsmalldosesofcontentintheformoflesson modulesorshort-termactivities.Thismethodallowslearners tocontrolallaspectsoftheirlearning,includingthetimein whichtheyreview,thepace,andthemethodbywhichthey completetheactivity.Providersreportedbeingmorelikelyto completethereviewiftheywereprovidedwithameanstouse ittocompletecertiﬁcationrenewal;thus,ifanagency providespatientoutcomestoproviders,itisbesttoincludeit asameansforcertiﬁcationrenewal.

Thispatientoutcome-speciﬁceducationdescribesanew methodofCEforEMSprovidersandalignswiththegoalsof CPD.Additionally,patientoutcome-lededucationhasbeen demonstratedtoimproveprovidercompetencyandimprove patientoutcomes.Morestudyisneededonawiderscaleto determinewhetherthistypeofeducationdeliveryensures morecompetencythananhour-basedmodel.Thistypeof learningmayalsobeappealingfromanoperational perspective,providing ﬂexibilitywithrespecttoscheduling educationalofferingsandreducingthetimethatEMS providersarerequiredtospendinaclassroom.

LIMITATIONS

Thesamplesizeisspeciﬁctothesethreeagencies (conveniencesample)andthosewhochosetocompletethe survey;assuchitmaynotbegeneralizabletotheentireEMS workforce.Thewindowforcompletingthesurveywasopen fortwoweeks;thus,theremayhavebeensomeproviderswho didnotworkinthatperiodandcouldnotcompletethe survey.Additionally,therequesttocompletethesurvey occurredduringoperationalhours;so,participationmay havebeenlimitedduetoother,moreurgenttasks.Byits nature,qualitativeanalysisisheavilydependentonthe backgroundandskillsoftheresearcherandmaybe inﬂuencedbypersonalbiases.

AlowproportionofproviderswhobegantheCEprocess actuallygainedCEcredits,indicatingthestepsorprocess maybetoodifﬁcultorcumbersome.Itisalsopossiblethat participantsdidnotneedtheCEcredits,aseachofthe participatingagenciesprovides100%oftherequiredhours.

Additionalresearchisneededtodeterminehowtobest integratethislearningmethodwithexistingworkpatterns andCEprograms.

CONCLUSION

Emergencymedicalservicesproviderssupportedthe personalandprofessionaldevelopmentandpatientcare improvementvalueofreviewingpatientoutcomes, includingmicrolearningactivities.Participation intherequiredactivitiestoobtaincontinuingeducationwas low.However,subjectswhodidparticipatedemonstrateda shiftinperceivedvaluefrommereacquisitionof knowledgetodevelopmentofimprovedpersonaland systemwideclinicalpractice.Furtherstudyiswarrantedto evaluatethegeneralizabilityofthese ﬁndingsandthe bestuserexperiencetofacilitatethecompletion ofCE.

AddressforCorrespondence:MichaelKaduce,MPS,FalckHealth Institute,1517W.BradenCt.,Orange,CA92868.Email: Michael. Kaduce@falck.com

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.AntonioFernan,ScottBourn,RemleP. Crowe,AlisonTreichel,andJ.BrentMyersareemployeesofESO. Therearenootherconflictsofinterestorsourcesoffunding todeclare.

REFERENCES

1.TerryM,PowellJ,GilmoreWS,etal.Derivingnationalcontinued competencyprioritiesforemergencymedicalservicesclinicians. PrehospEmergCare. 2023;27(4):439–48.

2.StudnekJR,FernandezAR,MargolisGS.Assessingcontinued cognitivecompetenceamongruralemergencymedicaltechnicians. PrehospEmergCare. 2009;13(3):357–63.

3.DeSiletsLD.TheInstituteofMedicine’sredesigningcontinuing educationinthehealthprofessions. JContinEducNurs. 2010;41(8):340–1.

4.BalmerJT.Thetransformationofcontinuingmedicaleducation(CME)in theUnitedStates. AdvMedEducPract. 2013;4:171–82.

5.IversN,JamtvedtG,FlottorpS,etal.Auditandfeedback:effectson professionalpracticeandhealthcareoutcomes. CochraneDatabase SystRev. 2012;(6):CD000259.

6.(2019).EMSAgenda2050TechnicalExpertPanel,NationalHighway TrafﬁcSafetyAdministration,DepartmentofTranportation.EMS Agenda2050:apeople-centeredvisionforthefutureofemergency

medicalservices,reportNo.DOTHS812664.Washington,DC: NationalHighwayTrafﬁcSafetyAdministration.

7.DorsettM,PanchalAR,StephensC,etal.Prehospitalairway managementtrainingandeducation:anNAEMSPpositionstatement andresourcedocument. PrehospEmergCare. 2022;26(sup1):3–13.

8.RedlenerM,OlivieriP,LooGT,etal.Nationalassessmentofquality programsinemergencymedicalservices. PrehospEmergCare. 2018;22(3):370–8.

9.DelbridgeTR,BaileyB,ChewJLJr.,etal.EMSagendaforthe future:whereweare wherewewanttobe. PrehospEmergCare. 1998;2(1):1–12.

10.ChahineS,KulasegaramKM,WrightS,etal.Acalltoinvestigatethe relationshipbetweeneducationandhealthoutcomesusingbigdata. AcadMed. 2018;93(6):829–32.

11.VithalaniV,SondheimS,CorneliusA,etal.Qualitymanagementof prehospitalairwayprograms:anNAEMSPpositionstatementand resourcedocument. PrehopEmergCare. 2022;26(sup1):14–22.

12.Page,Wolfberg&Wirth.Animaginarybarrier:howHIPAApromotes bidirectionalpatientdataexchangewithemergencymedicalservices. 2019.Availableat: https://nemsis.org/wp-content/uploads/2020/07/ HIPAA_An-Imaginary-Barrier-to-Data-Exchange.pdf AccessedJuly21,2024.

13.MorrisonL,CassidyL,WelsfordM,etal.Clinicalperformancefeedback toparamedics:whattheyreceiveandwhattheyneed. AEMEducTrain. 2017;1(2):87–97.

14.Eaton-WilliamsP,MoldF,MagnussonC.Exploringparamedic perceptionsoffeedbackusingaphenomenologicalapproach. Br ParamedJ. 2020;5(1):7–14.

15.O’HaraR,JohnsonM,SiriwardenaAN,etal.Aqualitativestudyof systemicinﬂuencesonparamedicdecisionmaking:caretransitionsand patientsafety. JHealthServResPolicy. 2015;20(1Suppl):45–53.

16.CampbellEGandRosenthalM.Reformofcontinuingmedical education:investmentsinphysicianhumancapital. JAMA. 2009;302(16):1807–8.

17.WoltersKluwer.ContinuingmedicaleducationwithUpToDate.2024. Availableat: https://www.wolterskluwer.com/en/solutions/uptodate/ about/continuing-medical-education.AccessedJuly21,2024.

18.CateOTandCarraccioC.Envisioningatruecontinuumofcompetencybasedmedicaleducation,training,andpractice. AcadMed. 2019;94(9):1283–8.

19.ZafarMA,DiersT,SchauerDP,etal.Connectingresidenteducationto patientoutcomes:theevolutionofaqualityimprovementcurriculumin aninternalmedicineresidency. AcadMed. 2014;89(10):1341–7.

20.KleinD,StaplesJ,PittmanC,etal.Usingelectronicclinicalpractice auditsasneedsassessmenttoproduceeffectivecontinuingmedical educationprogramming. MedTeach. 2012;34(2):151–4.

21.LammersRL,Willoughby-ByrwaMJ,VosDG,etal.Comparisonoffour methodsofparamediccontinuingeducationinthemanagementof pediatricemergencies. PrehospEmergCare. 2022;26(4):463–75.

22.GugiuMR,McKennaKD,PlattTE,etal.Aproposedtheoretical frameworkforclinicaljudgmentinEMS. PrehospEmergCare. 2023;27(4):427–31.

23.HubnerP,LobmeyrE,WallmullerC,etal.Improvementsinthequalityof advancedlifesupportandpatientoutcomeafterimplementationofa standardizedreal-lifepost-resuscitationfeedbacksystem. Resuscitation. 2017;120:38–44.

24.ScholzKH,HilgersR,AhlersmannD,etal.Contact-to-balloontimeand door-to-balloontimeafterinitiationofaformalizeddatafeedbackin patientswithacuteST-elevationmyocardialinfarction. AmJCardiol. 2008;101(1):46–52.

25.SamuelA,CerveroRM,DurningSJ,etal.Effectofcontinuing professionaldevelopmentonhealthprofessionals’ performanceand patientoutcomes:ascopingreviewofknowledgesyntheses. AcadMed. 2021;96(6):913–23.

26.CommissiononAccreditationforPre-HospitalContinuingEducation. AboutCAPCE.Availableat: https://www.capce.org/Home/About AccessedApril2024.

27.HsiehHFandShannonSE.Threeapproachestoqualitativecontent analysis. QualHealthRes. 2005;15(9):1277–88.

28.McGuireS,LukeA,KlassenA,etal.11Whatmedicswant:analysisof feedbackandpatientfollow-uprequestsfromout-of-hospitalproviders atanacademicemergencydepartment. AnnEmergMed. 2021;78(4):S5.

29.PollardJandBlackS.Doparamedics ﬁnditbenefecialtolearnthe diagnosisgiventotheirpatientsintheemergencydepartment? Emerg MedJ. 2015;32(5):421–1.

30.TorresAJandMcCoyRG.HowtobettervalueEMScliniciansaskey careteammembers. AMAJEthics. 2022;24(9):E898–905.

31.CashRE,RivardMK,ChrzanK,etal.Comparisonofvolunteerandpaid EMSprofessionalsintheUnitedStates. PrehospEmergCare. 2021;25(2):205–12.

32.JanssenA,CogginsA,TadrosJ,etal.Deliveringadaptiveonline learningusingelectronichealthdata:afeasibilityandacceptabilitystudy withemergencytrainees. ResSq. 2023.Availableat: https://doi.org/10. 21203/rs.3.rs-2395367/v1

ORIGINAL RESEARCH

ValidationoftheTurkishVersionoftheProfessional FulﬁllmentIndex

MerveEksioglu,MD*

AycaKoca,MD†

BurcuAzapogluKaymak,MD*

TubaCimilliOzturk,MD*

AtillaHalilElhan,PhD‡

SectionEditor:TehreemRehman,MD,MPH

*UniversityofHealthSciencesFatihSultanMehmetEducationandResearch Hospital,DepartmentofEmergencyMedicine,Istanbul,Turkey

† AnkaraUniversityFacultyofMedicine,DepartmentofEmergencyMedicine, Ankara,Turkey

‡ AnkaraUniversityFacultyofMedicine,DepartmentofBiostatistics, Ankara,Turkey

Submissionhistory:SubmittedMay14,2024;RevisionreceivedApril15,2024;AcceptedAugust16,2024

ElectronicallypublishedSeptember25,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.21199

Introduction: Clinicianburnoutrepresentsasignificantoccupationalhazardamongphysicians,witha notablyhighprevalenceamongemergencyphysicians.TheStanfordProfessionalFulfillmentIndex (PFI)wasdevelopedtocomprehensivelyassessvariousaspectsofdoctors’ workexperiences,including professionalfulfillment.InthisstudyweaimedtovalidatetheTurkishversionofthePFI(T-PFI),a16-item instrumentdesignedtomeasurephysicians’ professionalfulfillmentandburnout.

Methods: Inthiscross-sectionalstudy,wevalidatedtheT-PFIintwophases.Theinitialphaseinvolved translatingandculturallyadaptingtheoriginalPFIintoTurkish.Weevaluatedthecontentvalidityofthe translatedversionusingitemandscalecontentvalidityindices(I-CVIandS-CVI,respectively).The validatedT-PFIwasthendistributedamongabroadcohortofemergencyphysiciansviaanonlinesurvey tofurtherassessitsreliabilityandvalidity.TheassessmenttoolsincludedCronbach α,conﬁrmatory factoranalysis,andcontentvalidityindices.

Results: Of1,434physicianswhoweresentthesurvey,425fullycompletedit(29.6%).Therewasan almostequaldistributionof215femalesand210males.Only9.6%oftheparticipantsreportedhigh levelsofprofessionalfulfillment,whereasasignificantmajority(79.1%)weresusceptibletoburnout. TheCronbach α valuesfortheprofessionalfulfillmentandoverallburnoutscaleswere0.87and 0.90,respectively.ThecontentvaliditywasconfirmedbyI-CVIvaluesexceeding0.80andan S-CVI/averagerelevanceof0.92.Theconfirmatoryfactoranalysisdemonstratedanacceptablemodel fit afteradjustments.

Conclusion: TheT-PFIisareliableandvalidtoolforassessingprofessionalfulﬁllmentandburnout amongemergencyphysiciansinTurkey.Effectiveinterventionstomitigateburnoutareessentialto improvephysicianwell-beinginTurkishhealthcaresettings.[WestJEmergMed.2024;25(6)958–965.]

INTRODUCTION

Burnoutisasyndromeresultingfromchronicworkplace stressandoverload.1 Itischaracterizedbythreedimensions: emotionalexhaustion;depersonalization;andreduced personalaccomplishment.2 Studieshaveshownthat physiciansexperiencehigherratesofburnoutand lowersatisfactionwithwork-lifeintegrationcompared

tothegeneralpopulation.3 Emergencyphysicians,in particular,arehighlysusceptibletoburnout,withrates surpassingthoseseeninothermedicalspecialties.3 Researchontheworkexperiencesofemergencyphysicians haslargelyfocusedonindividualandworkplacefactors contributingtoburnout.4,5 Burnoutnotonlyaffects clinicians’ healthbutalsonegativelyimpactspatient

careoutcomesandtheoverallfunctioningof healthcaresystems.6,7

Toeffectivelymeasureburnoutamonghealthcare professionals,itiscrucialtoassessnotonlythenegative aspects,suchasemotionalexhaustionand depersonalization,butalsothepositiveaspectssuchas professionalsatisfaction,well-being,andoccupational fulfillment.8–11 Unliketraditionalburnoutscales,the StanfordProfessionalFulfillmentIndex(PFI)wasdeveloped toprovideacomprehensiveassessmentofphysicians’ work experiences.10 ThePFIisaconcisetoolspecificallydesigned forphysicians,offeringvaluableinsightsintotheirwellbeing,qualityoflife,andproductivity.Ithelpshealthcare organizationsevaluatebothburnoutandprofessional satisfactionlevels,enablingthedevelopmentofprogramsto enhancejobsatisfactionandoverallwell-being.However, forthePFItobeeffectiveacrossdifferentculturalcontexts, itmustbetranslatedandadaptedtoensureitsvalidity andrelevance.ThePFIhasbeensuccessfullyadapted incountriessuchasBrazilandJapan,butfurthervalidation isneededtoconfirmitspsychometricstrengthand culturalapplicability.12,13

InthisstudyweaimedtodevelopandvalidatetheTurkish PFI(T-PFI)toassesstheprofessionalfulfillmentand burnoutlevelsofemergencyphysiciansinTurkey.The T-PFIisbasedontheStanfordPFIandhasbeentranslated andadaptedto fitthespecificculturalandhealthcarecontext ofTurkey.Wethenevaluatedthepsychometricpropertiesof theT-PFItoensureitsreliabilityandvalidity.Ourgoalwas toprovidearobusttoolforaccuratelyassessingand monitoringtheprofessionalwell-beingofemergency physiciansinTurkey.

METHODS

Participants

Theresearchteamcollaboratedwiththeheadsofhighvolumeemergencydepartments(ED)inprominent healthcareinstitutionsacrossTurkeytorecruitparticipants. Recognizingthepotentialforalowresponserateduetothe demandingschedulesofemergencyphysicians,weemployed severalstrategiesrecommendedbytheliteraturetoimprove responseratesandmitigatenon-responsebias.14 These strategiesincludedsendingmultipleremindersandusing variouscommunicationplatformstodistributethe surveylink.

Weinvitedemergencyphysiciansafﬁliatedwiththese organizationsthroughemails,socialmediagroups,and internalcommunicationplatforms,suchasonlinemessaging systems.Toensureclarityforpotentialparticipants,we providedadetailedexplanationofthestudy’spurposeand objectives.Eligibleparticipantsincludedphysiciansactively workinginEDs.Whileourprimaryaimwastoidentify burnoutamongthesehealthcareprofessionals,weexpanded thestudytoincluderesidents,attendingphysicians,

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Emergencyphysiciansexperiencehighrates ofburnout,negativelyimpactingtheirwellbeingandpatientcare.

Whatwastheresearchquestion?

IstheTurkishversionoftheProfessional Ful ﬁ llmentIndex(T-PFI)validandreliable?

Whatwasthemajor ﬁndingofthestudy?

TheT-PFIshowedstronginternalconsistency (Cronbach α :0.87forPF,0.90forburnout).

Howdoesthisimprovepopulationhealth?

TheT-PFIoffersareliablemeasureof professionalful ﬁ llmentandburnout,enabling healthcareorganizationstomonitor physicianwell-being.

consultantphysicians,andfacultymembers.Withthis broaderscopeweaimedtocaptureadiverserangeof perspectives,ensuringcomprehensiverepresentationof healthcareprofessionalswithinemergencymedicine(EM).

Wesentthesurveyto1,434medicaldoctors.Participants providedelectronicinformedconsentbeforecompletingthe survey,conﬁrmingtheirvoluntaryparticipationand understandingofthestudy’sobjectivesandprocedures.To protecttheirprivacyandencouragecandidresponses,we designedthesurveytobeanonymous.Weincludedonlyfully completedquestionnairesinthestatisticalanalysis,resulting inaresponserateof29.6%.

Procedure

Weconductedthiscross-sectionalstudyintwophasesto validatetheT-PFI.Inthe ﬁrstphase,wetranslatedand culturallyadaptedtheoriginalPFIintoTurkish,adheringto WorldHealthOrganizationguidelinesforlinguisticaccuracy andculturalrelevance.15 Twobilingualemergency physiciansperformedtheforwardtranslationofthePFI fromEnglishtoTurkish.Apanelofexperts,includingthree facultymembers(oneprofessorandtwoassociateprofessors) andtwoemergencyphysicianswithaspecialinterestin physicianwell-being,reviewedandrevisedthetranslationto improveitsculturalrelevance.

ToassessthecontentvalidityofthetranslatedPFI,we calculateditemandscalecontentvalidityindices(I-CVIand S-CVI).Tenindependentexpertsrateditemrelevanceona four-pointscale,with1or2indicatingnorelevanceand3or4 indicatingrelevance.WesetacceptablethresholdsatI-CVI

>0.78andS-CVI >0.90,basedonguidelinesforexpertpanels of6–10members.16

Inthepre-testphase,weobtainedfeedbackfrom fivemale and fivefemaleemergencyphysiciansworkingincrowded EDs.Basedonthisfeedback,werefinedtheinstrument, resultinginthe finalversionoftheT-PFI(AppendixA).We thendistributedthevalidatedTurkishversiontoabroader cohortofrespondentsinEDsforfurthervalidation.The InstitutionalReviewBoardoftheUniversityofHealth SciencesFatihSultanMehmetEducationandResearch Hospitalapprovedallprocedures.

Measures

ThePFI,developedbyTrockeletal,assessesprofessional fulfillmentandburnoutamongphysicians.10 Theoriginal PFIcomprises16itemscategorizedintotwoscales:the ProfessionalFulfillment(PF)scale(sixitems)andtheOverall Burnoutscale(10items),measuringthreedimensions: professionalfulfillment;professionalexhaustion,and interpersonaldisengagement.Eachitemisevaluatedusinga five-pointLikertscale(0to4).HigherscoresonthePFscale (7.5orgreater)indicatehigherprofessionalfulfillment,while scoresexceeding3.325ontheburnoutscalesuggestpotential burnout.ThePFI,validatedin2018,evaluatesemotional exhaustion,interpersonaldisengagement,andprofessional achievement,expandingbeyondtraditionalburnout dimensionstoincludeintrinsicworkcomponentssuchas happiness,meaning,self-esteem,andsatisfaction.10

DataAnalysis

Wecomputeddescriptivestatisticsfordemographic characteristicsandT-PFIscores.Meanscoresforeachitem anddimensionwerecalculatedaccordingtoTrockel’ s instructions.10 WeevaluatedthereliabilityoftheT-PFI throughinternalconsistency,whichestimatestheextentto whichtheconstituentitemsofthescaleareinterrelated.The Cronbach α coefficientwasusedtoassessinternal consistency,withvaluesabove0.70indicating acceptability.17,18 Weevaluatedconstructvaliditythrough confirmatoryfactoranalysis(CFA)ofresponsestoall16 items,aimingtoassessmodel fit.Factorloadingswere calculatedusingmaximumlikelihoodestimation.To evaluatethemodel’sgoodnessof fit,weconsideredseveral indices,includingthechi-squarestatistic(χ2),rootmean squareerrorofapproximation(RMSEA),comparative fit index(CFI),andtheTucker-Lewisindex(TLI).The thresholdsfortheseindiceswereasfollows:TLI >0.90 (acceptable), >0.95(excellent);CFI >0.90(acceptable), >0.95(excellent);RMSEA <0.08(acceptable), <0.05 (excellent);andchi-squarestatisticdividedbythedegreeof freedom(<3acceptable).19,20 Erroritemswereonly correlatediftheybelongedtothesameconstruct. Forvalidationstudies,itisrecommendedtohavea minimumof50respondentsforeachcriterionandconstruct

validationstudiesthatinvolvecalculatingcorrelation coefﬁcients.However,largersamplesizesexceeding100are preferredtoenhancetherobustnessofthe ﬁndings.21

RESULTS

Wesentthesurveyto1,434medicaldoctors,and425fully completedit,resultinginaresponserateof29.6%.Among therespondents,215(50.6%)werefemaleand210(49.4%) weremale.Oftheparticipants,193(45.4%)identiﬁedas emergencyphysicians,156(36.7%)asEMresidents,and76 (17.9%)asmedicaldoctorswithoutaspecialty.Additionally,

Table1. Characteristicsofthestudygroup.

Sociodemographicvariablesn(%)

Sex

Male215(50.6)

Female210(49.4)

Profession

Emergencyphysician193(45.4)

EMresident156(36.7)

Medicaldoctor76(17.9)

Title

Facultymembers38(9.0) Academicstaff156(36.6) Non-academicstaff231(54.4) Workexperience(years)

<283(19.5) 2–5109(25.6)

6–10111(26.1) 11–1561(14.4) >1561(14.4)

Institution

Privatehospital13(3.1)

Statehospital122(28.7)

Cityhospital45(10.6)

Universityhospital85(20.0)

Educationalandresearchhospital160(37.5)

Weeklyhoursofwork <4023(5.4) 40–72330(77.6) >7272(16.9)

Thistablepresentsthedistributionofsociodemographicvariables amongemergencyphysicianswhoparticipatedinthestudy.The dataincludesinformationonsex,professionalrole,academictitle, yearsofworkexperience,typeofinstitutionwheretheyare employed,andtheirweeklyworkinghours.Eachcategoryand subcategoryisprovidedwiththenumberofindividualsandthe correspondingpercentageofthetotalparticipantpool. ED,emergencydepartment; EM,emergencymedicine.

38participantsheldanacademictitle.Themajority(94.5%) ofrespondentsreportedworkingmorethan40hoursper week. Table1 summarizesthedescriptivesociodemographic statisticsofthestudypopulation.

WeusedTrockel’scutpointstoanalyzetheresponses, whichindicatedthatonly9.6%ofrespondentswere consideredprofessionallyfulfilled.Conversely,asignificant 79.1%ofrespondentsreportedexperiencingburnout.The contentvalidityassessment,whichemployedbothI-CVI indexvaluesandtheS-CVI/averagerelevance(Ave),yielded excellentcontentvalidity.AllitemsdemonstratedanI-CVI valueexceeding0.80,withtheS-CVI/Avereaching0.92.The meanscores(standarddeviation)fortheitemsrangedfrom 3.34(2.94)to6.47(2.62),asshownin Table2.TheCronbach α scorefortheprofessionalfulfillmentscaleandtheoverall burnoutscalewere0.87and0.90,respectively.The Cronbach α valuesforeachdimensionarepresented in Table3.

TheCFArunontheinitialmodelwiththreefactorsand16 items(Model1)demonstratedlowCFIandTLIanddidnot meetcriteriaforgoodnessof fit.Factorloadingsofallitems weregreaterthan0.40;therefore,noitemwasremovedfrom theinitialmodel.Allfactorloadingsforeachitemare presentedin Table4.IndexmodificationssuggestedbyCFA wereappliedsubsequentlytoimprovethemodel fit.After

indexmodiﬁcations,Model2wassigniﬁcantlyimproved withacceptableCFI,TLIandRMSEA(Table5, Figure).

DISCUSSION

TheT-PFIhasdemonstrateditselftobearobustand reliablemeasurementtoolamongemergencyphysiciansin Turkey.OurstudyevaluatedtheT-PFI’sreliabilityand validity,revealinghighinternalconsistency,strongfactor loadings,andCFAresultsthatalignwithinternational standards.11 ThisissigniﬁcantgiventhatpreviouscrossculturaladaptationstudiesofthePFIwereconductedin BrazilandJapan,butnotinTurkey.12,13 Ourdiverse participantgroup,consistingofphysicianswithvarious demographiccharacteristicsandprofessionalroles,enhances thegeneralizabilityofour ﬁndings.Thisdiversityprovides valuableinsightsintotheexperiencesofemergency physicians,similartostudiesconductedinother contexts.Forinstance,Asaokaetalincludedhealthcare professionalsfromdisastermedicalassistanceteamsin Japan,whileSilva-Junioretalfocusedonworkplace physiciansinBrazil.12,13

Inourstudy,weachievedaresponserateof29.6%. Variousfactors,suchasthehighworkloadofemergency physicians,mayhavecontributedtothisresponserate. AccordingtoPhillipsetal,responseratesinsurveysofhealth

Table2. Meanscoresandstandarddeviationsofthesurveytoolitems.

4.Ifeelincontrolwhendealingwithdif ﬁcultproblemsatwork4.11(2.90)

5.Myworkismeaningfultome 5.74(3.10)

6.I’mcontributingprofessionallyinthewaysIvaluemost4.94(3.10)

Overallburnout 5.26(2.18)

Burnout:professionalexhaustion 5.98(2.28)

1.AsenseofdreadwhenIthinkabouttheworkIhavetodo4.92(2.90)

2.Physicallyexhaustedatwork

6.47(2.62)

3.Lackinginenthusiasmatwork 6.42(2.92)

4.Emotionallyexhaustedatwork 6.12(3.06)

Burnout:interpersonaldisengagement 4.78(2.48)

1.Lessempatheticwithmypatients 4.74(2.89)

2.Lessempatheticwithmycolleagues 3.92(2.92)

3.Lesssensitivetoothers’ feelings/emotions 5.04(2.98)

4.Lessinterestedintalkingwithmypatients 5.43(3.05)

5.Lessconnectedwithmypatients 5.38(3.04)

6.Lessconnectedwithmycolleagues 4.20(3.10)

ThistablepresentsthemeanscoresandSDsforitemsrelatedtoprofessionalfulﬁllmentandburnoutasassessedbytheTurkishProfessional FulﬁllmentIndex.

Table3. InternalconsistencyofthedimensionoftheTurkishProfessionalFulﬁllmentIndex. Cronbach α

ScaleThisstudyOriginalJapanesestudy

Professionalfulﬁllment0.870.910.91

Overallburnout0.900.92NA Professionalexhaustion0.840.860.80

Interpersonaldisengagement0.910.920.90

ThetabledisplaystheinternalconsistencyofthedimensionsoftheTurkishProfessionalFulﬁllmentIndexinthisstudy,measuredbythe Cronbach α,incomparisontotheoriginalscaleandresultsfromtheJapanesestudy.TheresultsoftheBrazilianstudyarenotincludedinthis tablesinceonlyglobalCronbach α ofthePFI(0.95)wascitedbySilva-Junioretal.13 NA:notapplicable(TheJapanesestudydidnotprovide theCronbach’s α ofoverallburnout).

Table4. FactorloadingsforeachitemoftheTurkishProfessionalFulﬁllmentIndex.

Standardizedfactor loadings

Professionalfulﬁllment1.Ifeelhappyatwork0.890.90 2.Ifeelworthwhileatwork0.820.83

3.Myworkissatisfyingtome0.780.78

4.Ifeelincontrolwhendealingwithdifﬁcultproblemsatwork0.630.63

5.Myworkismeaningfultome0.610.58

6.IamcontributingprofessionallyinthewaysIvaluemost0.620.59

Burnout:professionalexhaustion1.AsenseofdreadwhenIthinkaboutwork,Ihavetodo0.480.45

2.Physicallyexhaustedatwork0.710.70

3.Lackinginenthusiasmatwork0.920.93

4.Emotionallyexhaustedatwork0.890.89

Burnout:interpersonaldisengagement1.Lessempatheticwithmypatients0.820.83

2.Lessempatheticwithmycolleagues0.640.65

3.Lesssensitivetoothers’ feelings/emotions0.860.86

4.Lessinterestedintalkingwithmypatients0.880.88

5.Lessconnectedwithmypatients0.900.91

6.Lessconnectedwithmycolleagues0.590.56

ThistablepresentsthestandardizedfactorloadingsforeachitemoftheTurkishProfessionalFulfillmentIndex,asmeasuredintwodifferent models(Model1andModel2).Model1:beforemodificationindex;andModel2:aftermodificationindex.

professionstraineesvarywidely,rangingfrom26.6–100%, withmulti-institutionalsurveystypicallyhavinglower responseratescomparedtosingle-institutionsurveys.22 Our study,beingmulti-institutional,alsofacedthischallenge.To improvetheresponserate,weemployedmultiplemethods, includingemails,socialmediagroups,andinternal communicationplatforms,suchasonlinemessagingsystems, andwesentreminderstoparticipants.Byemployingthese strategies,weaimedtomaximizeparticipationandreduce non-responsebias,ensuringthatour ﬁndingsaccurately reﬂecttheperspectivesofabroadrangeofhealthcare professionalswithinEM.

Basedonoursurveyresults,wefoundthatemergency physiciansinTurkeyexperiencehighlevelsofburnout.Only

9.6%ofparticipantsreportedprofessionalfulfillment,while 79.1%werelikelyexperiencingburnoutbasedonTrockel’ s cut-offpoints.This findingisconsistentwithprevious researchthathaslinkedthe fieldofEMtohighratesof burnout,attributedtothechallengingnatureofthe specialty.3–5,23–29 Forinstance,ameta-analysisonburnout prevalenceandriskfactorsamongemergencyhealthcare workersfoundthatTurkeyhasthehighestprevalenceofhigh emotionalexhaustion.25

Across-sectionalsurveystudyconductedamongfaculty membersoftheAcademicEmergencyMedicineAssociation, whichalsousedthePFIscale,yieldedintriguingresults.30 In thatstudy,38.7%ofparticipantsreportedfeelingsatisﬁed withtheiroccupation,while39.1%reportedexperiencing

Table5. Confirmatoryfactoranalysis:models’ goodnessof fitoftheTurkishProfessionalFulfillmentIndexandotherversions. #ofitems χ2/df P CFITLIRMSEAAIC

Model1165.963 <0.0010.8900.8690.108672.301

Model2162.894 <0.0010.9600.9500.067358.761

Brazilianversion163.498 <0.0010.9500.08

Japaneseversion16 <0.0010.8970.9090.085

Thistablesummarizestheresultsoftheconfirmatoryfactoranalysis(CFA)conductedtoassessthegoodnessof fitforvariousmodelsofthe TurkishProfessionalFul fillmentIndexanditscomparisonswithotherinternationalversions. χ2/df:chi-square/degreeoffreedom.

AIC,Akaikeinformationcriterion; CFI,comparative fitindex;Model1,beforemodificationindex;Model2,aftermodificationindex; RMSEA,rootmeansquareerrorofapproximation; TLI,Tucker-LewisIndex.

Figure. Con firmatoryfactoranalysisoftheTurkishProfessional FulfillmentIndex,Model2(aftermodificationindex). This figurepresentstheconfirmatoryfactoranalysisfortheTurkish ProfessionalFulfillmentIndex.Themodeldepictsthe interrelationshipsbetweenthreelatentconstructs.Theobserved variablesassociatedwiththeconstructsofprofessionalful fillment (PF),professionalexhaustion(PE),andinterpersonal disengagement(ID)arepresented.

burnout.Theseresultsindicatealowerprevalenceofburnout comparedtoourstudy.Thediscrepancymaybeattributedto thefactthatthesampleincludedsolelyEDfacultymembers, whereasourstudygroupconsistedofaheterogeneous representationofemergencyphysicians,includingresidents andattendingphysicians.Consequently,our ﬁndingsare moregeneralizableduetothediversityofparticipants, providingvaluableinsightsintotheexperiencesofclinicians inEDs.

TheinternalconsistencyoftheT-PFIwasrobust,withthe Cronbach α forprofessionalfulﬁllmentat0.87andfor overallburnoutat0.90.Thesevaluesarecomparabletothose reportedintheoriginalUSvalidationstudyandtheJapanese study.10,12 IntheBrazilianvalidationstudy,onlyaglobal Cronbach α of0.95wasreported,indicatinghighreliability acrossdifferentculturalcontexts.13

TheinitialCFAfortheT-PFIindicatedlowCFIandTLI, whichdidnotmeetthecriteriaforgoodnessof fit.After applyingindexmodifications,the fitindicessignificantly improved,withModel2achievingaCFIof0.960andan RMSEAof0.067.Theseresultsalignwiththoseofthe Japanesestudy,whichalsodemonstratedacceptable fit indicesaftermodifications.12 TheBrazilianstudysimilarly reportedaCFIof0.950andRMSEAof0.08,underscoring themodel’srobustnessacrossdifferentpopulations.13 Inthe Japanesevalidationstudy,theCFAmodel fitwasmodest, andexploratoryfactoranalysisconfirmedathree-factor structuresimilartotheoriginalscale.Ourstudy’sfactor loadingsrangedfrom0.45–0.93,confirmingtheT-PFI’ s structuralvalidity.

Althoughwedidnotassessconvergentvaliditywith externalmeasures,the findingsfromtheJapanesestudyof thePFIdemonstratedsignificantpositivecorrelations betweenprofessionalfulfillmentandqualityoflife,aswellas betweenburnoutsubscalesanddepressivesymptoms.12 This alignswithliteraturesuggestingthathigherlevelsof professionalfulfillmentareassociatedwithbetterqualityof life,whilehigherlevelsofburnoutarelinkedtoincreased depressivesymptoms.10

Volume25,No.6:November2024WesternJournal

TheuseofthePFIindifferenthealthcaresettings, includingpharmacistsandphysicians,hasdemonstratedits versatilityandreliability.31,32 Forinstance,astudyonUS pharmacistsfoundthatprofessionalfulfillmentwas associatedwithdemographicsandworksettings,with communitypharmacistsreportingthelowestfulfillmentand highestburnout.32 Thispatternissimilartoour findings, whereemergencyphysiciansexposedtohighstressandlong workinghoursreportedsignificantburnoutandlow professionalfulfillment.

Theresultsofthisstudyhavesignificantimplicationsfor bothclinicalpracticeandhealthcarepolicy.Thehigh prevalenceofburnoutamongemergencyphysiciansin Turkeyhighlightstheurgentneedforsystematic interventions.Healthcareadministratorsshouldintegrate theT-PFIintoregularevaluationstomonitorphysician well-beingproactively.Thiscouldleadtoearly identificationofburnoutsymptoms,allowingfortimely interventions.Forinstance,BodenheimerandSinsky emphasizetheimportanceofthequadrupleaim,which includesimprovingclinicians’ worklifeasacrucial componentofenhancingpatientcare.9 Additionally, Shanafeltetalfoundthataddressingphysicianburnoutcan significantlyimprovepatientcarequalityandreduce healthcarecosts.3,7

The findingsofthisstudycaninformpolicy decisionsatbothorganizationalandgovernmentallevels. Policiesdesignedtoreduceworkhours,provide mentalhealthresources,andcreateasupportivework environmentarejustifiedbythedataindicatinghighburnout levels.Futureresearchshouldinvestigatethelongitudinal impactofburnoutandprofessionalfulfillmenton careerlongevityandpatientcarequalityamong emergencyphysicians.Collaborativeeffortsacross countriescouldfurtherrefinetheT-PFI,making itaglobalstandardforassessing physicianwell-being.

LIMITATIONS

Anotablelimitationofthisstudyisitsfocusonemergency physicians,whichmaylimitthegeneralizabilityofthe findingstoothermedicalspecialtiesordepartments.Future researchshouldincludeabroaderrangeofhealthcare professionalsacrossvariousspecialtiestoenhancethe comprehensivenessofthe findings.Additionally,wedidnot comparethedemographiccharacteristicsofrespondentsand non-respondents.Futurestudiesmightbenefitfromsuch comparisonstobetteraddressnon-responsebias. Longitudinalstudiesarealsorecommendedtounderstand thetemporaldynamicsofprofessionalfulfillmentand burnout.Investigatingorganizationalfactorscouldprovide deeperinsightsintocreatinghealthierworkenvironments forphysicians.

CONCLUSION

Inthisstudywesuccessfullytranslated,adapted,and validatedtheTurkishProfessionalFulfillmentIndex, establishingitsreliabilityasatoolforassessingprofessional fulfillmentandburnoutamongmedicaldoctorsinTurkish EDs.The findingshighlightasignificantprevalenceof burnout,emphasizingtheneedfortargetedinterventionsto enhancephysicianwell-beinginTurkishhealthcaresettings. Byprovidingareliablemeasureofprofessionalfulfillment andburnout,theT-PFIcanguidehealthcareorganizations indevelopingprogramstoimprovephysicians’ job satisfactionandoverallwell-being.

AddressforCorrespondence:MerveEksioglu,MD,Departmentof EmergencyMedicine,UniversityofHealthSciences,FatihSultan MehmetEducationandResearchHospital,HastaneSokakNo:1/9, Içerenköy,Ata ¸ sehir34752Istanbul,Turkey.Atasehir-Istanbul, Turkey.Email: mervekoyunoglu@gmail.com

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.PetersDandRajasingamD.Burnoutinhealthcare:medicaleducators’ role. BMJ. 2019;366:l5570.

2.MaslachC,SchaufeliWB,LeiterMP.Jobburnout. AnnuRevPsychol. 2001;52:397–422.

3.ShanafeltTD,WestCP,SinskyC,etal.Changesinburnoutand satisfactionwithwork-lifeintegrationinphysiciansandthegeneral USworkingpopulationbetween2011and2017. MayoClinProc. 2019;94(9):1681–94.

4.AroraM,AshaS,ChinnappaJ,etal.Reviewarticle:burnoutinemergency medicinephysicians. EmergMedAustralas. 2013;25(6):491–5.

5.BoutouA,PitsiouG,SourlaE,etal.Burnoutsyndromeamong emergencymedicinephysicians:anupdateonitsprevalenceandrisk factors. EurRevMedPharmacolSci. 2019;23(20):9058–65.

6.WestCP,DyrbyeLN,ShanafeltTD.Physicianburnout:contributors, consequencesandsolutions. JInternMed. 2018;283(6):516–29.

7.HanS,ShanafeltTD,SinskyCA,etal.Estimatingtheattributablecostof physicianburnoutintheUnitedStates. AnnInternMed. 2019;170(11):784–90.

8.TawﬁkDS,ScheidA,ProﬁtJ,etal.Evidencerelatinghealthcare providerburnoutandqualityofcare:asystematicreviewandmetaanalysis. AnnInternMed. 2019;171(8):555–67.

9.BodenheimerTandSinskyC.Fromtripletoquadrupleaim:careofthe patientrequirescareoftheprovider. AnnFamMed. 2014;12(6):573–6.

10.TrockelM,BohmanB,LesureE,etal.Abriefinstrumenttoassessboth burnoutandprofessionalfulﬁllmentinphysicians:reliabilityandvalidity, includingcorrelationwithself-reportedmedicalerrors, inasampleofresidentandpracticingphysicians. AcadPsychiatry. 2018;42(1):11–24.

11.WallaceJE,LemaireJB,GhaliWA.Physicianwellness:amissing qualityindicator. Lancet. 2009;374(9702):1714–21.

12.AsaokaH,SasakiN,KoidoY,etal.Reliabilityandvalidityofthe JapaneseversionoftheProfessionalFulﬁllmentIndexamong healthcareprofessionals:avalidationstudy. JOccupHealth. 2023;65(1):e12422.

13.Silva-JuniorJS,NiitumaOgataAJ,ScarpelliniBC,etal.Validity evidenceofBrazilianPortugueseversionoftheProfessionalFulﬁllment Index. AmJHealthPromot. 2022;36(7):1112–22.

14.PhillipsAW,ReddyS,DurningSJ.Improvingresponseratesand evaluatingnonresponsebiasinsurveys:AMEEguideno.102. Med Teach. 2016;38(3):217–28.

15.WorldHealthOrganization. ProcessofTranslationandAdaptationof Instruments.Geneva,Switzerland:WorldHealthOrganization,2005.

16.PolitDFandBeckCT.Thecontentvalidityindex:areyousureyouknow what’sbeingreported?Critiqueandrecommendations. ResNurs Health. 2006;29(5):489–97.

17.BlandJMandAltmanDG.Cronbach’salpha. BMJ. 1997;314(7080):572.

18.TavakolMandDennickR.MakingsenseofCronbach’salpha. IntJMed Educ. 2011;2:53–5.

19.XiaYandYangY.RMSEA,CFI,andTLIinstructuralequationmodeling withorderedcategoricaldata:thestorytheytelldependsonthe estimationmethods. BehavResMethods. 2019;51(1):409–28.

20.RoseM,BjornerJB,BeckerJ,etal.Evaluationofapreliminaryphysical functionitembanksupportedtheexpectedadvantagesofthePatientReportedOutcomesMeasurementInformationSystem(PROMIS). JClinEpidemiol. 2008;61(1):17–33.

21.deVetHCW,TerweeCB,MokkinkLB,etal.(2011). Measurementin Medicine:APracticalGuide.PracticalGuidestoBiostatisticsand Epidemiology.Cambridge,UK:CambridgeUniversityPress,2011.

22.PhillipsAW,FriedmanBT,UtrankarA,etal.Surveysofhealth professionstrainees:prevalence,responserates,andpredictivefactors toguideresearchers. AcadMed. 2017;92(2):222–8.

23.ZhangQ,MuMC,HeY,etal.Burnoutinemergencymedicine physicians:ameta-analysisandsystematicreview. Medicine (Baltimore). 2020;99(32):e21462.

24.SakamotoJT,LeeJ,LuDW,etal.Factorsdrivingburnoutand professionalfulﬁllmentamongemergencymedicineresidents:a nationalwellnesssurveybytheSocietyforAcademicEmergency Medicine. AEMEducTrain. 2022;6(Suppl1):S5–12.

25.AlanazyARMandAlruwailiA.Theglobalprevalenceandassociated factorsofburnoutamongemergencydepartmenthealthcareworkers andtheimpactoftheCOVID-19pandemic:asystematicreviewand meta-analysis. Healthcare(Basel). 2023;11(15):2220.

26.MoukarzelA,MicheletP,DurandAC,etal.Burnoutsyndromeamong emergencydepartmentstaff:prevalenceandassociatedfactors. BiomedResInt. 2019;2019:6462472.

27.KimoTakayesuJ,RamoskaEA,ClarkTR,etal.Factorsassociatedwith burnoutduringemergencymedicineresidency. AcadEmergMed. 2014;21(9):1031–5.

28.StehmanCR,TestoZ,GershawRS,etal.Burnout,dropout,suicide: physicianlossinemergencymedicine,part1. WestJEmergMed. 2019;20(3):485–94.[publishedcorrectionappearsin WestJEmerg Med.2019;20(5):840–1].

29.SonmezLOandGulM.Occupationalburnout,jobsatisfactionand anxietyamongemergencymedicinedoctorsinTurkey. PakJMedSci. 2021;37(3):757–63.

30.LuDW,LeeJ,AlvarezA,etal.Driversofprofessionalfulﬁllmentand burnoutamongemergencymedicinefaculty:anationalwellnesssurvey bytheSocietyforAcademicEmergencyMedicine. AcadEmergMed. 2022;29(8):987–98.

31.GustafsonKA,BatesB,ClineKM,etal.Concurrentvalidityofthe ProfessionalFulﬁllmentIndexinasampleofpharmacyresidentsand preceptors. AmJHealthSystPharm. 2023;80(11):678–86.

32.FadareOO,DoucetteWR,GaitherCA,etal.UseoftheProfessional FulﬁllmentIndexinpharmacists:aconﬁrmatoryfactoranalysis. Pharmacy(Basel). 2021;9(4):158.

ORIGINAL RESEARCH

UseofParenteralAntibioticsinEmergencyDepartments:Practice PatternsandClassConcordance

MeganElli,MS

TimothyMolinarolo,PharmD*†

AidanMullan,MA‡

LauraWalker,MD,MBA*

SectionEditor:ShahramLotﬁpour,MD,MPH

*MayoClinic,DepartmentofEmergencyMedicine,Rochester,Minnesota † MayoClinic,DepartmentofPharmacy,Rochester,Minnesota ‡ MayoClinic,DepartmentofQuantitativeHealthSciences,Rochester,Minnesota

Submissionhistory:SubmittedApril9,2023;RevisionreceivedMay29,2024;AcceptedJune11,2024

ElectronicallypublishedSeptember6,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.17998

Introduction: Weaimedtoassessantibioticstewardshipbyquantifyingtheuseof ﬁrst-doseintravenous (IV)vsoral-onlyantibioticsandthefrequencywithwhichantibioticclasswaschangedfordischarged patients.Secondaryaimsincludedthefollowing:evaluationoftherelativelengthofstay(LOS); differencesinprescribingpatternsbetweencliniciantypes;differencesbetweenacademicand communitysettings;assessmentofprescribingpatternsamongemergencydepartment(ED)diagnoses; andfrequencyofreturnvisitsforpatientsineachgroup.

Methods: ThiswasaretrospectivecohortstudyincludingpatientspresentingtoEDswithinfectionswho weredischargedfromourMidwesthealthcaresystemconsistingof17communityhospitalsandone academiccenter.Weincludedinfectiontype,antibioticclassandrouteofadministration,typeof infection,LOS,returnvisitwithintwoweeks,cliniciantype,anddemographics.Datawerecollected betweenJune1,2018–December31,2021andanalyzedusingdescriptivestatistics.

Results: Wehad77,204EDvisitsforpatientswithinfectionsduringthestudyperiod,ofwhom3,812 receivedIVantibioticsduringtheirvisit.Thereweremorewomen(62.4%)thanmenincluded.Ofthe 3,812patientswhoreceivedIVantibiotics,1,026(34.3%)weredischargedonadifferentclassof antibioticsthantheyreceived.ThemostcommonchangeswerefromIVcephalosporintooralquinolone orpenicillin.PatientstreatedwithIVantibioticspriortodischargehadalongerLOSintheED(median differenceof102minuteslongerforthosewhoreceivedIVantibiotics).Therewasnotasigniﬁcant differenceintheuseofIVantibioticsbetweentheacademiccenterandcommunitysitesincludedin thestudy.

Conclusion: AdministeringIVantibioticsasa ﬁrstdosepriortooralprescriptionsupondischargeis common,asisshiftingclassesfromtheIVdosetotheoralprescription.Thisoffersanopportunityfor interventiontoimproveantibioticstewardshipforEDpatientsaswellasreducecostandlengthofstay. [WestJEmergMed.2024;25(6)966–974.]

INTRODUCTION

Background

Acuteinfectionsareacommonreasonforpatientsto presenttotheemergencydepartment(ED).Therewere approximately130millionEDvisitsin2018withinthe UnitedStates.Acuteinfectionsaccountforapproximately 15millionvisitstotheEDannually.1,2 Acuterespiratory

infections,skininfections(eg,cellulitisandsubcutaneous abscess),andurinarytractinfections(UTI)areamongthe mostcommoninfectionsevaluatedintheEDin recentyears.3

Priorworkhasaddressedinpatientconcernsabout transitioningfromintravenous(IV)tooralantibiotics,some ofwhichareapplicabletotheEDsetting.Theseinclude

concernsfornoclearbeneﬁt;reducedbioavailability; inapplicabilityofbioavailabilitystudiesperformedon healthyindividuals;improvedclinicaloutcomeswithIV therapy;andconcernthatanoralrouteofadministration maycontributetoaninfectionlingering.Theseconcernsare systematicallyaddressedandincludecommentaryonthe bioavailabilityofmostantibiotics,whichisgreaterthan90%. Manyoftheclassescanachieveaserumconcentrationthatis thesameviaoralorIVroutes.TreatmentwithIVantibiotics comparedwithoralantibioticshasbeennotedtobemore expensivethanoralantibioticsalone,andIVantibioticsalso havetheriskofpossiblecomplicationsassociatedwith IV-lineinsertionanduse.4 Cephalosporinsareanotable exceptionandhavebeenstudiedinthesettingof pyelonephritiswithrecommendationsforasingleIVdosein theEDpriortodismissal(Guptaetal,below).However,the serumconcentrationachievableviaoraladministrationis adequatetotreatmildandmoderateinfections,5 aswewould expecttoseeinapatientwhoisdischargedhome.

Importance

Antibioticstewardshipisincreasinglyimportantdueto therisingratesofantibioticresistanceincludingmethicillinresistant Staphylococcusaureus andmultidrugresistant bacteriainthesettingofpneumoniaandUTIs.6–8 Thereare alsorisksofantibiotic-relatedcomplicationssuchas Clostridioidesdif fi cile colitis.Itis,therefore,imperativethat antibioticsbeadministeredfortheshortestduration required,throughanappropriateroute,andprescribedonly whennecessary.9 Thereisscantinformationregardingthe concordancebetween firstdoseofIVantibioticsand subsequentprescriptionfromtheED.Therearedifferences inthe financialburdenandtimeassociatedwiththe administrationrouteofantibiotics.Oralantibioticsaremore costeffectivethanIVantibioticsandoftenprovidesimilar microbialcoverage.4 Multiplestudieshavecomparedoral andIVantibioticswhentreatingUTIs,cellulitis,or pneumoniaseparately,butstudiesarelackingthatevaluate oralandIVantibioticsfortreatinginfectionsinabroader sense. 10–12 Inmanycases,oralantibioticsmaybe appropriateandasefficaciousasIVantibiotics,providing moretime-efficientandcost-effectivecareforpatientswho aredischargedfromtheED.

GoalsofThisInvestigation

Ourprimaryaimwastodeterminewhetherweadhereto bestpracticesinprescribingantibioticsforcommon infectiousconditionsbeingtreatedonanoutpatientbasis afteranEDvisit.Thisincludestherouteofadministration andconcordancebetweenanydosesgivenintheEDand subsequentprescriptions.Secondaryaimsincludedthe following:evaluationoftherelativelengthofstay(LOS)for patientsreceivingadoseofIVantibioticswhoare subsequentlydischargedcomparedtothosegivenoral

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Antibioticstewardshipisimperativedueto risingratesofantibiotic-relatedinfections andresistance;itiscrucialthatantibioticsbe prescribedappropriately.

Whatwastheresearchquestion?

Wequanti ﬁ edtheuseof ﬁ rst-doseIVvsoralonlyantibioticsandthefrequencythat antibioticclasswaschanged.

Whatwasthemajor ﬁndingofthestudy?

Of3,812patients(4.9%,95%CI4.8 – 5.1%) whoreceivedIVantibioticsintheED,1,273 (33.4%,95%CI31.9 – 34.9%)were prescribedadifferentantibioticclass whendischarged.

Howdoesthisimprovepopulationhealth?

Byrecognizinginconsistenciesinpatients treatedwithantibioticsintheEDpriorto discharge,thispresentsfutureopportunities toimproveuponantibioticstewardship.

antibioticsonly;differencesinprescribingpatternsbetween cliniciantypes(physician,physicianassistant/nurse practitioner)andinacademicvscommunitysettings; assessmentofprescribingpatternsamongEDdiagnoses(eg, skin/softtissue,urinary,pulmonary);andpatternsofreturn visitsforpatientsreceivingadoseofparenteralvsoral antibioticsonly.

METHODS

StudyDesignandSetting

ThisstudyadherestotheSTROBEguidelinesfor reportingobservationalstudies.13 Thiswasamulticenter retrospectivecohortstudy.Weincludedpatientsinasingle academiccenterand17communityEDsafﬁliatedwithour institutionlocatedthroughouttheMidwest.

SelectionofParticipants

Weincludedpatientswhowereevaluatedanddischarged fromtheEDwithadiagnosisofinfection,basedon InternationalClassiﬁcationofDiseases(ICD-10)codes,and wereprescribedoralantibioticsupondischargefromtheED. SomepatientsreceivedoralantibioticsfollowingadoseofIV antibioticsprovidedpriortodischarge.Patientswhowere admittedtothehospitalorplacedinEDobservationduring their ﬁrstEDvisitwereexcludedfromouranalysis.

Measurements

Ourdatasetincludedthefollowing:visitidentifier;legal gender;genderidentify;primarylanguage;ageatvisit; antibioticsgiveninED(yes/no);allergies;EDlocation;ED arrivaltime;EDdeparturetime;chiefcomplaint;primary diagnosis;diagnosislist; finaldisposition; firstattending,last attending,resident,advancedpracticepractitioner(APP), primarynurse;returnvisitidentifier;returnvisitlocation; daysbetweenreturnvisits;returnvisitarrivaltime;return visitdeparturetime;returnvisitchiefcomplaint;returnvisit primarydiagnosis;returnvisitdisposition;antibioticorder identification;outpatientantibioticorderdate;outpatient antibioticprescription;andoutpatientantibioticprescriber andspecialty.

Wecategorizedantibioticsbytheirrouteof administration(parenteralvsoral)andbypharmacologic class(aminoglycoside,carbapenem,cephalosporin,epoxide, glycopeptide,lincomycin,macrolide,nitrofurantoin, nitroimidazole,penicillin,quinolone,sulfa,ortetracycline). Topicalisincludedamongtheclassesofantibioticsduetoits distinctuse.Antibioticswereconsideredconcordantifthe antibioticprovidedparenterallywasinthesameclassasthe oralantibioticprescribedupondischarge.Prescribing patternswereevaluatedbasedoncredentialswithsubgroups ofphysicians(MD/DO/MBBS)andadvancedpractice providers(nursepractitioner[NP]/physicianassistant[PA]). Practicesettingsweredeﬁnedasanacademiccenterthat includesanemergencymedicineresidencyprogramand community-basedsettings.TheICD-10diagnoseswere groupedbasedonorgansystemwithpresumedbacterial etiology(urinary,skin/softtissue,pulmonary, gastrointestinal,otolaryngological,animalbite,insectbite, dental,orthopedic,ophthalmologic,osteomyelitis)and/or organismtype(fungalandparasitic)andcategoriesforfever ofunknownorigin,postoperativeinfections,prophylaxis, andotherinfections,whichisacatch-allforuncommon diagnoses.AcompletelistoftheICD-10associated diagnosesincludedwithinthestudyisavailablein AppendixA.Lengthofstay(LOS)ismeasuredintime elapsedfrompresentationtotheEDuntilthetimeof discharge.Returnvisitswereconsideredpotentiallyrelated totheindexvisitforinfectioniftheyoccurredwithin twoweeks.

Outcomes

Theprimaryoutcomewastherouteofadministrationof antibioticsforadiagnosedinfectionandconcordanceof prescriptionoralantibioticswithanyparenteraltreatment given.SecondaryoutcomesincludedLOSwithintheED, differencesinprescribingIVororalantibioticsbetween physiciansandAPPs,differencesbetweenacademicand communitysetting,andassociationbetweentreatmentand EDreturnvisits.

Analysis

Wesummarizedcontinuousfeaturesweresummarized withmeanswithstandarddeviations,aswellasmedianswith interquartileranges;categoricalfeaturesweresummarized withfrequencycountsandpercentages.Wecalculated conﬁdenceintervals(CI)forpercentagesusinganexact binomialdistribution.Demographicsandvisit characteristicswerecomparedbetweenpatientswhoreceived IVantibioticsandpatientswhodidnot,usingWilcoxon rank-sumtestsandchi-squaredtests.

Forthemainoutcomesofinterest,wecomparedtherates oftreatmentwithIVantibioticsbyEDpracticeandclinician typeusingchi-squaredtests.Similarly,therateoftwo-week EDreturnswascomparedbetweenpatientstreatedwithboth IVandprescriptionantibioticsandpatientstreatedonlywith prescriptionantibiotics,usingchi-squaredtests.We comparedEDLOSbetweenpatientsreceivingIVantibiotics intheEDandpatientsnottreatedwithIVantibiotics,using Wilcoxonrank-sumtests.Testresultswerereportedwiththe mediandifferenceinLOStimesalongwith95%CIs calculatedbybootstrapresampling.Alltestsweretwo-sided and P -valueslessthan0.05wereconsideredsigniﬁcant. AnalysiswasperformedusingRversion4.1.2(RFoundation forStatisticalComputing,Vienna,Austria).14

RESULTS

CharacteristicsofStudySubjects

Patientdemographicsandvisitcharacteristicsare summarizedin Table1.Ofnote,thereweresigniﬁcantly morefemalethanmalesubjects(62.4%vs37.6%).The medianageofpatientswasdifferentbetweenpatientswho didanddidnotreceiveIVantibiotics.Therewasamedian ageof50.0yearsforthosewhodidnotreceiveIVantibiotics comparedtoamedianageof55.5yearsforthosewhodid. Themajorityofthepatientsincludedwithinthestudywere Englishspeaking(97.2%).

MainResults

Atotalof77,204EDvisitsforpatientswithinfections occurredbetweenJune1,2018–December31,2021among allsitesandwereincludedforanalysis.Therewere3,812 patients(4.9%,95%CI4.8–5.1%)whoreceivedIV antibioticsintheED.NearlyallthepatientswhoreceivedIV antibioticswithintheEDreceivedcephalosporins(3,637 patients,95.4%),withpenicillins(114patients,3.0%)and glycopeptides(32patients,0.8%)beingthenextmost common.Theprimaryinfectiousdiagnosesaresummarized within Table2.PatientsdiagnosedwithaUTIwerethe largestgrouptreatedwithIVantibiotics(63.3%).When comparingpyelonephritiswithotherUTIs,pyelonephritis wasmuchmorelikelytobetreatedwithIVantibioticsthan allotherUTIs(28.3%vs6.5%, P < 0.001).Insectbites(0.2%) andbitewounds(33of4,047visits,0.8%)receivedIV antibioticstheleastoften.

Table1. Demographicsandvisitcharacteristics.

Patientgender

Female(n = 48,157)45,600(62.1%)2,557(67.1%)

Male(n = 29,045)27,790(37.9%)1,255(32.9%)

Unknown2(0.0%)0(0.0%)

Patientage < 0.001

Mean(SD)49.2(24.5)53.4(23.6)

Median(Q1,Q3)50.0(29.0,69.0)55.5(33.0,73.0)

Primarylanguage < 0.001

English71,389(97.3%)3,686(96.7%)

Non-English1,918(2.6%)121(3.2%)

Unknown/didnotdisclose85(0.1%)5(0.1%)

EDpracticetype 0.14

Academiccenter22,591(30.8%)1,130(29.6%)

Communitypractice50,801(69.2%)2,682(70.4%)

Cliniciantype

NP/PA28,976(39.5%)1,231(32.3%)

Physician44,314(60.5%)2,578(67.7%)

EDlengthofstay < 0.001

Mean(SD)164.3(128.8)278.3(183.6)

Median(Q1,Q3)139.0(76.0,219.0)241.0(180.0,314.0)

Changeinantibioticclass

NoIVantibiotics73,392(100.0%) –

Changedantibiotics0(0.0%)1,273(33.4%)

Sameantibiotics0(0.0%)2,539(66.6%)

IV,intravenous; Abx,antibiotics; ED,emergencydepartment; Q1, ﬁrstquartile; Q3,thirdquartile; NP/PA,nursepractitioneror physicianassistant.

Table3 comparestheIVantibioticadministeredintheED tooralantibioticprescribedupondischarge.Themost commonchangesinantibioticclasswerefromIV cephalosporintooralquinolone(304visits,23.9%of changes)andfromIVcephalosporintooralpenicillin(231 visits,18.1%ofchanges).Themostcommonoralantibiotic classprescribedatdischargefromtheEDwascephalosporin. Themostcommonlyprescribedtopicalantibioticwaswithin theclassofcephalosporinsaswell.Thetypeofantibiotic prescribedtopatientstreatedparenterallywasdifferentfrom thethoseprescribedfororal-onlytreatment(P < 0.001). AmongthepatientstreatedwithIVantibiotics,1,273(33.4%, 95%CI31.9–34.9%)receivedaprescriptionforadifferent antibioticclassatdischarge.

Secondaryoutcomes

Foroursecondaryendpoints,wefoundthatpatients treatedwithIVantibioticspriortodischargehadalonger LOSwithintheED(mediandifference102minuteslonger

forthosewhoreceivedIVantibiotics;95%CI97–106 minutes; P < 0.001).Physiciansweremorelikelytotreat patientswithIVantibioticscomparedtoAPPS(5.5%vs 4.1%; P < 0.001).Whilethisisstatisticallysigniﬁcant,the overallpercentagedifferenceissmall.Therewasno signiﬁcantdifferenceintheuseofIVantibiotics betweentheacademiccenterandcommunitysites (4.8%vs5.0%; P = 0.14).

Whenweassesseddiagnosis-basedpatterns,wefoundthat amongpatientsgivenIVantibioticsduringtheEDvisit,the groupmostlikelytobeprescribedadifferentclassof antibioticwasthosewithpulmonaryinfections(279of361 visits,77.3%)followedbygastrointestinal(54of87visits, 62.1%).Therewasonepatientwithanophthalmologic infection,andtheclassofantibioticswaschangedupon dismissal(Table4).PatientstreatedwithIVantibioticsfor UTIswereleastlikelytochangeantibioticclassatdismissal (20.7%of391visits).TheIVandoralantibioticsclass administeredbasedontheinfectiontypeissummarized

Table2. Summaryofantibiotic,antifungal,antiparasitictreatmenttypeforemergencydepartmentinfections.

PrimaryEDdiagnosis

IVAbxintheED (N = 2,989)

NoIVAbxintheED (N = 56,289)

Percentofdiagnosed withIVAbxintheED

UTIsnotdiagnosedaspyelonephritis1,145(38.3%)16,591(29.5%)6.5%

Pyelonephritis747(25.0%)1,888(3.35%)28.3%

Skin/softtissueinfection540(18.1%)20,109(35.7%)2.6%

Pulmonaryinfection361(12.1%)6,708(11.9%)5.1%

Gastrointestinalinfection87(2.9%)4,211(7.5%)2.0%

ENTinfection36(1.2%)667(1.2%)5.1%

Bitewound33(1.1%)4,014(7.1%)0.8%

FUO15(0.5%)207(0.4%)6.8%

Otherinfection10(0.3%)66(0.1%)13.2%

Dentalinfection9(0.3%)454(0.8%)1.9%

Insectbite2(0.1%)1,096(1.9%)0.2%

Orthopedicinfection2(0.1%)62(0.1%)3.1%

Ophthalmologicinfection1(0.0%)68(0.1%)1.4%

Fungalinfection1(0.0%)61(0.1%)1.6%

Prophylaxis0(0.0%)52(0.1%)0%

Post-opinfection0(0.0%)22(0.0%)0%

Parasiticinfection0(0.0%)8(0.0%)0%

Osteomyelitis0(0.0%)4(0.0%)0%

IV,intravenous; Abx,antibiotic/antifungal/antiparasitic; ED,emergencydepartment; UTI,urinarytractinfection; ENT,otolaryngological; FUO,feverofunknownorigin.

Table3. Comparisonofintravenousandprescriptionantibiotics,antifungals,andantiparasiticagents.

IVantibioticclassadministeredintheED

Oralantibiotic class prescribedat discharge

Aminoglycoside (N = 2)

Carbapenem (N = 6)

Cephalosporin (N = 3,637)

Glycopeptide (N = 32) Lincomycin (N = 1) PCN (N = 114)

Sulfa (N = 1)

19)

Aminoglycoside0(0%)0(0%)1(0.0%)0(0%)0(0%)0(0%)0(0%)0(0%)

Carbapenem0(0%)3(50%)0(0%)0(0%)0(0%)0(0%)0(0%)0(0%)

Cephalosporin0(0%)0(0%)2,431(66.8%)6(19%)0(0%)4(3.5%)0(0%)3(16%)

Epoxide0(0%)1(17%)1(0.0%)0(0%)0(0%)0(0%)0(0%)0(0%)

Glycopeptide0(0%)0(0%)2(0.1%)0(0%)0(0%)0(0%)0(0%)0(0%)

Lincomycin0(0%)0(0%)25(0.7%)6(19%)1(100%)5(4.4%)0(0%)0(0%)

Macrolide0(0%)0(0%)172(4.7%)0(0%)0(0%)4(3.5%)0(0%)0(0%)

Nitrofuran1(50%)0(0%)86(2.4%)0(0%)0(0%)0(0%)0(0%)0(0%)

Nitroimidazole0(0%)0(0%)50(1.4%)0(0%)0(0%)1(0.9%)0(0%)1(5.3%)

PCN0(0%)0(0%)231(6.4%)2(6%)0(0%)88(77.2%)0(0%)0(0%)

Quinolone0(0%)2(33%)304(8.4%)3(9%)0(0%)5(1.8%)0(0%)0(0%)

Sulfa1(50%)0(0%)179(4.9%)8(25%)0(0%)2(1.8%)1(100%)0(0%)

Tetracycline0(0%)0(0%)153(4.2%)6(19%)0(0%)4(3.5%)0(0%)15(79%)

Topical0(0%)0(0%)2(0.1%)1(3%)0(0%)1(0.9%)0(0%)0(0%)

Shadedcellsindicateconcordantintravenousandprescriptionantibioticclass. IV,intravenous; ED,emergencydepartment; PCN,penicillin.

Tetracycline (N =

Table4. Changeinantibioticsandantifungalsbyinfectiondiagnosis. PrimarydiagnosisChangedantibiotics(N = 1,026)Sameantibiotics(N = 1,963)

Ophthalmologicinfection1(100%)0(0.0%)

Pulmonaryinfection279(77.3%)82(22.7%)

Gastrointestinalinfection54(62.1%)33(37.9%)

ENTinfection21(58.3%)15(41.7%)

Insectbite1(50.0%)1(50.0%)

Otherinfection5(50.0%)5(50.0%)

Orthopedicinfection1(50.0%)1(50.0%)

Skin/softtissueinfection256(47.4%)284(52.6%)

Dentalinfection4(44.4%)5(55.6%)

FUO6(40.0%)9(60.0%)

Bitewound7(21.2%)26(78.8%)

Urinaryinfection391(20.7%)1,501(79.3%)

Fungalinfection0(0.0%)1(100%)

Prophylaxis0(0.0%)0(0.0%)

Post-opinfection0(0.0%)0(0.0%)

Parasiticinfection0(0.0%)0(0.0%)

Osteomyelitis0(0.0%)0(0.0%)

1Percentagesarecalculatedrow-wise,relativetothetotalnumberofpatientswithineachprimarydiagnosisgroup. ENT,otolaryngological; FUO,feverofunknownorigin.

within Table5.Amongthe3,812patientswhoreceived parenteralantibioticsduringtheirEDvisit,749(19.6%),had areturnvisitwithintwoweekscomparedto11,601 (15.8%)of73,392patientswhoreceivedonlyoral antibiotics(P < 0.001).

DISCUSSION

OuranalysisofEDvisitsbypatientswithcommon infectiousdiseaseswhoweretreatedwithantibioticsrevealed thatthereareopportunitiesforimprovementinselectionof antibioticsintermsofadministrationrouteandhome-going prescriptionsinourhospitalsystem.Recommendationsfrom theInfectiousDiseaseSocietyofAmerica(IDSA)are availableforthethreemostcommonareasofinfection amongourpatients:urinary;skin;andpulmonarysources. TheIDSAguidelinesregardingtreatmentforUTIs recommendoraltreatmentforuncomplicatedcystitis,and whileoralantibioticsarealsoappropriateforacute pyelonephritis,thereisanoptiontoprovideaone-timeIV doseofantibiotics,suchasalong-actingcephalosporin,prior toinitiationoforaltherapy.15 Forskininfections,theIDSA guidelinesuseamild,moderate,andseveregradingfor cellulitis.Onlymildiscategorizedasappropriatefororal therapy;moderateandseverearerecommendedtoreceiveIV antibiotics.TherearemultipleappropriateoralandIV optionsfortreatmentofbitewounds.16 First-linetreatment optionsforoutpatientcommunity-acquiredpneumonia includeoralamoxicillin,macrolides,anddoxycyclinefor

patientswithfewriskfactors,andamoxicillin-clavulanatein conjunctionwithatypicalcoverage.17 Healthcare-associated pneumoniatreatmentrecommendationsofteninclude multiplemedicationstypicallyincludingarequiredIVagent, suchasvancomycin,precludingdischarge.18

PatientswhopresenttotheEDforcareareoftencomplex; clinicalassessmentofmultiplefactorsincludingclinical gestalt,inadditiontolaboratoryandimaging ﬁndings,may cueacliniciantohaveahighersuspicionforasevere infection,thuspromptingthemtoprovideIVtreatment. Additionally,therecouldbesomediagnosticuncertainty promptingadesiretoinitiateempirictreatmentpriorto attainingadeﬁnitivediagnosis.Patientsoftenimprovewhile underourcare,anditispossiblethatapatientisexpectedto beadmittedtothehospitalandprovidedIVantibioticsand eitherimprovesenoughfordismissal,orperhapstheydonot wanttobeadmitted.Thenumberofscenariosisnearly limitless.Thereisnoclearanswerastohowdecisionsare madetodeviatefromrecommendations,anditmaybean arearipeforadditionalresearchtounderstand thebasis.

WeidentiﬁedthatforUTIs,wehadthehighest concordanceratewhenanIVdoseofantibioticsis prescribed.Thisisanopportunitytoexploretherelativecost ofIVvsoraltherapies.Usingdrugs.com,wefoundthatanIV doseof1gramceftriaxonecostsapproximately$11.47,prior toreconstitution.Adoseoforalcefdinircostsunder$2.In additiontothecostofthemedications,thereareadditional

Table5. Intravenousantibioticsandantifungalsadministeredbyinfectiontype. IVantibioticsintheED

Primarydiagnosis

= 2,830)

= 103)

= 29)

(N = 17)

FUO15(100%)0(0%)0(0%)0(0%)0(0%)

Otherinfection10(100%)0(0%)0(0%)0(0%)0(0%)

Orthopedicinfection2(100%)0(0%)0(0%)0(0%)0(0%)

Insectbite2(100%)0(0.0%)0(0%)0(0%)0(0%)

Fungalinfection1(100%)0(0%)0(0%)0(0%)0(0%)

Urinaryinfection1,879(99.3%)4(0.2%)2(0.1%)0(0%)7(0.4%)

Pulmonaryinfection354(98.1%)5(1.4%)0(0%)2(0.6%)0(0%)

GIinfection74(85.1%)12(13.8%)0(0%)0(0%)1(1.1%)

Skin/softtissue

Infection

462(82.6%)38(7.0%)26(4.8%)13(2.4%)1(0.2%)

ENTinfection22(61.1%)13(36.1%)0(0%)0(0%)1(2.8%)

Dentalinfection2(22.2%)6(66.7%)1(11.1%)0(0%)0(0%)

Bitewound7(21.2%)24(72.7%)0(0%)2(6.1%)0(0%)

Ophthalmologic infection 0(0%)1(100%)0(0%)0(0%)0(0%)

1Percentagesarecalculatedrow-wise,relativetothetotalnumberofpatientswithineachprimarydiagnosisgroup. IV,intravenous; ED,emergencydepartment; GI,gastrointestinal; ENT,otolaryngological; FUO,feverofunknownorigin; Abx,antibiotics.

chargesassociatedwithIVcatheterplacementand medicationadministration.19

Useofasingle-doseglycopeptide(vancomycin)isoneof themoreproblematicexampleswithinourstudy. Unsurprisingly,wefoundthattherewerenoepisodesof concordancebetweenIVandoraladministrationfor vancomycin.Thecostforvancomycinisapproximately$200 forIVsolution.19 Inadditiontothecostforthemedication andadministration,thecostoftimeincreaseswith vancomycin,givenitslongeradministrationtimecompared tootherIVantibioticsororal-onlytherapy.

Ourstudyshowedthatthereisasignificantdifferencein LOS,whichimpactsEDthroughputandcrowding,aswellas patientqualityoflife.Wecannotattributethisdifference solelytotheprovisionofIVantibiotics,anditmaybedueto otherconfoundingfactors.However,inatimeinwhichED crowdingandpromptthroughputisamatterofpatient safety,itshouldnotbeneglected.Thereisanadditionalcost totheinstitutionfortheoccupancyofabedintheED. SchreyeretalcalculatedthepersonnelcostforasinglebedhourintheEDtobe$58.20.20 Whileover3,000patientswho receivedIVantibioticshaveanaverageLOSof100minutes greaterthanthepatientstreatedwithoralantibiotics,we find asubstantial financialimpactinadditiontoa quality-of-careeffect.20

Wefoundnodifferenceinprescribingpatternsbetween communityandacademicsettingsandonlyasmalldifference betweenphysicianandAPPs.Thismayreﬂectpractice

patternsestablishedbyinstitutionalnorms,training programsthatperpetuateasimilarculturebeingpassedon fromsupervisortotrainee,orsimplycommonpracticesin emergencymedicine.

The finaloutcomeweexaminedwasthelikelihoodtohave asecondvisitwithintwoweeksandwhethertherewasa differenceintheIV-oralvsoral-onlygroups(19.6%vs 15.8%).Weweresurprisedto findthatthepatientswho receivedIVantibioticsweremorelikelytoreturn.Thiscould havebeenrelatedtodiscordancebetweenIVantibiotics administeredwithintheEDandoralantibioticsthatpatients receivedupondischarge.Oritmayreflectamoresevere diseasethanwasappreciatedbythetreatingteam,resulting intheadministrationofIVantibiotics,orpatientswho declinedadmission.Furtherinvestigationintothecourseof thesepatientsmayshedadditionallightontheclinical decision-makingaroundmedicationadministration, prescription,andanticipatedtrajectoryoftheirillness.

Anadditional ﬁndingthatwediscoveredwasthe predominanceofwomenasrecipientsofIVantibiotics.This isconsistentwiththehigherincidenceofUTIsinwomenas comparedtomen,21 whichincombinationwiththehigh numbersofpatientswhoreceivedIVantibioticswithUTI/ pyelonephritiscouldaccountforthis ﬁnding.Giventhe highercostofcareduetoIVmedicationadministrationand longerdurationoftimespentintheED,itisimportantto considerthedisparitiesindownstreameffectsoftreatment betweengenders.

FutureOpportunities

Opportunitiesforfurtherresearchincludeinvestigating anyvariationinaveragedurationofillness,costofcare,or patientsatisfactionbetweenpatientswhoreceiveoral antibioticsalonecomparedwithpatientswhoinitiallyreceive IVantibiotics.EvaluatingthereasonsforadministeringIV antibioticsinitiallyandthereasonforchangingfromone classofIVantibioticstoanotherclassoforalantibioticsin theEDsettingisworthfurtherinquiryaswell.Identifyingthe underlyingcauseforprescribingbehaviorsthatarenot adherenttorecommendedbestpracticeswillreveal opportunitiesforeducationandintervention.Providing educationregardingoralbioavailabilityandefficacyof appropriateantibioticsmaybehelpful.Thesemayinclude educationonpharmacokinetics,implementationof electronichealthrecorddecisionsupport,processesfor prescriberandpharmacistcollaboration,andmore. Additionally,clarificationoftheIDSAguidelinesaround first-linetreatmentmayresultinimprovedLOSintheED andotherpatient-orientedoutcomes.Inparticular,theuseof anIVdoseofvancomycinpriortodismissalonotheragents isaprimeareaforinterventionwithitsassociatedcostsand durationofadministration.

LIMITATIONS

Thiswasaretrospectivecohortstudy,whichhasthe associatedlimitationsrelatedtobias.Ourstudysamplewas foundtobeskewedtowardfemalegendercomparedtothe generalpopulation.Additionally,thelargenumberof primaryEnglish-speakingpatientsmaybeanindicatorthat thisstudyisnotgeneralizabletoEDsinmorediversesettings. Ourdatawasnotabletodetecttheclinicalsigniﬁcance relatedtothereturnvisits.MorepatientswhoreceivedIV antibioticsreturnedtotheED,butitisnotclearwhetherthis wasrelatedtotheunderlyinginfection,whetherIV antibioticswereprescribedduetoaclinicaljudgmentthatthe patientappearedmoreillandwasathigherriskofdisease progression,orwhetherotherfactorsinﬂuencedthis trajectory.Nosurrogatesforpatientacuitywereincludedin ouranalysis.Inclusionofanillnessseverityscorecould improvetheabilitytounderstandthedecisiontoprovide parenteralantibiotics,aswellasinformthecontextregarding returnvisitsandprovideadditionalunderstandingofthe differenceinLOS.Whencomparingprescribingdifferences betweenphysiciansandAPPswedidnotcontrolforpractice setting,whichrangesfromaNP/PAwithindependent practiceatacriticalaccesshospitalorwithinanacademic EDandmayormaynotincludedirecton-sitesupervision. NeitherdidwecontrolforthepresenceofanED-based pharmacisttoassistwithprescribingrecommendations.

CONCLUSION

Wefoundthatpatientswithinouranalysiswhowere treatedwithintravenousantibioticsintheEDoftenreceived

adifferentclassoforalantibioticsupondischarge.Wealso foundthatadministeringIVantibioticsasa ﬁrstdosepriorto anoralantibioticbeingprescribedupondischargefromthe EDwascommonbutmaynotbenecessary.Byrecognizing theseinconsistencies,therearefutureopportunitiesto improveuponantibioticstewardshipandadherencefor prescribingoralantibioticsthatareconcordantwithIV antibioticsthatareadministered.

AddressforCorrespondence:MeganElli,MS,MayoClinic, DepartmentofEmergencyMedicine,1216SecondStreet Southwest,Rochester,MN55902,GeneroseG410. Email: Zellmer.Megan@mayo.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.CairnsC,AshmanJJ,KangK.Emergencydepartmentvisitratesby selectedcharacteristics:UnitedStates,2018. NCHSDataBrief. 2021;401:1–8.

2.SocietyforAcademicEmergencyMedicine(SAEM).Challengesin diagnosinginfectioninED.Availableat: https://www.saem.org/ education/industry-sponsored-education/challenges-in-diagnosinginfection-in-the-ed.AccessedMay22,2024.

3.CairnsC,KangK,SantoL.NationalHospitalAmbulatoryMedicalCare Survey:2018emergencydepartmentsummarytables.Availableat: https://www.cdc.gov/nchs/data/nhamcs/web_tables/ 2018-ed-web-tables-508.pdf.AccessedMay22,2024.

4.ScarboroughM,LiHK,RombachI.Oralversusintravenousantibiotics forboneandjointinfections:theOVIVAnon-inferiorityRCT. Health TechnolAssess. 2019;23(38):1–92.

5.BeiqueLandZvonarR.Addressingconcernsaboutchangingtheroute ofantimicrobialadministrationfromintravenoustooralinadult inpatients. CanJHospPharm. 2015;68(4):318–26.

6.BreyreAandFrazeeBW.Skinandsofttissueinfectionsinthe emergencydepartment. EmergMedClinNorthAm. 2018;36(4):723–50.

7.RiderACandFrazeeBW.Community-acquiredpneumonia. Emerg MedClinNorthAm. 2018;36(4):665–83.

8.KhoshnoodS,HeidaryM,MirnejadR,etal.Drug-resistantgramnegativeuropathogens:areview. BiomedPharmacother. 2017;94:982–94.

9.ShresthaJ,ZahraF,CannadyP. AntimicrobialStewardship.Treasure Island,FL:StatPearlsPublishing,2023.

10.PohX-E,WuK-H,ChenC-C,etal.Outcomesforpatientswithurinary tractinfectionafteraninitialintravenousantibioticsdose beforeemergencydepartmentdischarge. InfectDisTher. 2021;10(3):1479–89.

11.BrindleR,WilliamsOM,BartonE,etal.Assessmentofantibiotic treatmentofcellulitisanderysipelas:asystematicreviewandmetaanalysis. JAMADermatol. 2019;155(9):1033–40.

12.KimuraT,ItoM,OnozawaS.Switchingfromintravenoustooral antibioticsinhospitalizedpatientswithcommunity-acquiredpneumonia: areal-worldanalysis2010–2018. JInfectChemother. 2020;26(7):706–14.

13.vonElmE,AltmanD,EggerM.Thestrengtheningreportingof observationalstudiesinepidemiology(STROBE)statement: guidelinesforreportingobservationalstudies. AnnInternMed. 2007;147(8):573–7.

14.RCoreTeam.(2021). R:Alanguageandenvironmentforstatistical computing.Vienna,Austria:RFoundationforStatistical Computing.

15.GuptaK,HootonTM,NaberKG,etal.Internationalclinicalpractice guidelinesforthetreatmentofacuteuncomplicatedcystitisand pyelonephritisinwomen:a2010updatebytheInfectiousDiseases

SocietyofAmericaandtheEuropeanSocietyforMicrobiologyand InfectiousDiseases. ClinInfectDis. 2011;52(5):e103–20.

16.StevensDL,BisnoAL,ChambersHF,etal.Practiceguidelinesforthe diagnosisandmanagementofskinandsofttissueinfections:2014 updatebytheInfectiousDiseasesSocietyofAmerica. ClinInfectDis. 2014;59(2):147–59.

17.MetlayJP,WatererGW,LongAC,etal.Diagnosisandtreatmentof adultswithcommunity-acquiredpneumonia.Anofﬁcialclinical practiceguidelineoftheAmericanThoracicSocietyandInfectious DiseasesSocietyofAmerica. AmJRespirCritCareMed. 2019;200(7):e45–67.

18.KalilAC,MeterskyML,KlompasM,etal.Managementofadultswith hospital-acquiredandventilator-associatedpneumonia:2016 clinicalpracticeguidelinesbytheInfectiousDiseasesSocietyof AmericaandtheAmericanThoracicSociety. ClinInfectDis. 2016;63(5):e61–111.

19.Drugs.com.Availableat: http://drugs.com.AccessedMay22,2024.

20.SchreyerKEandMartinR.Theeconomicsofanadmissionsholding unit. WestJEmergMed. 2017;18(4):553–8.

21.GeerlingsSE.Clinicalpresentationsandepidemiologyofurinarytract infections. MicrobiolSpectr. 2016;4(5).

ORIGINAL RESEARCH

DiagnosticandPrognosticValueofSCUBE-1 inCOVID-19Patients

VildanOzer,MD*

OzgenGonencCekic,MD†

OzlemBulbul,MD*

DavutAydın,MD‡

EserBulut,MD§

FirdevsAksoy,MD∥

MehtapPehlivanlarKucuk,MD¶ SuleymanCanerKarahan,MD# EbruEmelSozen,MD**

EsraOzkaya,MD††

Authorscontinuedatendofpaper

SectionEditor: JosephR.Shiber,MD

*KaradenizTechnicalUniversity,SchoolofMedicine,DepartmentofEmergency Medicine,Trabzon,Türkiye

† SBUKanuniTrainingandResearchHospital,DepartmentofEmergency Medicine,Trabzon,Türkiye

‡ SBUKanuniTrainingandResearchHospital,DepartmentofChestDiseases, DivisionofIntensiveCareMedicine,Trabzon,Türkiye

§ SBUKanuniTrainingandResearchHospital,DepartmentofRadiology, Trabzon,Türkiye

∥ KaradenizTechnicalUniversity,SchoolofMedicine,DepartmentofInfectious DiseasesandClinicalMicrobiology,Trabzon,Türkiye

Af ﬁliationscontinuedatendofpaper

Submissionhistory:SubmittedNovember17,2023;RevisionreceivedAugust23,2024;AcceptedAugust26,2024

ElectronicallypublishedOctober2,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18586

Introduction: Theworkloadofphysiciansincreasedduetothenumberofpatientspresentingwith suspicionofcoronavirus2019(COVID-19)andtheprolongedwaittimesintheemergencydepartment duringtheCOVID-19pandemic.Signalpeptide-CUB-EGFdomain-containingprotein1(SCUBE-1)isa proteinpresentinplateletsandendothelialcells;itisactivatedbyinﬂammationfromCOVID-19andmay beassociatedwithCOVID-19’sknownthromboticrisk.WeaimedtodeterminewhetherSCUBE-1levels arediagnosticallycorrelatedinsuspectedCOVID-19patients,andwhetherSCUBE-1correlatedwith severityofdiseaseand,therefore,mightbeusefultoguidehospitalization/dischargedecisions.

Methods: ThesuspectedCOVID-19patientscaredforattertiaryhealthcareinstitutionsforoneyear betweenMay2021–May2022wereexaminedinthisstudy.ThesubjectswerebothsuspectedCOVID19patientsnotultimatelyfoundtohaveCOVID-19andthosewhowerediagnosedwithCOVID-19.By modifyingthediseaseseverityscoringsystemspresentinCOVID-19guidelinesin2021,theCOVID-19positivepatientgroupwasclassiﬁedasmild,moderate,severe,andcritical,andcomparedusingthe SCUBE-1levels.Moreover,SCUBE-1levelswerecomparedbetweentheCOVID-19positivegroupand theCOVID-19negativegroup.

Results: Atotalof507patientswereconsideredforthepresentstudy.Afterexcluding175patientsfor incompletedataandalternatecomorbidorganfailure.wereporton332patients(65.5%).Ofthese332 patients,80(24.0%)wereCOVID-19negative,and252(76.0%)wereCOVID-19positive.Of252(100%) patientsdiagnosedwithCOVID-19,74(29.4%)wereclassiﬁedasmild,95(37.7%)moderate,45(17.8%) severe,and38(15.1%)critical.TheSCUBE-1levelswerestatisticallydifferentbetweenCOVID-19 positive(8.48 ± 7.42nanogramspermilliliter[ng/mL])andCOVID-19negative(1.86 ± 0.92ng/mL) patients(P < 0.001).IntheCOVID-19positivegroup,SCUBE-1levelsincreasedwithdiseaseseverity (mild = 3.20 ± 1.65ng/mL,moderate = 4.78 ± 2.26ng/mL,severe = 13.68 ± 3.95ng/mL,andcritical = 21.87 ± 5.39ng/mL)(P < 0.001).TheinitialSCUBE-1levelsofdischargedpatientsweresigniﬁcantly lowerthanthoserequiringhospitalization(discharged = 2.89ng/mL[0.55–8.60ng/mL];wardadmitted = 7.13ng/mL[1.38–21.29ng/mL],andICUadmitted = 21.19ng/mL[10.58–37.86ng/mL])(P < 0.001).

Conclusion: TheSCUBE-1levelswerefoundtobedifferentiatedbetweenpatientswithandwithout COVID-19andtobecorrelatedwiththeseverityofillness.[WestJEmergMed.2024;25(6)975–984.]

INTRODUCTION

Thecoronavirus2019(COVID-19)pandemichashada hugeimpact1,2 withmorethan770millionconﬁrmedcases andmorethan6.9milliondeathsreportedworldwideasof September2023.3 Althoughhospitalandemergency department(ED)admissionratesdecreasedinthe ﬁrstperiod oftheCOVID-19pandemic,theseratesreturnedtoprepandemiclevelsovertime.Previousstudiesshowedthatthe durationoftimespentbypatientsintheEDincreasedduring thepandemicperiod.4,5 ThisledtoEDcrowding,whichin turncontributedtoincreasedinhospitalmortality.6 Forthis reason,itwouldbeusefultodeterminewhichpatientscanbe managedasoutpatients,andwhichneedadmissionandto whatlevelofcare.Itisalsocrucialtopromptlydiagnoseand providetreatmentforthispatientgrouptomanagediseaserelatedprognosisbecausethemortalityrateofpatientswho visithospitalswithCOVID-19andrequireintensivecare admissionishigh.7

Asamemberofthesignalpeptide-CUB-epidermalgrowth factordomain-containingprotein(SCUBE)genefamily, SCUBE-1isacellsurfaceglycoproteinpredominantlylocated inplateletsand,toalesserextent,inendothelialcells.The SCUBE-1isstoredinthealphagranulesofinactiveplatelets andmigratestotheplateletsurfaceafteractivationby thrombinandreleasedassmallsolubleparticlesthatare incorporatedintothethrombus.Unlikeothermembersofthe SCUBEgenefamily,SCUBE-1tendstocauseinﬂammation andthrombosisandcanbeevaluatedasaprognosticfactorin plateletactivationandthromboticdiseases.8,9 Infectionwith COVID-19predisposespatientstovenousandarterial thromboembolismsduetoexcessiveinﬂammation,hypoxia, immobilization,anddisseminatedintravascular coagulation.10 Previousstudieshaveshownthatthe thromboticcomplicationrateincreaseswithincreasing severityofthedisease.1,2,11,12

WeaimedtodeterminewhetherSCUBE-1levelsare diagnosticallycorrelatedinsuspectedCOVID-19casesand todeterminewhetherSCUBE-1correlatedwithseverityof diseaseand,therefore,mightbeusefultoguide hospitalization/dischargedecisions.

METHODS

StudyDesign

Thestudyhadaprospectiveandobservationalcohort designandrecruitedpatientswithsuspectedCOVID-19who visitedtertiaryhealthcareinstitutionsthatmanagedpatients withCOVID-19,foroneyearbetweenMay2021–May2022. ThepatientsintheCOVID-19(+)groupweresuspectedof havingCOVID-19andwerediagnosedwithCOVID-19by laboratorytests,imaging,andreal-timepolymerasechain reaction(RT-PCR)testing.TheCOVID-19( )groupwere thosepatientswhohadsymptomsbutdidnothaveCOVID19andweredischarged.Patientexclusioncriteriawereas

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Duringthepandemic,theincreaseinthe lengthoftimethatpatientsspentinthe EDandtheresultantcrowdingledto highermortality.

Whatwastheresearchquestion?

CanSCUBE-1levelsserveasadiagnostic markerinCOVID-19andbecorrelatedwith theseverityofthedisease?

Whatwasthemajor ﬁndingofthestudy?

SCUBE-1levelswerehigherinCOVID-19 positivepatients(P < 0.001)andincreased withdiseaseseverity(P < 0.001).

Howdoesthisimprovepopulationhealth?

UsingSCUBE-1asabiomarkerenables timelydiagnosisofCOVID-19andseverity assessmentwhenRT-PCRresults aredelayed.

follows:incompletedatarecords;lackofconsent;aged <18years;andconditionsthatmayalterSCUBE-1levelsdue topredispositiontothrombosis,includingpregnancy,acute renalfailure,acutemyocardialinfarction,acuteischemic cerebrovasculardisease,acutemesentericischemiaatthe timeofdiagnosis,peripheralarterialdisease,liverfailure, heartfailure,ormalignancy.

ThediseaseseverityofpatientsintheCOVID-19(+) groupwasdeterminedinJune2020usingtheclassification introducedinguidelinespublishedbytheUSNational CenterforImmunizationandRespiratoryDiseases,Division ofViralDiseases.13 Usingtheseguidelines,patientsare classifiedas,1-mildtomoderate,2-severe,or3-critical.To determinewhethertherewasadifferencebetween “mild” and “moderate,” patientsinthe “mildtomoderate” group basedontheSCUBE-1levelwereclassifiedfurtheras “mild” vs “moderate” inJuly2023byfollowingthe “COVID-19 TreatmentGuide” publishedbytheUSNationalInstitutes ofHealth.14

Asourgoalinthisstudywastodeterminewhetherthere weredifferencesintheSCUBE-1level,basedonseverityof disease,wemadeseveralmodificationstothescoringsystems fromtheguidelinesandclassifiedpatientsas1-mild, 2-moderate,3-severe,or4-critical.Accordingly,takinginto accounttheclinicalsymptomsandradiographic findingsof

thepatients,theCOVID-19(+)groupwasclassiﬁedas “mild” (individualswhohadanyofthesignsorsymptomsof COVID-19[eg,fever,cough,sorethroat,malaise,headache, musclepain,nausea,vomiting,diarrhea,lossoftasteand smell]butnoshortnessofbreath,dyspnea,orabnormalchest imaging); “moderate” (individualswhoshowedevidenceof lowerrespiratorydiseaseduringclinicalassessmentor imagingandwhohadanoxygensaturationmeasuredby pulseoximetry[SpO2]of94%orhigheronroomairatsea level); “ severe ” (individualswhohadSpO2 <94%,respiration rate >30breathsperminute,or >50%lunginvolvementon imaging);and “critical” (individualswhohadrespiratory failure,septicshock,and/ormultipleorgandysfunction).13,14

Patientswhohad “mild” disease,weretobedischargedso thattheirtreatmentwouldcontinueathome.Forthosewho had “moderate” disease,treatmentwastobeathomeorin hospital.Forthosewith “ severe ” disease,treatmentwas plannedasadmissiontotheCOVID-19ward,andforthose with “critical” disease,treatmentwasplannedasadmission totheCOVID-19intensivecareunit(ICU).Hospitallength ofstay,ICUstay,requirementformechanicalventilation, high-ﬂowoxygen,positiveinotropicsupport,andoutcomes wererecordedduringhospitalizationofthepatientsadmitted tothewardortheICU.

Forthepresentstudy,basedonG*Poweranalysis,itwas determinedthattheCOVID-19( )groupwouldhave100 participantsandtheCOVID-19(+)groupwouldhave280 patients.ThepatientsintheCOVID-19(+)groupwere dividedintofourgroupsbasedondiseaseseverity,with70in eachgroup.Beforethestudycommenced,ethicalapproval wasreceived(ApprovalNo:2021/137).

BiochemicalMeasurements

Bloodsamples

Afterconsentforparticipationinthestudywasobtained, bloodsampleswerecollectedinbiochemistrytubeswith separatorgelsandroutinebloodtests(completebloodcount, urea,creatinine,sodium,potassium,aspartate aminotransferase,alaninetransaminase,totalbilirubin, lactatedehydrogenase[LDH],creatininephosphokinase, D-dimer,ferritin,troponin,andC-reactiveprotein[CRP]) wereperformed.TomeasuretheSCUBE-1level,thetubes werecentrifugedat1800 × gfor10minutesafterclottingat roomtemperaturefor20minutes,andserumportionswere carefullytransferredto1.5-milliliter(mL)cappedtubes,and storedat 80°Cuntilanalyzed.

DeterminationoftheSCUBE-1Level

Inhumansera,theSCUBE-1levelwasdeterminedusing anenzyme-linkedimmunosorbentassay(ELISA)kit (Elabscience,Wuhan,Hubei,China;CatNo:E-EL-H5405, Lot:UPJ28DN4SW),accordingtothe manufacturer’srecommendations.

TransferringtheSamplestoELISAPlatesandPreparing forMeasurements

Serumsamplesstoredat 80°Cwerethawedatroom temperature.TheSCUBE-1standardswereprepared followingthekitprocedures,and100-microliter(μL)samples wereaddedtothewellsandtothetestserumsamples.The platewascoveredwithfoilandincubatedonashakerat37°C for90minutes.Aftertheliquidintheplatewasremoved,100 μLofbiotinylated-Ab/AgSCUBE-1solutionwasaddedto eachwell.Theplatewasthencoveredwithfoilandincubated onashakerat37°Cfor60minutes.Afterincubation,the liquidwasremoved,andtheplatewaswashedthreetimes withbufferusingaplatewasher.Then,100 μLof streptavidin-HRPsolutionwasaddedtoeachwell.After incubation,theliquidwasremoved,andtheplatewas washed ﬁvetimeswithbufferusingaplatewasher.

StainingofSamplesandMeasurements

Forstaining,90 μLofsubstratesolutionwasaddedto eachwellandincubatedfor15minutesinadark environmentat37°C.Then,50 μLofcounter-staining solutionwasaddedtoeachwell,andtransformationtothe coloryellowwasobservedineachsampleandthestandards. Theabsorbanceofthesampleswasmeasuredat450 nanometersonaVERSAmicroplatereader(Molecular Devices,LLC,SanJose,CA),andtheresultswererecorded innanograms(ng)permL.

StatisticalAnalysis

WeusedtheStatisticalPackageforSocialSciences23.0 (SPSSStatistics,IBMCorp,Armonk,NY)fordataanalysis. Categoricaldataarepresentedasnumbersandpercentages. TheKolmogorov-Smirnovtestdeterminednormalityofthe numericaldatadistributions.Datawithanormal distributionareshownasmean ± standarddeviation,and non-normallydistributeddataareshownwithmedianand quartilevalues.Weusedthechi-squaretestintheanalysisof categoricaldata.Intheanalysisofnumericaldata conformingtonormaldistribution,theStudent t -testwas usedtocomparetwogroups.Intheanalysisofnon-normally distributednumericaldata,weusedtheMann-WhitneyU testfortwo-groupcomparisonsandtheKruskal-Wallistest withBonferronicorrectionformultiple-groupcomparisons. ThediagnosticvalueoftheSCUBE-1levelwasexamined usingreceiveroperatingcurve(ROC)analysis.

RESULTS

Atotalof507patientswereconsideredforthisstudy. Afterexcluding175patients(23withincompletedata,two whowithdrew,fourwithacutemyocardialinfarction,eight withacutestroke,35withacutekidneyfailure,onewith peripheralarterydisease,twowithliverfailure,48withheart failure,48withmalignancy,two <18years,andtwo pregnantpatients),wecompletedthestudywith332patients.

Amongthese332patients,80wereCOVID-19( )and252 wereCOVID-19(+).Amongthe252patientsdiagnosedwith COVID-19,74(29.4%)wereclassiﬁedasmild,95(37.7%)as moderate,45(17.8%)assevere,and38(15.1%)ascritical. Demographicandclinicalcharacteristics,vaccinationstatus, laboratoryresults,andcomparisonsbetweenCOVID-19( ) andCOVID-19(+)patientsareshownin Table1

ThecomparisonoftheSCUBE-1levelbetweenthe COVID-19( )andCOVID-19(+)groupsispresentedin Table2.Therewasasignificantdifferenceinthediagnostic valueofSCUBE-1forCOVID-19betweentheCOVID-19 ( )andCOVID-19(+)groups(P < 0.001).TheCOVID-19 (+)patients(8.48 ± 7.42)hadahighermeanSCUBE-1level thanCOVID-19( )patients(1.86 ± 0.92)(Table2).The ROCanalysisandinitialSCUBE-1levelcut-offvaluesfor COVID-19diagnosis(areaunderthecurve[AUC]0.891, confidenceinterval[CI]0.852–0.922, P < 0.001)areshownin Figure1 and Table3.FortheCOVID-19(+)group(n = 252) theSCUBE-1levelincreasedwithanincreaseindisease severity(P < 0.001).ComparisonsbetweenSCUBE-1levels forCOVID-19(+)patients,classifiedaccordingtodisease severity,areshownin Table2 and Figure2.

Amongthe252patientsintheCOVID-19(+)group, 118wereadmittedtotheCOVID-19wardand38tothe COVID-19ICU,inlinewithcurrentguidelinesandthe protocolpublishedbytheRepublicofTürkiyeMinistryof Health.Ninety-sixpatients,whowereCOVID-19(+)anddid notrequirehospitalizationbasedontheclinicaland laboratoryexaminations,weredischarged.Thirteenofthe96 dischargedpatients(13.5%)returnedtotheEDwithin14days afterdischarge,butonlytwoofthese13werehospitalized. Thesetwopatientswereaddedtothehospitalizedpatient groupforthestatisticalanalysis.Therefore,although94 patientsweredischarged,120patientswereconsideredtobe hospitalizedbecauseofCOVID-19.ThemeanSCUBE-1level ofthedischargedpatientswassigniﬁcantlylowerthanthatof patientsrequiringhospitalization(P < 0.001).TheCOVID-19 (+)patientsweredividedintothreegroupsbasedontheir outcomes:discharged;admittedtotheward;andadmittedto theICU.TheSCUBE-1levelsandcomparisonsbypatient outcomesarepresentedin Table2.Thecut-offvaluesforthe safedischargeofpatientsweredeterminedwithROCanalysis (AUC0.868,CI0.820–0.907, P < 0.001)(Figure3).TheROC curvesandoptimalcut-offvaluesareshownin Table4 and Figure3.

DISCUSSION

TheresultsfromthisstudyshowthatthemeanSCUBE-1 levelofCOVID-19(+)patientswashigherthanforCOVID19( )patientsandthattheSCUBE-1levelincreaseswith severityofthedisease.Moreover,therewasasigniﬁcant differencebetweenpatientsrequiringhospitalization outcomes,suchasdischarged,admittedtotheward, admittedtotheICU,andtheSCUBE-1level.

Manystudieshaveexaminedthediagnosticand prognosticeffectivenessofreversetranscription-polymerase chainreaction(RT-PCR)usingimmunological, biochemical,andhematologicalparametersinpatientswith COVID-19,anddifferencesbetweenstudieshavebeen reported.AlthoughRT-PCRisknownasthegoldstandard diagnosticmethodtodiagnoseCOVID-19,ithassome drawbacks,15,16 suchasincorrectsamplecollectionandlow viralloads,whichmaycausefalsenegativeresults.15 Another studyreportedthat3%ofpatientspresentingwithCOVID19symptomshadCOVID-19-relatedtomography ﬁndings; however,whileRT-PCRtestresultsforthesepatientswere negative,serialRT-PCRtestsduringfollow-upofthese patientswerepositive.17 Consideringthattheaveragesix-day incubationperiodofthesevereacuterespiratorycoronavirus 2(SARS-CoV2),RT-PCRresultsintheearlysymptomatic periodorduringtherecoveryperiodmayyieldnegative results.18 Inbothcases,itmaybecomedifﬁculttocontrolthe infectionandpreventapandemicbecausecontagiousness continues,whichmaycausedelayedtreatmentinpatients.19 Moreover,thediagnosisrequiresalongerperiod,andserial testingisexpensive.Therefore,simplerdiagnostictestsare requiredinresource-limitedregionswhereRT-PCRcyclers andhighlytrainedtechniciansarenotemployed.

TheeffectivenessofIgMandIgGantibodiesdetectionin thediagnosisofCOVID-19hasbeeninvestigatedpreviously, anditwasshowninserology-basedteststhatsensitivity increasedasthetimefromsymptomonsetincreased,and sensitivitywasrelativelylowerbeforesevendays.20–23 Furthermore,thediagnosticvalueofbiochemicaland hematologicalbiomarkers(eg,D-dimer,CRP, procalcitonin,LDH,ferritin,lymphocytecount,and leukocytecount)wasexaminedforCOVID-19,andthese biomarkerswereshowntohavelowdiagnosticefﬁciencyand higherprognosticvaluethandiagnosticvalue.24–31 Thedata obtainedsuggeststhatlow-cost,easy-to-obtain,andeasy-tousebiomarkersthatprovideresultsinashorttimeandoffer highdiagnosticefﬁciencyarerequired.

TheSCUBE-1ishighlyexpressedinvascularendothelial cellsandplateletsandisknowntoincreaseinthrombotic diseaseswithplateletandendothelialactivation.32,33 Therefore,SCUBE-1hasbeenusedfordiagnosticor prognosticevaluationofmanythromboticdiseases.Ina previousstudy,theSCUBE-1levelinpatientsdiagnosed withpulmonaryembolismwashigherthaninthecontrol group,anditwasstatedthattheSCUBE-1levelcouldbe usedforthediagnosisofpulmonaryembolismatacut-off pointof >46ng/mLwith82%sensitivityand91% speciﬁcity.34 Similarly,XiaoetalreportedthatSCUBE-1 maybeapotentialbiomarkerforthediagnosisofpulmonary embolism.35 Cakiretaldeterminedthediagnosticvalueof SCUBE-1inaorticdissectionandreportedthatitcouldbe usedforthediagnosisofaorticdissectioninpatientswith aorticdissectionatlevels >19.75ngperdeciliterwith95%

Table1. Thedemographic,clinical,andbiochemicalcharacteristicsofCOVID-19negativeandCOVID-19positivegroups.

Variables COVID-19negative (n:80,%24.1) COVID-19positive (n:252,%75.9)

Demographics

Age,years65.0(24.0–95.0)63.0(18.0–92.0)0.73

Sex,male,n(%)36(45.0)104(41.3)0.56

Medicalhistory

Diabetesmellitus,n(%)20(25.0)55(21.8)0.55

Hypertension,n(%)37(46.3)114(45.2)0.87

CVA,n(%)7(8.8)12(4.8)0.18

CAD,n(%)6(7.5)31(12.3)0.23

Asthma/COPD,n(%)19(23.8)34(13.5)0.03

Smoking,n(%)13(16.3)38(15.1)0.80

Symptoms

Fever,n(%)18(22.5)54(21.4)0.84

Cough,n(%)28(35.0)165(65.5) <0.001

Dyspnea,n(%)33(41.3)153(60.7)0.002

Runnynose,n(%)5(6.3)7(2.8)0.15

Anorexia,n(%)9(11.3)48(19.0)0.11

Lossoftaste,n(%)0(0.0)6(2.4)0.16

Lossofsmell,n(%)0(0.0)1(0.4)0.57

Myalgia,n(%)24(30.0)93(36.9)0.26

Fatigue,n(%)38(47.5)143(56.7)0.15

Headache,n(%)19(23.8)52(20.6)0.55

Nausea/vomiting,n(%)18(22.5)45(17.9)0.36

Diarrhea,n(%)14(17.5)7(2.8) <0.001

Vaccination, n(%)51(63.7)161(63.9)0.98

Laboratoryresults Creatinine,(mg/dL)1.07

Valuesareexpressedasmean ± SD,n(%)ormedian(interquartilerange)unlessotherwisestated. CAD,coronaryarterydisease; COPD,chronicobstructivepulmonarydisease; CRP,C-reactiveprotein; CVA,cerebrovascularaccident; LDH,lactatedehydrogenase; N/L ratio,neutrophil/lymphocyteratio; PCT,procalcitonin; WBC,whitebloodcell.

Table2. TheSCUBE-1*levelsofCOVID-19negativeandCOVID-19positivegroupsaccordingtodiseaseseverityandpatientoutcomes.

COVID-19( ),(n = 80)1.86 ± 0.92 <0.001*

COVID-19(+),(n = 252)8.48 ± 7.42

COVID-19(+)(n = 252)Mild,(n = 74)3.20 ± 1.65 <0.001¥

Moderate,(n = 95)4.78 ± 2.26

Severe,(n = 45)13.68 ± 3.95

Critical,(n = 38)21.87 ± 5.39

COVID-19(+)(n = 252)Discharged,(n = 94)2.89(0.55–8.60) <0.001¥

Ward-admitted,(n = 120)7.13(1.38–21.29)

ICU-admitted,(n = 38)21.19(10.58–37.86)

*Mann-WhitneyUtest.

¥Kruskal-WallistestwithMann-Whitneycorrection,resultsachievedfromthecomparisonofthethreegroupswerestatisticallysigniﬁcant.

*SCUBE-1,signalpeptide-CUB-EGFdomain-containingprotein1.

Figure1. TheinitialSCUBE-1*levelcut-offandconﬁdence intervalvaluesforCOVID-19diagnosis.

*SCUBE-1,signalpeptide-CUB-EGFdomain-containing protein1.

AUC,areaunderthecurve; NPV,negativepredictivevalue.

sensitivityand76%speciﬁcity.36 Furthermore,Daietal concludedintheirstudythattheSCUBE-1levelmayhave diagnosticvalueinpatientswithacutecoronarysyndrome andacuteischemicstroke.37

Studiesthatexaminedthediagnosticeffectivenessof SCUBE-1werebasedontherelationshipbetweenSCUBE-1 andthrombusformationduetoendothelialdysfunction. SARS-CoV-2doesnotpossessprocoagulant

characteristics,32,38 butvascularendothelialcelldamage occursduetoanexcessiveinﬂammatoryresponsetriggered byCOVID-19.Thediagnosticrelationshipbetween SCUBE-1andCOVID-19wasdeterminedinthepresent studybecausehypercoagulability,plateletactivation,and endothelialdysfunctionmaydevelopwiththeresulting vasculopathy.34,39,40 Oneofthemainresultsofthisstudywas thattheSCUBE-1levelwashigherinCOVID-19(+)patients thaninCOVID-19( )patients.Inthiscontextitshowedthat theSCUBE-1levelisaneffectivebiomarkerforthediagnosis ofCOVID-19,anditcanbeusedtodiagnoseCOVID-19in EDs.However,thecurrentassaystudiedhereiscomplicated andlaborintensiveandwouldtakeatleast210minutesto perform,evenunderoptimumconditions.41

Clinicalsignsandsymptoms(eg,cough,dyspnea,fever, diarrhea,nausea,vomiting,lossoftasteandsmell, respiratoryrate,saturation,andradiographic ﬁ ndings) wereusedtodeterminetheseverityofCOVID-19. 12 Many biochemicalparameters(eg,elevatedCRP, thrombocytopenia,andanelevatedferritinlevel)arepoor prognosticfactorsinCOVID-19,andtheyhavenotbeen usedtode ﬁ nediseaseseverityperthecurrentliterature.42–44 Inthiscontext,determiningthediseaseseverityofpatients atthetimeofadmissionbyusingabiochemicalparameter, suchastheSCUBE-1level(withorwithoutthepresent

Table3. SCUBE-1*cut-offvalueofCOVID-19negativeandCOVID-19positivegroups.

SensitivitySpeciﬁcityPPVNPV

SCUBE-1cut-offvalue(ng/mL)(%)(95%CI)(%)(95%CI)(%)(95%CI)(%)(95%CI)

*SCUBE-1,signalpeptide-CUB-EGFdomain-containingprotein1. CI,conﬁdenceinterval; NPV,negativepredictivevalue; PPV,positivepredictivevalue.

Figure2. *SCUBE-1levelsofCOVID-19patients.All P-values <0.001. *SCUBE-1,signalpeptide-CUB-EGFdomain-containingprotein1.

Figure3. Thecut-offandconﬁdenceintervalvaluesforthesafe dischargeofCOVID-19(+)patientsaccordingtoROCanalysis. ROC,receiveroperatingcharacteristiccurve; SCUBE-1,signal peptide-CUB-EGFdomain-containingprotein1.

scoringsystems),canbeusedatanearlystagetodistinguish betweensevereandcriticalpatientswithCOVID-19to reducemortalityandenabletimelytreatment. Microvascularandmacrovascularthrombotic

complicationsmaydevelopinarterial,venous,and capillaryvascularbedsbecausethromboin ﬂ ammatory processesintensifyduringCOVID-19,particularlywith increasingseverityofthedisease.39, 45 Ina2022study conductedbyTopraketal,anelevatedSCUBE-1levelwas associatedwiththromboticcomplications,diseaseseverity, andinhospitalmortalityinpatientswithCOVID-19. 46 In thepresentstudy,theSCUBE-1levelwaselevatedin patientswithCOVID-19,andastheseverityofthedisease increasedtheSCUBE-1levelalsoincreased.

ThestudyconductedbyCaliketalreportedalow mortalityrateofpatientswhopresentedearlytohospitaland receivedearlyantiviraltreatment.47 Earlydiagnosis, appropriatetriage,andearlytreatmentofpatientswho presenttohealthcareinstitutionswithsymptomsofCOVID19andareconsideredCOVID-19(+)maypreventtheriskof contamination,reducetheneedforintensivecare,andreduce theneedforhospitalizationbyenablingrapiddecisionmakinginthebestinterestsofpatients.Fromthis perspective,biomarkersarerequiredtoguidecliniciansin

Table4. OptimalSCUBE-1*cut-offvaluesofCOVID-19positivepatientsaccordingtopatientsrequiringhospitalizationoutcomes.

Patientsrequiring hospitalizationoutcome

*SCUBE-1,signalpeptide-CUB-EGFdomain-containingprotein1. CI,conﬁdenceinterval; ICU,intensivecareunit; NPV,negativepredictivevalue; PPV,positivepredictivevalue.

hospitalization/dischargedecisionsandward/intensivecare unitadmissionofpatientsdiagnosedwithCOVID-19.48,49 Suchabiomarkermaycontributetobetterdecision-making attheEDordischargestageandEDoccupancybyreducing patientwaittimes.

Inthepresentstudy,whentheEDoutcomesofthe patientsweregroupedasdischarge,wardadmission,or intensivecareadmission,andwhentheSCUBE-1levelswere compared,theSCUBE-1levelofdischargedpatientswas lowerthanthatofpatientswhorequiredhospitalization, andtheSCUBE-1levelofpatientswhorequiredICU admissionwashigherthanfortheothergroups.Giventhese results,itcanbearguedthattheSCUBE-1levelmayassist clinicianstopredictdiseaseseverityandassistinmaking decisionsregardinghospitalizationordischarge.Inaddition, becauseoftheriskofmicro-andmacrovascularthrombosis, ahighSCUBE-1levelmeasuredintheearlystages ofthediseasemayindicatetherequirementformore intensiveantithrombotictreatmenttoprevent thromboticcomplications.

TheRT-PCRisthegoldstandardforconfirmingthe presenceofSARS-CoV2,andthetimetoobtaintheresultfor asingletestisapproximatelytwohours.50 However,samples collectedinhospitalsweretransportedtospecificlaboratories becausePCRtestscouldnotbeperformedineverylaboratory duringthepandemicperiod,51,52 whichresultedindelaysin receivingthetestresults.Previousstudieshaveshownthatthe confirmationtimeoftheSARS-CoV-2virususingRT-PCR was6–48hoursduringthepandemicperiod.53,54 Inaddition, testssuchastheRT-PCRonlyidentifySARS-CoV2anddo notprovidedataontheseverityofCOVID-19.Considering theselimitations,theuseofRT-PCRkitsforsurveillanceor screeningpatients,preventingincreasedpatientdensityin healthcareinstitutions,andreducingpatientwaittimesmaybe difficult.53 Therefore,thereisaneedfornovelbiomarkersto enabletherapiddetectionofindividualswithCOVID-19, eveninprimaryhealthcareinstitutionsandtoguidephysicians regardingthedischargeorhospitalizationofpatients accordingtocut-offvalues.

ThetesttimeofSCUBE-1isapproximately3.5hours,and thesampleiseasytoobtainfromabloodsample,which enablestherapididentiﬁcationofpatientswithCOVID-19.41 ThepresentstudyalsorevealedthattheSCUBE-1levelis associatedwiththeseverityofdisease,whichfacilitates decision-makingregardingdischargeoradmissiontothe wardortheICU,whichmayassistinreducingpatient densityinhealthcareinstitutions,reducepatientwaittimes, andeffectivelyimprovepatientmanagement.

LIMITATIONS

Thereweresomelimitationstothisstudy.First,the targetednumberofpatientswasnotrecruitedowingtothe decreasedseverityandincidenceofCOVID-19worldwide. Second,becausethenumberofSCUBE-1kitswaslimited,

SCUBE-1measurementswerelimitedtoasingleplasma sample.SerialSCUBE-1measurementsduringpatient treatmentmayhavealteredthecorrelationbetweenthe SCUBE-1levelanddiseaseseverity.

CONCLUSION

EventhoughRT-PCRtestingusuallyproducesa diagnosisofCOVID-19inashorttime,theexcessivesample loadaccumulatedinlaboratoriesduringthepandemic increasedthetimetocompletionandincreasedpatientwait times.Inthepresentstudy,wefoundthattheSCUBE-1level differsbetweenpatientswithandwithoutCOVID-19andit wascorrelatedwiththeseverityofthedisease.Accordingly, besidesguidingphysiciansregardingthediagnosisof COVID-19andtheseverityofthediseaseamongpatients whopresentathealthfacilitiesduringpandemicperiods whereresultsofRT-PCRtestsmaybedelayed,SCUBE-1 mayassistcliniciansinmanaginginﬂammatorydiseasesthat predisposetothrombosis.

AUTHORSANDAFFILIATIONSCONTINUED PolatKosucu,MD‡‡ YunusKaraca,MD* SuleymanTuredi,MD

¶KaradenizTechnicalUniversity,SchoolofMedicine, DepartmentofChestDiseases,DivisionofIntensiveCare Medicine,Trabzon,Türkiye

#KaradenizTechnicalUniversity,SchoolofMedicine, DepartmentofBiochemistry,Trabzon,Türkiye

** SBUKanuniTrainingandResearchHospital,Departmentof InfectiousDiseasesandClinicalMicrobiology, Trabzon,Türkiye

††KaradenizTechnicalUniversity,SchoolofMedicine, DepartmentofMedicalMicrobiology,Trabzon,Türkiye ‡‡KaradenizTechnicalUniversity,SchoolofMedicine, DepartmentofRadiology,Trabzon,Türkiye

AddressforCorrespondence:VildanOzer,MD,KaradenizTechnical University,SchoolofMedicine,DepartmentofEmergencyMedicine, FarabiStreetNo.64,Ortahisar,Trabzon,Türkiye61080. Email: vildan.ozer@ktu.edu.tr

REFERENCES

1.BajemaKL,OsterAM,McGovernOL,etal.Personsevaluatedfor2019 novelcoronavirus. MorbMortalWklyRep. 2020;69(6):166–70.

2.ZhuW,XieK,LuH,etal.Initialclinicalfeaturesofsuspectedcoronavirus disease2019intwoemergencydepartmentsoutsideofHubei,China. J MedVirol. 2020;92(9):1525–32.

3.WorldHealthOrganization.COVID-19weeklyepidemiological update.2023.Availableat: https://covid19.who.int/ AccessedSeptember27,2023.

4.LuceroA,SokolK,HyunJ,etal.Worseningofemergencydepartment lengthofstayduringtheCOVID-19pandemic. JAmCollEmerg PhysiciansOpen. 2021;2(3):e12489.

5.GrifﬁnG,KrizoJ,MangiraC,etal.TheimpactofCOVID-19on emergencydepartmentboardingandin-hospitalmortality. AmJEmerg Med. 2023;67:5–9.

6.JankeAT,MelnickER,VenkateshAK.Hospitaloccupancyand emergencydepartmentboardingduringtheCOVID-19pandemic. JAMANetwOpen. 2022;5(9):e2233964.

7.PetrilliCM,JonesSA,YangJ,etal.Factorsassociatedwithhospital admissionandcriticalillnessamong5279peoplewithcoronavirus disease2019inNewYorkCity:prospectivecohortstudy. BMJ. 2020;369:m1966.

8.TuCF,SuYH,HuangYN,etal.Localizationandcharacterizationofa novelsecretedproteinSCUBE1inhumanplatelets. CardiovascRes. 2006;71(3):486–95.

9.TurkmenS,MenteseS,MenteseA,etal.ThevalueofsignalpeptideCUB-EGFdomain-containingprotein1andoxidativestressparameters inthediagnosisofacutemesentericischemia. AcadEmergMed. 2013;20(3):257–64.

10.YangRB,NgCKD,WassermanSM,etal.Identiﬁcationofanovelfamily ofcell-surfaceproteinsexpressedinhumanvascularendothelium. JBiolChem. 2002;277(48):46364–73.

11.PatelA,JerniganDB,AbdirizakF,etal.Initialpublichealthresponse andinterimclinicalguidanceforthe2019novelcoronavirusoutbreakUnitedStates. MorbMortalWklyRep. 2020;69(5):140–6.

12.FanaroffACandLopesRD.COVID-19thromboticcomplicationsand therapeuticstrategies. AnnuRevMed. 2023;74:15–30.

13.NationalCenterforImmunizationandRespiratoryDiseases(U.S.). DivisionofViralDiseases.Interimclinicalguidanceformanagementof patientswithconﬁrmedcoronavirusdisease(COVID-19).2020. Availableat: https://stacks.cdc.gov/view/cdc/89980 AccessedSeptember27,2023.

14.NationalInstitutesofHealth.Coronavirusdisease2019(COVID-19) treatmentguidelines.2023.Availableat: https://www. covid19treatmentguidelines.nih.gov/.AccessedSeptember26,2023.

15.HarahwaTA,YauTHL,Lim-CookeMS,etal.Theoptimaldiagnostic methodsforCOVID-19. Diagnosis. 2020;7(4):349–56.

16.MittalR,ChourasiaN,BhartiVK,etal.Blood-basedbiomarkersfor diagnosis,prognosis,andseveritypredictionofCOVID-19: opportunitiesandchallenges. JFamilyMedPrimCare. 2022;11(8):4330–41.

17.XieX,ZhongZ,ZhaoW,etal.ChestCTfortypicalcoronavirusdisease 2019(COVID-19)pneumonia:relationshiptonegativeRT-PCRtesting. Radiology. 2020;296(2):E41–5.

18.FraterJL,ZiniG,d’OnofrioG,etal.COVID-19andtheclinical hematologylaboratory. IntJLabHematol. 2020;42(Suppl1):11–8.

19.GuneysuF,YurumezY,GucluE,etal.Thediagnosticprocessof COVID-19intheemergencydepartment:laboratoryandimaging methods. RevAssocMedBras. 2020;66(Suppl2):58–64.

20.LiuL,LiuW,ZhengY,etal.Apreliminarystudyonserological assayforsevereacuterespiratorysyndromecoronavirus2 (SARS-CoV-2)in238admittedhospitalpatients. MicrobesInfect. 2020;22(4–5):206–11.

21.GuoL,RenL,YangS,etal.Proﬁlingearlyhumoralresponseto diagnosenovelcoronavirusdisease(COVID-19). ClinInfectDis. 2020;71(15):778–85.

22.ZhaoJ,YuanQ,WangH,etal.AntibodyresponsestoSARS-CoV-2in patientswithnovelcoronavirusdisease2019. ClinInfectDis. 2020;71(16):2027–34.

23.CassanitiI,NovazziF,GiardinaF,etal.PerformanceofVivaDiag COVID-19IgM/IgGrapidtestisinadequatefordiagnosisofCOVID-19in acutepatientsreferringtoemergencyroomdepartment. JMedVirol. 2020;92(10):1724–7.

24.BattagliniD,Lopes-PachecoM,Castro-Faria-NetoHC,etal.Laboratory biomarkersfordiagnosisandprognosisinCOVID-19. FrontImmunol. 2022;13:857573.

25.IwamuraAPD,SilvaMRT,HümmelgenAL,etal.Immunityand inﬂammatorybiomarkersinCOVID-19:asystematicreview. RevMedVirol. 2021;31(4):e2199.

26.ShenY,ChengC,ZhengX,etal.Elevatedprocalcitoninispositively associatedwiththeseverityofCOVID-19:ameta-analysisbasedon10 cohortstudies. Medicina. 2021;57(6):594.

27.MesaAM,CesarEC,MontanezEM,etal.Acutelunginjurybiomarkers inthepredictionofCOVID-19severity:totalthiol,ferritinandlactate dehydrogenase. Antioxidants. 2021;10(8):1221.

28.ToppG,BouyeaM,CaggianoNC,etal.Biomarkerspredictive ofextubationandsurvivalofCOVID-19patients. Cureus. 2021;13(6):e15462.

29.BergerJS,KunichoffD,AdhikariS,etal.PrevalenceandoutcomesofDdimerelevationinhospitalizedpatientswithCOVID-19. Arterioscler ThrombVascBiol. 2020;40(10):2539–47.

30.SmilowitzNR,KunichoffD,GarshickM,etal.C-reactiveprotein andclinicaloutcomesinpatientswithCOVID-19. EurheartJ. 2021;42(23):2270–9.

31.AliA,MohamedS,ElkhidirI,etal.Theassociationoflymphocyte count,CRP,D-dimer,andLDHwithseverecoronavirus disease2019(COVID-19):ameta-analysis. SudanJMedSci. 2020;15(5):9–19.

32.WoolGDandMillerJL.TheimpactofCOVID-19diseaseonplatelets andcoagulation. Pathobiology. 2021;88(1):15–27.

33.ConnorsJMandLevyJH.COVID-19anditsimplicationsforthrombosis andanticoagulation. Blood. 2020;135(23):2033–40.

34.TurkmenS,SahinA,GunaydinM,etal.ThevalueofsignalpeptideCUB-EGFdomain-containingprotein-1(SCUBE1)inthediagnosisof pulmonaryembolism:apreliminarystudy. AcadEmergMed. 2015;22(8):922–6.

35.XiaoL,WangM,YangS,etal.Thediagnosticpotentialofplasma SCUBE-1concentrationforpulmonaryembolism:apilotstudy. ClinRespirJ. 2023;17(4):263–9.

36.CakırA,PayzaU,AksunS,etal.Validityofsignalpeptide-CUB-EGF domain-containingprotein-1(SCUBE-1)inthediagnosisofaortic dissection. SignaVitae. 2020;17(1):112–6.

37.DaiDF,ThajebP,TuCF,etal.PlasmaconcentrationofSCUBE1,a novelplateletprotein,iselevatedinpatientswithacutecoronary syndromeandischemicstroke. JACC. 2008;51(22):2173–80.

38.Gomez-MesaJE,Galindo-CoralS,MontesMC,etal.Thrombosisand coagulopathyinCOVID-19. CurrProbCardiol. 2021;46(3):100742.

39.VargaZ,FlammerAJ,SteigerP,etal.Endothelialcellinfectionand endotheliitisinCOVID-19. Lancet. 2020;395(10234):1417–8.

40.PanigadaM,BottinoN,TagliabueP,etal.HypercoagulabilityofCOVID19patientsinintensivecareunit:areportofthromboelastography ﬁndingsandotherparametersofhemostasis. JThrombHaemost. 2020;18(7):1738–42.

41.Elabscience.HumanSCUBE1(signalpeptide,CUBandEGF-like domain-containingprotein1)ELISAkit.Availableat: https://www. elabscience.com/p-human_scube1_signal_peptide_cub_and_ egf_like_domain_containing_protein_1_elisa_kit-19928.html AccessedNovember27,2023.

42.StringerD,BraudeP,MyintPK,etal.TheroleofC-reactiveproteinasa prognosticmarkerinCOVID-19. IntJEpidemiol. 2021;50(2):420–9.

43.PranataR,LimMA,YonasE,etal.Thrombocytopeniaasaprognostic markerinCOVID-19patients:diagnostictestaccuracymeta-analysis. EpidemiolInfect. 2021;149:e40.

44.ParaO,CarusoL,PestelliG,etal.Ferritinasprognosticmarkerin COVID-19:theFerVidstudy. PostgradMed. 2022;134(1):58–63.

45.ThachilJ,CushmanM,SrivastavaA.AproposalforstagingCOVID-19 coagulopathy. ResPractThrombHaemost. 2020;4(5):731–6.

46.ToprakK,KaplangorayM,PaliceA,etal.SCUBE1isassociatedwith thromboticcomplications,diseaseseverity,andin-hospitalmortalityin COVID-19patients. ThrombRes. 2022;220:100–6.

47.Ba ¸ saranNC,UyarogluOA,DizmanGT,etal.Outcomeof noncriticalCOVID-19patientswithearlyhospitalizationand earlyantiviraltreatmentoutsidetheICU. TurkJMedSci. 2021;51(2):411–20.

48.ErtekinB,Yortanlı M,OzelbaykalO,etal.Theperformanceofrisk scoringsystemsusedforsevereCOVID-19casesintheemergency department. PhnxMedJ. 2022;4(3):123–8.

49.ScheideggerD,FumeauxT,SchaffertB,etal.COVID-19pandemic: triageforintensive-caretreatmentunderresourcescarcity. SwissMed Wkly. 2020;150:w20229.

50.YuCY,ChanKG,YeanCY,etal.Nucleicacid-baseddiagnostictests forthedetectionSARS-CoV-2:anupdate. Diagnostics(Basel). 2021;11(1):53.

51.ArulebaRT,AdekiyaTA,AyaweiN,etal.COVID-19diagnosis:areview ofrapidantigen,RT-PCRandartiﬁcialintelligencemethods. Bioengineering. 2022;9(4):153.

52.YüceM,FiliztekinE,ÖzkayaKG.COVID-19diagnosis-areviewof currentmethods. BiosensBioelectron. 2021;172:112752.

53.MeiX,LeeHC,KyDiao,etal.Artiﬁcialintelligence-enabledrapid diagnosisofpatientswithCOVID-19. NatMed. 2020;26(8):1224–8.

54.ChaimayoC,KaewnaphanB,TanliengN,etal.RapidSARS-CoV-2 antigendetectionassayincomparisonwithreal-timeRT-PCRassay forlaboratorydiagnosisofCOVID-19inThailand. VirolJ. 2020;17(1):177.

ORIGINAL RESEARCH

FeasibilityofEmergencyDepartment-InitiatedHIV Pre-ExposureProphylaxis

EzraBisom-Rapp,BS*

KishanPatel,MD†‡

KatrinJaradeh,MD*†‡

TunaC.Hayirli,MD,PhD§

ChristopherR.Peabody,MD,MPH†‡

SectionEditor: IoannisKoutroulis,MD,MBA,PhD

*UniversityofCalifornia,SanFranciscoSchoolofMedicine, SanFrancisco,California

† UniversityofCaliforniaSanFrancisco,DepartmentofEmergencyMedicine, SanFrancisco,California

‡ ZuckerbergSanFranciscoGeneralHospitalandTraumaCenter, DepartmentofEmergencyMedicine,SanFrancisco,California

§ HarvardMedicalSchool,Boston,Massachusetts

Submissionhistory:SubmittedAugust13,2024;RevisionreceivedSeptember6,2024;AcceptedSeptember10,2024

ElectronicallypublishedOctober22,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.33611

Introduction: Pre-exposureprophylaxis(PrEP)forHIV usingantiretroviralmedicationinnon-infected individualstopreventHIV hasimmensepotentialtoslowthespreadofthevirus.However,uptakehas beeninsufﬁcient,andstarkracialdisparitiesexistinbothHIVacquisitionandPrEPusage,makingPrEP accessahealthequityissue.Apromisingvenuetoengagehigh-riskpopulationsinPrEPcareisthe emergencydepartment(ED);however,existingEDPrEPinitiativeshavebeencostlyorhavehadlimited success.Wehypothesizethattwostrategiescouldovercomethesebarriers:prescribingPrEPduringan EDvisitandprovidingpatientswithaninitialsupplyofPrEPmedicationintheED.Here,wedescribethe resultsofaqualitativestudyexploringmultidisciplinaryemergencycliniciansandHIVclinicians’ needs andviewsaboutthefeasibilityofsuchaninitiative.

Methods: Weconducted22semi-structuredinterviewswithmultidisciplinarycliniciansfromanurban, safety-netmedicalcenterintheEDandtheon-siteHIVclinicthatprovidesPrEPservices.Weperformed thematicanalysistosummarizechallengesandpotentialsolutionsdescribedbyparticipants.

Results: Participants’ responsesfellintothreethematiccategories:operationalchallenges;patient-level considerations;andpotentialimpacts.OperationalchallengeshighlightedthedifﬁcultyofPrEPinitiation inabusyEDandcliniciansupportneeds.Patient-levelconsiderationsincludedthecomplex psychosocialneedsofEDpatientswhocouldbeneﬁtfromPrEP.Finally,participantsanticipatedthatan ED-basedPrEPinitiationprogramcouldpositivelyimpactbothindividualpatientsandpublichealth.

Conclusion: InterviewswithemergencydepartmentandHIVclinicstaffrevealedimportant considerationsandpotentialsolutionsforED-initiatedPrEPworkflows.Cliniciansinbothspecialties wereenthusiasticaboutsuchaninitiative,whichcouldfacilitateitssuccess.Thisstudylaysthe groundworkforthefuturedesignofanefficientandinnovativeworkflowtoleveragetheEDasan essentialentrypointintoHIVpreventionservices.[WestJEmergMed.2024;25(6)985–992.]

INTRODUCTION

Despitesubstantialprogressinunderstandingand managinghumanimmunodeﬁciencyvirus(HIV),thevirus remainsapressingconcernintheUnitedStatesduetoits persistentprevalence,associatedmultisystemichealth impacts,andcoststothehealthcaresystem.1–4 This underscorestheneedforinnovativeapproachesto prevention.HIVpre-exposureprophylaxis(PrEP) theuse ofantiretroviralmedicationinnon-infectedindividualsto preventHIVacquisition hasemergedasagroundbreaking

strategythatdrasticallyreducestheriskofcontractingHIV andhasreceivedaGradeArecommendationfromtheUS PreventativeServicesTaskForce.5,6

WhilePrEPisasafeandeffectiveHIVpreventiontool, challengespersistthathavelimiteditsreal-worldimpacton theHIVepidemic.Uptakehasbeenpoor:just36%ofpeople atriskforHIVwithaPrEPindicationwereprescribedthe medicationin2022.7 Wideracialandsocioeconomic disparitiesexistinbothHIVacquisitionandPrEPuptakein theUS,8–10 leavingvulnerablepopulations disproportionatelyexposedtotheriskofHIVinfection. Addressingthesedisparitiesnecessitatesinnovative approachesthatextendbeyondtraditionalsexualhealth clinicsettings.Emergencydepartments(ED)areoftenthe onlyhealthcareaccesspointforunderservedpopulationsand havebeenidentifiedasapromisingvenuetoengagehigh-risk populationsinHIVpreventionservices.11 Researchinan urban,county-runEDestimatesthatatleast1in20ED patientsarePrEP-eligible,andofthePrEP-eligible participantswhohadpreviouslyheardofPrEP,75%were willingtostartatthatvisit.12 Otherstudieshavesimilarly demonstratedPrEP’sacceptabilityamongEDpatients.13,14 WhilemanyEDpatientscouldbenefitfromPrEP, connectingthemtolongitudinalcareatPrEPclinicsremains challenging.Astudyinahigh-volumeurbanEDshowedthat while68.6%ofpatientswhocompletedanHIVrisk assessmentwereinterestedinPrEP,11%ofthoseinterested wereabletostartPrEPmedicationafterspeakingwitha PrEPeducator,and3%ofinterestedpatientswhowere providedwithinformationaboutdrop-inPrEPclinichours receivedaprescription.14 OtherEDshavesimilarly implementedED-basedPrEPeducatorsornavigatorswith varyingdegreesofsuccess.15,16 However,thisstrategymay becost-prohibitiveformanysafety-netEDs,astheymaynot beabletofundorstaffPrEP-specificpositionstoassistwith linkages.Moreworkisneededtodeterminehowtoaddress lowPrEPinitiationrates,whileleveragingexistingresources andinterdisciplinarypersonneltoprovideequitable,highvaluecare.

Wehaveidentiﬁedtwostrategiesstudiedinothercontexts thatcouldbeadaptedtoaddressbarriersfacedbypreviously reportedED-basedPrEPprograms.First,researchindropinsexualhealthclinicsshowsthatprescribingPrEPduringa patient’sinitialvisit(“same-dayPrEPinitiation”),before receivingalllabresults,increasesthelikelihoodofPrEP initiationandcontinuationwithoutcompromisingpatient safety,comparedtothestandardmodelofrequiringmultiple visitsfortestingandcounselingbeforeprescribingPrEP medication.17 Second,providinga14-daysupplyof antiretroviralmedication “starterpacks” topatientswhotest positiveforHIVintheEDhasbeenshowntoincreasethe likelihoodofengagementinfollow-upHIVcare,compared topatientswhodidnotreceivemedicationstarterpacks.18 Wehypothesizethatadaptingbothofthesestrategiesto

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

Theemergencydepartmentisapromising venuefortheinitiationofHIVpreexposureprophylaxis(PrEP)for underservedpopulations.

Whatwastheresearchquestion?

Whatchallengesandfacilitatorsexistforthe creationofanED-basedPrEPinitiationand carelinkageprogram?

Whatwasthemajor ﬁndingofthestudy?

BusyEDswithlimitedcliniciansupportand patients ’ complexpsychosocialneedsare factorstoconsiderinthecreationofan ED-basedPrEPinitiative.

Howdoesthisimprovepopulationhealth?

PrescribingPrEPduringapatient ’ sinitial EDvisitbeforereceivi ngalllabresultscould increasethelikelihoodofPrEPinitiationand continuationinmarginalizedcommunities.

createaworkflowthatemploys “same-dayPrEPinitiation” withintheinitialEDvisitandPrEP “starterpacks” could overcomechallengesand financialbarriersfacedby previouslydescribedprogramssolelyfocusedoncounseling andreferraltoservices.However,thefeasibilityand acceptabilityofsuchaworkflowamongEDandHIVclinic staffisunknown.

Inthispaper,wepresentaqualitativestudyinvestigating thefeasibilityof,andidentifyingbarriersandfacilitatorsto, theimplementationofasame-dayPrEPinitiationworkﬂow intheEDviathematicanalysisofinterviewsconductedwith multidisciplinarycliniciansfromanurbanEDandasafetynetHIVclinicthatprovidescomprehensivePrEPservices. Thisexploratorystudylaysthegroundworkforcreationand evaluationofsuchaworkﬂow.

METHODS

StudyDesignandParticipantSelection

Weconducted22half-hourinterviewsatalarge,urban, safety-netmedicalcenterinamajorCaliforniacitywith15 multidisciplinarycliniciansintheEDandsevenintheon-site HIVclinicbetweenMay25–July12,2023andJuly15–July 24,2024.Wechosethissitebecauseitisacounty-run,safetynetinstitutionthathasbothanEDandHIVcliniconthe samecampusandservesmainlypubliclyinsuredand uninsuredpatients.Moreover,manyofthiscenter’spatients

experiencepsychosociallycomplexcircumstancessuchas houselessnessorsubstanceusedisordersandareathighrisk forHIVfromsharinginjectiondrugequipment,engagingin sexwork,orhavingcondomlessanalsexwithmultiple partners.Weidentiﬁedparticipantsbypurposivesampling, collaborativelydraftingalistofpotentialsubjectsto representadiversearrayofprofessionalexperiences, yearsofpractice,anddegreeofpastinvolvementinharm reductioninitiatives.

Weintentionallyrecruitedsubjectswhoserveinavariety ofclinicalrolestoidentifyopportunitiesforinterprofessional collaborationandstaffingefficiency.Additionally, participantswereidentifiedbysnowballsamplingbyasking participantstoreferadditionalstaffmemberswhocouldadd auniqueperspectivetothedataset.Recruitmentcontinued iterativelyinthiswayuntilthematicsaturationwasreached, meaningthatinterviewersagreedthatinterviewswere yieldingsimilardata.Participantsreceived$25giftcardsfor theirparticipation.Ourinstitution’sinstitutionalreview boarddeterminedthisstudytobeexempt.

Interventions

Interviewsweresemi-structured,usinganinterviewguide createdcollaborativelybyauthorsEBRandKJanditeratively modifiedbyallauthors(SupplementalTable1).Theinterview guideinvitedparticipantstodiscusstheirroleintheED, previousexperiencewithED-basedHIVprevention interventions,perceivedneedforHIVprevention interventionsamongtheirpatientpopulation,andanticipated barriersandsolutionsforsuccessfulPrEPinitiationand linkagetocare.Allparticipantsprovidedverbalconsentto participateinthestudy(SupplementalTable2).Besidesrole, noidentifyinginformationwascollected.Allinterviewers (EBR,KP,andKJ)haveexperiencecreatingcurriculumor workflowsformedicalprofessionalsandstudentsontopics relatedtotheneedsofmarginalizedpopulations.We consideredhowinterviewerpositionalitycouldpotentially impactparticipants’ responses;noissueswerefound,andnone oftheinterviewershadsupervisorialresponsibilitiesoverthe participantstheyinterviewed.EBRconducted14interviews; KPconductedsixinterviews,andKJconductedtwo interviews.Interviewswererecordedandtranscribedusing Zoomteleconferencingsoftwarev5.13.11(ZoomVideo Communications,Inc,SanJose,CA),andtranscriptswere reviewedforaccuracybyEBR.Recordingsandtranscripts werestoredsecurelywithinourinstitution’sapprovedsecure, enterprise,cloud-based filecollaborationsoftware.Sampling andinterviewswereconducteduntilnonewinformationwas generated,indicatingthatthematicsaturationwasreached.

Analysis

Weperformedinductivethematicanalysisofinterview transcriptsaftercompletinginterviewsfollowingthe groundedtheoryapproach.Allowingparticipants’ responses

toguidethemedevelopmentenabledustoprioritizeproblem areasnotalreadyexploredintheliterature.19 Analysiswas conductedfromJuly15–August15,2023andJuly 25–August2,2024.Attheinitialstage,authorsEBR,KP, andKJinductivelydevelopedapreliminarycodebookby reﬂectingoninterviewsandanalyzingonefulltranscript together.Thiscodebookwasiterativelyrevisedduring codingindiscussionwiththeresearchteam.EBRcoded12 interviewsandKPcoded10interviewsusingATLAS.tiweb softwarev5.11.0-2023-08-02)(ATLAS.tiGmbH,Berlin, Germany).Afterindependentlyreviewingeachother’ s codes,EBRandKPreconciledanydiscrepanciesby discussingthemwithinthecontextofthefullinterview transcript,andKJarbitrateddisagreements.Therewasa highdegreeofconcordancebetweencoders,andcomplete agreementregardingcodesandthemeswasreached. Throughoutthisprocess,higherorderthemeswere developedthroughiterativediscussionsamongtheresearch team.Frequentmeetingsateachanalyticstepenhanced consistencyofcodingandanalysis.Fundersdidnot participateinanyportionofdatacollectionoranalysis.

Inthispaper,weuse “HIVclinic” todescribethestudysite atwhichinterviewstookplace,and “PrEPclinic” toindicate anyclinicthatcouldformpartnershipswithEDstofacilitate follow-upPrEPcare.

RESULTS

Oursampleincludedeightattendingphysicians(ﬁveinthe EDandthreeintheHIVclinic);fourresidentphysiciansand fellows(threeintheEDandoneintheHIVclinic);onenurse practitionerintheED; ﬁveregisterednurses(threeintheED andtwointheHIVclinic);twosocialworkersintheED; andtwopharmacystaff(oneintheED,andoneintheHIV clinic)(Table1).

Participantsself-identifiedtheirexperiencewithdelivering PrEPas “ none, ”“limited,” or “extensive.”“Limited” PrEP experienceisdefinedasnotwithinthehealthcare professional’susualscopeofpractice,buttheyhavebeen involvedinconnectingEDpatientstoPrEPatleastonce. “Extensive” PrEPexperienceindicatesthattheyhavebeen involvedinED-prescribedPrEPmultipletimesandare comfortableinthispractice.Thirteenparticipantshadno experiencedeliveringPrEP(nineintheEDandfourinthe HIVclinic);sixhadlimitedexperience(fiveintheEDand oneintheHIVclinic);andthreehadextensiveexperience (oneintheEDandtwointheHIVclinic).Themesdiscussed byparticipantsfellintothreedistinctcategories:operational challenges;patient-levelconsiderations;andpotential impactsofanED-initiatedPrEPworkflow.

OperationalChallenges

OperationalchallengestoimplementinganED-based PrEPinitiationworkﬂowwerereportedconsistentlyby participantsfromboththeEDandtheHIVclinic(Table2).

Table1. Participantdemographicsamong22healthcare professionalsfromtheemergencydepartment(ED)andHIVclinicat ZuckerbergSanFranciscoGeneralHospital,anurbansafety-net medicalcenter.

*LimitedPrEPexperienceisdeﬁned as “notwithintheclinician’s usualscopeofpractice,buttheyhavebeeninvolvedinconnecting EDpatientstoPrEPatleastonce.”

**ExtensivePrEPexperienceisdeﬁnedashavingbeen “involved inED-prescribedPrEPmultipletimesandiscomfortablein thispractice.” ED,emergencydepartment; PrEP,pre-exposureprophylaxis.

Themainconcernsparticipantsidentiﬁedweretime, capacity,andresourceconstraintsintheED.Emergency clinicianswerewaryofinitiatingconversationsabout preventativehealthunrelatedtothepatient’schiefconcern, asdoingsocouldextendEDlengthofstayandimpedeED operations.Oneemergencyphysicianexpressedtrepidation aboutclinicians’ abilitytohave “meaningfulconversations [aboutsexualhealthandHIVrisk]whiletryingtocarefor [many]peopleinthewaitingroomatthesametime” Needingtoconductlengthysearchesforclinicalguidelines frommultiplesourceswasalsoarecurringfactor.

Tocircumventthesetimeandcapacityconstraints,nearly allparticipantsdiscussedelectronichealthrecord(EHR) toolsorelectronicallyaccessibleworkflowsasfactorsthat couldfacilitaterapidPrEPprescriptionanddecreasethe “knowledgebaseneeded[tocorrectlyprescribePrEP]” (HIV Clinicphysician).TheEHRtoolsincludedpremadeorder setstofacilitateorderingofallrequiredlabs,prescription formulations,andreferrals;achecklistoftopicstoreview withpatientsintegratedintomedicalnotetemplates;andprewrittennotesanddischargeinstructions.Integratingthese toolsintoonepageintheEHRorelectronicallyaccessible workflowsystemcouldallayconcernsoverthecognitive burdenofimplementingthisworkflowthatsomeemergency cliniciansconsideredtobeoutsidetheirroutinescope ofpractice.

Interdisciplinarycollaborationswerecommonly discussed.Theemergencyphysicianshighlightedthe importanceofinterdisciplinarycollaborationsbetween physicians,nurses,pharmacists,socialworkers,andpatient navigators, “sothattheburdenisn’tsolelyon[oneclinician] toexplaineverythingtothepatient” (emergencymedicine resident).Manyfeltthatthiscouldeasethetimeandcapacity constraintsfeltbyphysiciansintheirdailypractice.This

Table2. Representativequotesfromparticipantinterviewsillustratingoperationalchallengestoimplementinganemergencydepartmentinitiatedpre-exposureprophylaxisworkﬂow.

Sub-themeswithinoperationalchallengesAssociatedquotations

TimeandcapacityconstraintsforPrEPcare within theED

Challengesassociatedwithequitably identifyinghigh-needPrEPcandidates

Needforstaffeducation

“Thisproposedprojectiscomingata nationalcrisisofEDcrowding sothere’s reallylittleexcesscapacityanywherein,inandoutsidetheEDforadditionaltasks, withoutmoreresources.” (EDMD-01)

“We’reusuallyshortstaffed,andthere’s ::: competingprioritiesusuallyduringthe shift,andsomethinglikethiswillprobablyfalltothewaysideintermsofprioritylist, butnottosayit’snotimportant.” (EDRN-3)

“Ineed[clinicians]toofferthistopeople,butnotjustthosewho[m]youassumeare atrisk,because ::: wehavealotofpatientswhoaresexworkers,andtheydon’t necessarilytellusthey’resexworkers.” (EDNP-01)

“IthinkIwouldjustneedabitmoreinformationmyself Ihavenorealexpertisein thisarea.Andso,Iwoulddeﬁnitelywanttohaveabetterunderstandingof ::: whatI waslookingfor ::: ThenIwouldfeelprettycomfortablehavingaconversationas longas Iwaseducatedenough.” (EDSW-02)

ED, emergencydepartment; EDMD,emergencyphysician; EDRN,registerednurseintheED; EDSW,EDsocialworker; PrEP,pre-exposureprophylaxis.

sentimentwassimilarlyechoedbythenurses,pharmacists, andsocialworkersinterviewedwhofelttheirparticipation couldstrengthentheprogram.Onenursenotedthat, althoughdischargeplanningiscoordinatedbyprescribers, theseplanssometimesresult “fromtheadvocacyofbedside nurses ” (EDnurse);formalizingandencouraging interdisciplinarycollaborationcouldincreasepatient identiﬁcationandlinkage.However,anothernursenoted thatputtingtoomuchburdenonalreadybusynursingstaff wouldlimitthesuccessoftheprogram.

Participantsanticipatedchallengesinidentifyingpatients whowouldbeidealPrEPcandidates.Someworriedabout thepotential “biasofwho[clinicians]thinkisatrisk” (ED nursepractitioner),reinforcingstereotypes,orincomplete identiﬁcationofneedbasedonpatientreluctancetodisclose riskybehaviors.However,manyhadmixedfeelingsabout thesolutionstheyidentiﬁed.Whilemanyparticipants suggestedautomatedEHRpop-upreminderstriggeredby certainchiefconcernsorchartedriskfactors,they anticipatedthattheseremindersmaybeignoredbyclinicians inundatedbyaburgeoningnumberofsimilaralerts.Some suggesteduniversalscreeningforHIVriskfactorstoavoid biasedPrEPofferingbutanticipatedfeasibilityandprivacy issues.Finally,manysuggestedpostersencouragingpatients toself-identifyasPrEPcandidates,butothersfeltthis strategywouldmisshigh-needpatientswithlimitedhealth literacyorreluctancetodiscloseriskfactors.

Participantsemphasizedtheimportanceofstaff education,withmanyreportingthattheircomfortwiththe programwouldbecontingentonadequatestafftraining. Therewasnoconsensusonideallengthandmodalityof educationalsession,andsomeemphasizedtheneedfora varietyoftrainingformats livelectures,teamhuddles, emails,posters tomeettheneedsofstaffwithavariety oflearningstylesandschedules.Withregardtotraining content,emergencycliniciansdesiredmoreinformation aboutthepotentialharmthatPrEPtherapycould causepatients,particularlyrelatedtoliverandkidney health.However,HIVclinicstaffconsistentlyemphasized thatEDstafftrainingshouldhighlightthatPrEPcarries

alowriskforsevereinjuryifcliniciansadhereto prescriberguidelines.

Patient-LevelConsiderations

Inadditiontooperationalconcerns,participants frequentlyreportedthatanED-basedPrEPinitiation programmustaddressfactorsthatinfluencepatients’ ability toenterandmaintainengagementwithlongitudinalPrEP care(Table3).Theneedformechanismsforreliable connectiontofollow-upcare,includingplansforpatients whopresentafterbusinesshours,wasdiscussedubiquitously. MultipleHIVclinicparticipantsemphasizedcollecting alternateformsofcontactinformationthatmightnotusually beaskedaboutduringtriage,includingemailaddresses, friends’ orcaseworkers’ contactinformation,andcampsite locationsforunhousedpatients.ManyfromboththeEDand HIVclinicemphasizedtheimportanceofhavingaclearly definedfollow-upstructureorspecificpersontoconduct outreachtopatientswithabnormallabresultsthatresulted afterdischargefromtheED.

Patientnavigatorswerediscussedasanessentialresource tofacilitateconnectiontocare.Multipleparticipantsfrom theHIVclinicmentionedthevalueofhavinganavigator basedattheHIVclinicmeetthepatientintheED,as opposedtorelyingonED-basedsupportstaffforconnection tocare.Theyfeltthatthiscouldbuildtherapeuticrapport, decreasepatientanxietyassociatedwithseekingservicesina newhealthcaresettingas “theiceisalreadybroken” (HIV Clinicphysician),andincreaselikelihoodofretentionincare.

Duetothepsychosocialcomplexityofthetargetpatient population,participantsproposedthreecomponentsofa PrEPinitiationworkﬂowthatareessentialforsuccess.First, manyparticipantsdiscussedtheneedtoprovidepatientsan initialsupplyofPrEPintheED,asopposedtosendingthe prescriptiontoapharmacyforthepatienttopickupafter discharge.Participantsreportedhavingexperiencedgreater successindischargemedicationinitiationwhenthisstrategy wasemployedinsimilarinitiatives.Second,participants emphasizedtheimportanceofEDspartneringwithPrEP clinicsthatcouldmeetadditionalpsychosocialneeds “that

Table3. Representativequotesfromparticipantinterviewsillustratingpatient-levelconsiderationswhenimplementinganemergency department-initiatedpre-exposureprophylaxisworkﬂow.

Sub-themeswithinpatient-level considerationsAssociatedquotations

Ensuringadequatefollow-up

Anticipatedbarrierstoaccessingand adheringtomedications

“In termsofobtaininggoodcontactinformationfrompeople andIdon’tjustmeanphone numbersbythat.So,itcouldbecampsites,placestheyfrequent,friends’ contactnumbers ifsomebodydoesn’thaveareliablemeansofcommunication,becausewecertainlydon’t wanttocreatebarriersforthepeoplewhoneedPrEPthemost.” (IDMD-03)

“IdoworrythatmanypatientswhoareatriskforHIVmaynothavethefacultiestobe abletotakeadailymedication.” (EDMD-05)

EDMD, emergencyphysician; IDMD,infectiousdiseasesphysician; PrEP,pre-exposureprophylaxis.

arereallyattheforefrontof[thepatient’s]life” (HIVClinic nurse).Forexample,manyparticipantsfromboththeED andHIVclinicanticipatedthataPrEPclinicthatprovides substanceusedisordersupportandwrap-aroundservices couldbetteraddressbarrierstoadherenceinhigh-risk patientsthanaclinicthatonlyoffersbasicsexualhealthcare. Third,participantsdiscussedtheimportanceoflanguageinclusivepatienteducationmaterialsandstaffforpatients withlimitedEnglishproﬁciency,aswellaspatienteducation materialsthatusesimplelanguage.

PotentialImpacts

Participantsnearuniversallyanticipatedahighneed foranED-basedPrEPinitiationwork ﬂ ow,andthe majoritywerenoticeablyexcitedforthepotentialrolloutof suchaninitiative( Table4 ).Participantsstatedthatthis initiativecouldhaveasigni ﬁ cantimpactonpatientswho areathighriskforHIVinfectionandhavehighlevelsof psychosocialcomplexity,especiallyunhousedpatientsand thosewithsubstanceusedisorders.Furthermore, participantsanticipatedthatthisprogramcouldhave positiveimplicationsonpopulation-levelhealthdisparities inHIVacquisitionbyreachinghigh-riskgroupsthat currentHIVpreventionini tiativeshavenotbeenable toengage.

Participantshaddifferingviewsontheeffectthatthis programcouldhaveonEDstaffandoperations.BothED andPrEPclinicparticipantsworriedthatemergency clinicianswouldresisttheadditionaltasksrequiredforthis programand ﬁnditincompatiblewithanacutecaresetting. Oneemergencyattendingphysician,speakingtoherprevious experienceleadinganED-basedharmreductioninitiative,

notedthatconsiderableeducationalinterventionswere neededto “changethe[ED]cultureandmake[prescribing harmreductionmedication]somethingthatwaswithinthe purviewoftheemergencyteam.” However,manyEDstaff anticipatedthattheircolleagueswould findfulfillmentinthe opportunitytoengagepatientsinpreventativemedicineasa respitefrommanagingfrequentacutecrises.Additionally, someEDparticipantsviewedthisproposedworkflowaspart ofalargerperceivedcultureshiftinEDstoconsider upstreamfactorsthatcausepatientstoseekoutacutecare, evenfulfillingamoralobligation: “Weoweittoour communitytobeabletoprovide[PrEPintheED]” (emergencymedicineresident).

DISCUSSION

Thisstudydemonstratedthatstaffatbothabusy,urban medicalcenterEDandsafety-netHIVclinicseetheneedfor HIVpreventionservicesamongEDpatientsandare amenabletothecreationofaworkﬂowtoengageat-risk patientsincarewhileintheED,assumingthatdescribed challengesareaddressed.Weintegratedthemesthatarose intoacohesiveframeworkthatcouldbeusedtoguide workﬂowdevelopment(Figure).

Ourdataprovidesvaluableinsightintopotential interventionstoprepareandprimeEDstafftoconsiderPrEP asanoptionforpatients.Forexample,discordantviewson PrEPsafetyandsideeffectsbetweenEDandHIVclinicstaff demonstratetheneedforeducationalinterventionstargeted atthissubject.Furthermore,theperspectivesofEDstaff excitedaboutthepotentialworkflow suchas finding fulfillmentinconnectingpatientswithPrEPorevokinga moralobligationamongstaff couldbeincorporatedinto

Table4. Representativequotesillustratingparticipant-reportedpotentialimpactsofanemergencydepartment-initiatedpre-exposure prophylaxisworkﬂow.

Sub-themeswithinPotentialImpactsAssociatedquotations

Individualandpublichealthbeneﬁts ofengagingatriskpopulationsinHIVpreventionservices

Impactsonstaff

DifferingviewsontheED’sevolvingroleinthe healthcaresystem

“Alotofpeoplecometoouremergencydepartment,Isaysometimesasa lastresort,butalsoasa ﬁrstresort,becausetheydon’tknowwhereelseto go andIthinkthatitwilltouchalotmorepeoplethanIthinkwethinkitwill atthispoint.” (EDRN-01)

“Ithinkthat[clinicians]aregoingtolovebeingabletoprovidepeoplewitha wellnessactinsteadofmeetingpeopleintheirmomentofcrisis.Because that’swhatwedoallthetime.Sohowgreatisitthatwecanactuallyhelp peoplethatwanttohelpthemselvestostayhealthy.” (EDRN-01)

“Somepeopleliterallydon’tseethisastheroleoftheER,andit’sgoingto takesome,like,armtwisting.” (EDNP-01)

“Andespeciallyherethecontextof acommunityhospital, alotof peoplewhochoosetoworkhereknowthat[communityprotectionis]partof ouremergencydepartmentjob.” (EDRN-02)

“Ithinkmostpeopleinemergencymedicinerecognizethatourroleas emergencyphysiciansisconstantlyexpandingandcontractinginrelationship towhatishappeningintheworld.” (EDMD-04)

ED, emergencydepartment; EDMD,emergencyphysician; EDNP,nursepractitionerintheED; EDRN,registerednurseintheED.

Figure. Themesintegratedintoacohesiveframeworktoguidethe creationandimplementationofanemergencydepartment-based PrEPinitiationworkﬂow.

ED,emergencydepartment; EHR,electronichealthrecord; PrEP,pre-exposureprophylaxis.

interventionsdesignedtoshiftEDculturetobemore amenabletoprescribingPrEP.

Inourinterviews,EHRtoolssurfacedasapromising interventiontoaddressclinicianconcerns.Cliniciansfeltthat ordersetsandtemplatescoulddecreasetheamountoftime requiredforcliniciansto findandfollowup-to-date guidelines.Indeed,EHRtemplateshavebeenshownto improveclinicalguidelineadherenceforotherhealth issues.20,21 Electronicplatformsthatallowcliniciansto accessstandardizedworkflowsquickly,whichemergency clinicianscommonlyrequested,couldfurthersupplement thesetoolstoincreaseclinicianconfidencewith theprogram.22

Additionally,clinicians’ concernsoverequitable identiﬁcationofPrEPcandidates,aswellastrepidation aboutEHR-basedpop-upHIVriskalerts,arereﬂectedin research.Astudyconductedinalarge,managedcarehealth systemshowedthatanEHR-basedHIVriskpredictive modelbasedsolelyonPrEPguidelinesfromtheCentersfor DiseaseControlandPrevention(CDC)waslesssensitivein identifyingHIVriskinBlackpatientscomparedtoWhite patients,whereasamultivariablealgorithmgenerated throughmachinelearningwasequallysensitiveamong races. 23 ThisdemonstratesthepotentialforEHRtoolsto ameliorateracialdisparitiesinPrEPofferrate,aswellasthe needforthoughtfuldesignofthesetoolstoavoid exacerbatinghealthdisparities.

WhereaspreviousstudiesaboutPrEPintheEDfocused onlyonreferraltoservices,11,14 EDandHIVclinicstaffboth anticipatedthatimmediatePrEPinitiationwithastarter packprovidedintheEDwouldincreasechancesoffollowup.Thispredictionissupportedbyexistingliterature; immediatePrEPinitiationatthepatient’ s ﬁrstvisithasbeen showntofacilitateincreasedPrEPuptakeandpersistencein drop-insexualhealthclinicsettingsandisdescribedasatool

toaddressPrEPaccessbarriersinthe2021CDCPrEP guidelines.5,17 Thisstrategymeritsfurtherinvestigationfor adaptationtoanacutecaresetting.

Furthermore,ourdatasupportsapartnershipmodel betweentheEDandasingleoutpatientclinic.Thiscould streamlinethereferralprocessanddecreasecostsby removingtheneedforanED-basedPrEPnavigatorsor educatorstoguidepatientsthroughcomplexhealthcare systems.Direct “warmhandoffs” toPrEPclinicbasednavigatorscouldalsofacilitatetheearlydevelopment oftherapeuticrapportwithpatientsandpromote retentionincare,addressingthelow-yieldofprograms basedonprovidingresourcesheetswithPrEP clinicinformation.

Asnotedbyparticipants,theservicesofthepartnering PrEPclinicshouldmatchtheneedsofthetargetpopulation servedbytheED,includingtreatmentforsubstanceuse disorderinlocalitiesdisproportionatelyaffectedbythe opioidepidemic.Additionally,EDsmustequiptheir partneringPrEPclinicswiththeinformationneededto facilitatesuccessfulconnectiontothespeciﬁcservices providedbythatPrEPclinic.Forexample,PrEPclinicsthat performoutreachtotheunhousedmayneedinformation suchasthelocationofpatients’ encampments,whereas clinicsthatdonotoffertheseoutreachservicesmay not ﬁndthisinformationuseful.Thelocalcontextof bothservicesneededandservicesavailableshouldbetaken intoconsiderationwhendesigninganED-initiated PrEPprogram.

LIMITATIONS

Toourknowledge,thisexploratorystudyistheonlyone ofitskindtoevaluatetheneedsandviewpointsof multidisciplinarycliniciansregardinganED-initiatedPrEP workﬂow.However,ourstudydoesnotcapturethepatient perspectiveonaprograminwhichapatientwouldinitiate PrEPintheED,animportanttopicforfutureresearch. Additionally,thisstudywasconductedatalargemedical centerinacitywitharobustsocialsafetynetandmaynotbe generalizabletosmallercommunityhospitals.

CONCLUSION

Ourstudydescribesanticipatedchallengesandfacilitators ofinitiatingofapre-exposureprophylaxisworkﬂowinthe EDfromtheperspectiveofmultidisciplinaryemergencyand HIV-clinicclinicians.Theperspectivesofthe multidisciplinaryparticipantsinterviewedareessentialfor developingacomprehensive,successfulworkﬂow. Recommendationsdescribedhereprovideaframeworkfor thecreationofanovelPrEPinitiationprogram. CollaborationsbetweentheEDandpreventativemedicine programsmayhaveprofoundimplicationsforhealthequity asacutecarefacilitiesexpandtheirroleinthecommunityto

facilitateaccesstopreventativecaretothosewhohaveno othersourcefortheseservices.

LatinoMSMages21–30inSanAntonio,TX. JHIVAIDSSocServ. 2019;18(2):206–11.

10.RansomeY,BogartLM,KawachiI,etal.Area-levelHIVriskand socioeconomicfactorsassociatedwithwillingnesstousePrEPamong BlackpeopleintheU.S.South. AnnEpidemiol. 2020;42:33–41.

AddressforCorrespondence:ChristopherR.Peabody,MD,MPH, UniversityofCaliforniaSanFrancisco,DepartmentofEmergency Medicine,521ParnassusAve.,SanFrancisco,CA94143.Email: christopher.peabody@ucsf.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.ChristopherR.Peabodyisaconsultant forFujiFilm – SonoSite.Thispublicationreceivedfundingfromthe UCSFInquiryFundingOfficeMedicalStudent “SummerExplore” ResearchFellowship,securedbyEzraBisom-Rapp.Thereareno otherconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.FarnhamPGP,HoltgraveDRP,SansomSLP,etal.Medicalcosts avertedbyHIVpreventioneffortsintheUnitedStates,1991–2006. JAIDS. 2010;54(5):565–7.

2.DeAnda-DuranI,KimbroughAD,BazzanoLA.HIVandhypertension epidemiology. CurrOpinCardiol. 2023;38(4):297–303.

3.NgDK,JacobsonLP,BrownTT,etal.HIVtherapy,metabolicand cardiovascularhealthareassociatedwithglomerularhyperﬁltration amongmenwithandwithoutHIVinfection. AIDS. 2014;28(3):377–86.

4.NanniMG,CarusoR,MitchellAJ,etal.DepressioninHIVinfected patients:areview. CurrPsychiatryRep. 2015;17(1):1–11.

5.CentersforDiseaseControlandPrevention.USPublicHealthService: preexposureprophylaxisforthepreventionofHIVinfectionintheUnited States 2021update:aclinicalpracticeguideline.2021.Availableat: https://www.cdc.gov/hiv/pdf/risk/prep/cdc-hiv-prep-guidelines-2021. pdf.AccessedAugust19,2024.

6.USPreventiveServicesTaskForce.Preexposureprophylaxisforthe preventionofHIVinfection:USpreventiveservicestaskforce recommendationstatement. JAMA. 2019;321(22):2203–13.

7.FanfairRNandMerminJH.Dearcolleague:preliminarydataonpreexposureprophylaxiscoveragereleased.2023.Availableat: https:// www.cdc.gov/hiv/policies/dear-colleague/dcl/20231017.html AccessedAugust15,2024.

8.HuangYA,ZhuW,SmithDK,etal.HIVpreexposureprophylaxis,by raceandethnicity UnitedStates,2014–2016. MMWRMorbMortal WklyRep. 2018;67(41):1147–50.

9.GarciaMandSawG.Socioeconomicdisparitiesassociatedwith awareness,access,andusageofpre-exposureprophylaxisamong

11.ShullJA,AttysJM,Amutah-OnukaghaNN,etal.Utilizingemergency departmentsforpre-exposureprophylaxis(PrEP). JACEPOpen. 2020;1(6):1427–35.

12.HaukoosJS,WhiteDAE,RowanSE,etal.HIVriskandpre-exposure prophylaxiseligibilityamongemergencydepartmentpatients. AIDS PatientCareSTDS. 2021;35(6):211–9.

13.AronsonB,KulieP,CastelAD,etal.Evaluatingpatientperceptionson PrEPscreeningandreferralintheacutecaresetting. AIDSBehav. 2023;27(3):1024–9.

14.RidgwayJP,AlmirolEA,BenderA,etal.Whichpatientsinthe emergencydepartmentshouldreceivepreexposureprophylaxis? Implementationofapredictiveanalyticsapproach. AIDSPatientCare STDS. 2018;32(5):202–7.

15. MahalJ,DeccyS,SeuR.Linkingemergencydepartmentpatientsatrisk forhumanimmunodeﬁciencyvirustopre-exposureprophylaxis. AmJ EmergMed. 2022;54:87–90.

16.ZhaoZ,JonesJ,Arrington-SandersR,etal.Emergencydepartmentbasedhumanimmunodeﬁciencyviruspreexposureprophylaxisreferral program-usingemergencydepartmentsasaportalforpreexposure prophylaxisservices. SexTransmDis. 2021;48(8):E102–4.

17.KamisKF,MarxGE,ScottKA,etal.Same-dayHIVpre-exposure prophylaxis(PrEP)initiationduringdrop-insexuallytransmitted diseasesclinicappointmentsisahighlyacceptable,feasible,andsafe modelthatengagesindividualsatriskforHIVintoPrEPcare. Open ForumInfectDis. 2019;6(7):ofz310.

18.WhiteDAE,JewettM,BurnsM,etal.Implementingarapidantiretroviral therapyprogramusingstarterpacksforemergencydepartment patientsdiagnosedwithHIVinfection. OpenForumInfectDis. 2023;10(7):ofad292.

19.KigerMEandVarpioL.Thematicanalysisofqualitativedata:AMEE guideno.131. MedTeach. 2020;42(8):846–54.

20.GrotellLA,BrysonL,FlorenceAM,etal.Postpartumnotetemplate implementationdemonstratesadherencetorecommendedcounseling guidelines. JMedSyst. 2021;45(1):14.

21.ChoudharyD,BrownB,KhawarN,etal.Implementationofelectronic medicalrecordtemplateimprovesscreeningforcomplicationsin childrenwithtype1diabetesmellitus. PediatricHealthMedTher. 2020;11:219–23.

22.SchwartzHEM,StarkNR,SowaCS,etal.Buildingbackbetter:applying lessonsfromtheCOVID-19pandemictoexpandcriticalinformation access. JEmergMed. 2021;61(5):607–14.

23.MarcusJL,HurleyLB,KrakowerDS,etal.Useofelectronichealth recorddataandmachinelearningtoidentifycandidatesforHIV pre-exposureprophylaxis:amodellingstudy. LancetHIV. 2019;6(10):e688–95.

ORIGINAL RESEARCH

DistributionandUtilization:AProspectiveCohortStudy

AlexanderJ.Ulintz,MD*

SeemaS.Patel,MD†

KatherineAnderson,MD†

KevinWalters,MD†

TylerJ.Stepsis,MD†

MichaelS.Lyons,MD,MPH*

PeterS.Pang,MD,MS†

SectionEditor:FaithQuenzer,DO,MPH

*TheOhioStateUniversityCollegeofMedicine,DepartmentofEmergencyMedicine, Columbus,Ohio

† IndianaUniversitySchoolofMedicine,DepartmentofEmergencyMedicine, Indianapolis,Indiana

Submissionhistory:SubmittedOctober27,2023;RevisionreceivedJune12,2024;AcceptedAugust15,2024

ElectronicallypublishedSeptember19,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18513

Objective: Foodinsecurityisaprevalentsocialriskamongemergencydepartment(ED)patients. PatientswhomaybenefitfromfoodinsecurityresourcesmaybeidentifiedviaED-basedscreening; however,manypatientsexperiencedifficultyaccessingresourcesafterdischarge.Co-locating resourcesinorneartheEDmayimproveutilizationbypatients,butthisapproachremainslargely unstudied.Thisstudycharacterizedtheacceptanceanduseofafoodvoucherredeemableatahospital foodmarketforpatientswhoscreenedpositiveforfoodinsecurityduringtheirEDvisit.

Methods: Thisprospectivecohortstudy,conductedatasinglecounty-fundedED,includedconsecutive adultpatientswhopresentedonweekdaysbetween8 AM–8 PM fromJuly–October2022andconsentedto researchparticipation.Weexcludedpatientswhorequiredresuscitationonarrivalorcouldnotprovide writteninformedconsentinEnglish.Studyparticipantscompletedapaperversionofthetwo-question HungerVitalSignscreeningtool,administeredbyresearchstaff.Participantswhoscreenedpositive receivedauniquelynumbered$30foodvoucherredeemableatthehospital’sco-locatedfoodmarket. Voucherredemptionwasquantiﬁedthroughregularevaluationofmarketreceiptrecordsat30-day intervals.Theprimaryoutcomewastheproportionofredeemedvouchers.Secondaryoutcomes includedtheproportionofparticipantsscreeningpositiveforfoodinsecurity,proportionofparticipants acceptingvouchers,andassociateddescriptivestatistics.

Results: Ofthe396eligibleindividualsapproached,377(95.2%)consentedandcompletedfood insecurityscreening.Mostweremiddle-aged(median53years,interquartilerange30–58years),191 werefemale(50.4%),242wereBlack(63.9%),and343werenon-Hispanic(91.0%).Oftheparticipants, 228(60.2%)screenedpositiveforfoodinsecurityand224receivedvouchers(98.2%),ofwhich86were redeemed(38.4%)amedianofninedaysaftertheEDvisit.

Conclusion: Ahighproportionofparticipantsscreenedpositiveforfoodinsecurityandacceptedfood vouchers;however,lessthanhalfofallvoucherswereredeemedattheco-locatedfoodmarket.These resultsimplyEDfoodvoucherdistributionforfoodinsecurityisfeasible,butco-locationofresources alonemaybeinsufﬁcientinaddressingthesocialriskandalludestoalimitedunderstandingoffacilitators andbarrierstoresourceutilizationfollowingED-basedsocialneedsscreening.[WestJEmergMed. 2024;25(6)993–999.]

INTRODUCTION

Socialrisks,definedasadversesocialconditions associatedwithpoorhealth,arecommonamongemergency department(ED)patientsandinfluencetheirhealth outcomes.1,2 Foodinsecurityisoneprevalentsocialrisk amongEDpatientsthatisassociatedwithprogressionand exacerbationofchronicdisease,frequentacutecareuse,and increasedall-causemortality.3–10 PreviousstudiesofEDbasedfoodinsecurityscreeningandreferraltoresources identifiedpatientswiththissocialrisk;however,many patientswhoreceivedresourcereferralshaddifficulty connectingtotheseresourcesafterdischarge.11,12 ColocatingresourcesinorneartheEDforpatientswhoscreen positiveforfoodinsecurityrepresentsonepotentialsolution thatdirectlyconnectspatientstotargetedresources; however,littleevidencesupportsthismodel.

Withinthecontextoffoodinsecurityresourceco-location, casereportsdescribea “foodbagprogram” pilotedduring theCOVID-19pandemic,whichprovidedanydischarged EDpatientwitha1–2daysupplyofshelf-stablefood.13 However,thiswasprimarilyanoperationalprojectthat prioritizedfooddistributionandwaslimitedinits conclusionsregardingfoodinsecurityscreeningandresource uptakeduetoco-location.13 Thus,itremainspoorly understoodwhetherco-locatingresourcesinorneartheED toaddressthissocialriskleadtoincreasedresource acceptanceorutilization.Inthisstudywesoughtto characterizetheacceptanceanduseofa$30foodvoucher redeemableatahospitalfoodmarketbystudyparticipants whoscreenedpositiveforfoodinsecurityduringtheir EDvisit.

METHODS

StudyDesign,Setting,andParticipants

Thisprospectivecohortstudyscreenedconsecutive consentingpatientspresentingtotheEDforfoodinsecurity andprovidedafoodvouchertoparticipantswhoscreened positive.ThisstudywasapprovedbytheIndianaUniversity SchoolofMedicineInstitutionalReviewBoard(IRB #13829).Thestudysitewasa95-roomcounty-fundedED withover100,000annualvisits.Allpatientsarrivingtothe EDare ﬁrsttriagedinthewaitingroomorambulancebayby anurse;patientswhodonotrequireemergentstabilization basedonappearanceandchiefcomplaintareevaluatedin oneof24intakerooms.Intakeroomsareprivaterooms whereacompletenursingevaluationandthe ﬁrstpatientphysicianinteractiontakesplace.

ThisstudyincludedEDpatientsaged ≥18yearswho presentedonweekdaysbetween8 AM–8 PM fromJuly1–October31,2022andwereevaluatedinanintakeroom duringtheirEDvisit.Thestudyprotocolexcludedpatients whowereminors,wereplacedinanon-intakeroom(dueto requiringimmediateresuscitationora1:1nursing intervention),orwhowereunabletoprovideinformed

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue?

FoodinsecurityiscommonamongED patients.Screeningidenti ﬁ esindividuals withthissocialrisk,butlittleevidence guidesreferral.

Whatwastheresearchquestion?

Willpatientswhoscreenedpositiveforfood insecurityintheEDaccepta$30foodvoucher redeemableataco-locatedhospitalmarket?

Whatwasthemajor ﬁndingofthestudy?

98.2%ofpatientswhoscreenedpositivefor foodinsecurityacceptedavoucher,butonly 38.4%hadredeemedthematamedianofnine dayslater.

Howdoesthisimprovepopulationhealth?

Referraltoco-locatedresourcesforfood insecurityisfeasible,butprogramsshould consideraccessibilityandpatientpreferences inaddressingsocialrisksidenti ﬁ edintheED.

consentinEnglish.Researchassistantsusedtheelectronic healthrecordtoassesspatientsforeligibilityupontheir arrivalintriageandapproachedeligibleparticipantsfor consentoncemovedtoaprivateintakeroom.Eligible participantsweremadeawareofthepurposeofthestudyand thattheirsurveyresponsescouldmakethemeligiblefora foodvoucheratthehospitalfoodmarket.

ScreeningToolDistributionandScreeningDataCollection

ConsentedparticipantsreceivedapaperversionoftheUS DepartmentofAgriculturebinaryquestionHungerVital Signscreeningtool(“Withinthepast12monthsweworried whetherourfoodwouldrunoutbeforewegotmoneytobuy more, ” and “Withinthepast12monthsthefoodwebought justdidn’tlastandwedidn’thavemoneytogetmore”).14 Participantsusedaprovidedwritinginstrumenttocheckyes ornotoeachscreeningtoolquestion.Apaperversionwas chosenbasedonpriorworkbyGonzalesetal(2021),which demonstratedthat75%ofpatientspreferredfoodinsecurity screeningviapaperasopposedtoverbalresponses;however, theconsentprocessdidnotifyparticipantsthatiftheycould notreadorwrite,theresearchassistant(RA)couldreadthe screeningtoquestionstothem.15 Ifthepatientdeclined assistanceinreadingor ﬁllingoutthescreeningtool,theRA lefttheroomfor10minutesandreturnedtocollectthe completedscreeningtool.TheRAdirectlyenteredscreening

toolresultsintoapredeﬁneddatacollectioninstrument, REDCap.Wecollectedandmanagedstudydatausing REDCapelectronicdatacapturetoolshostedatIndiana University,whichincludedthepatient’shealthrecord numberforlongitudinaltracking.Paperscreeningtoolswere thendestroyedviathehospital’ sconﬁdential-document disposalsystem.

Intervention

Iftheparticipantscreenednegativeforfoodinsecurity,no furtherinterventionwasperformed.Iffoodinsecuritywas identiﬁedonthescreeningtool,theparticipantreceiveda$30 foodvoucherredeemableatthehospital’sco-locatedfood market.16 The “FreshforYou” hospital-basedmarketwas designedbythehealthsystemtoaddressfoodinsecurityby providingpatients,visitors,andstaffeasyaccesstofresh produce,preparedfoods,healthysnacks,convenience ingredients,kitchenutensils,andpantrystaplesataffordable pricesandinaconvenientlocation.17 Thefoodmarketis openweekdaysfrom10:30 AM–6 PM andislocated approximately600feetfromtheEDentrance,positioned nearabusstopandtheparkinggarage.Priortothisstudy,a similarscreeningtoolandvoucherreferralsystemhadbeen usedinselectoutpatientpractices;thevalueofthevoucher waschosenasitwassimilartotheoutpatientpractices.Each foodvoucherhadarandomthree-digitcodeontheback,and thiswasrecordedintheRedCapdatabasepriorto distributionbytheresearchstaff.Inadditiontothefood voucher(studyintervention),participantsscreeningpositive forfoodinsecurityalsoreceivedresourcesthatwerestandard ofcarepriortothisstudy,whichincludedprintedcommunity resourcesforfoodinsecurity.Ifafoodvoucherwas redeemed,itwasmarkedwiththedateandtimeof redemptionbymarketstaff.At30-dayintervals,thestudy teamqueriedhospitalfood-marketreceiptrecordsto determinewhetheravoucherhadbeenredeemedand secondarilyperformedpatientchartreviewforprimarycare follow-upvisits.

Outcomes

Theprimaryoutcomewastheproportionoffood vouchersredeemedbyparticipantswhoscreenedpositivefor foodinsecurity.Secondaryoutcomesincludedthe proportionofparticipantsscreeningpositiveforfood insecurity,proportionofindividualswithoutpatientfollowupaftertheirEDvisit,anddemographicdescriptionsof thesegroups.

StatisticalMethods

Theanalysisplanincludeddescriptivedataanalysiswith frequencies,proportions,andmedianswithinterquartile range(IQR).Wedidnotcalculateanapriorisamplesizedue toenrollmentbeinglimitedbythenumberofavailable vouchers.Aposthocpowercalculationfordetecting

differencesbetweenparticipantswhoredeemedordidnot redeemafoodvoucherdemonstratedlessthan80%power; thus,statisticalcomparisonwasnotperformed,andonly descriptivestatisticsarereported.Allstatisticalanalyses wereperformedusingSTATAICversion17(StataCorp, LLC;CollegeStation,TX).

RESULTS

Researchassistantsapproached396eligibleindividuals, ofwhom379consentedandcompletedthefoodinsecurity screeningtool(Figure1).Noindividualwasscreenedmore thanonceduringthestudyperiod.Mostparticipantswere middle-aged(median53years,IQR30–58years);191 identiﬁedasfemale(50.4%);242asBlack(63.9%);and343as non-Hispanic,(91.0%);234reportedhavingaprimarycare physician(61.7%).Mostparticipants(228)screenedpositive forfoodinsecurity(60.2%)(Table1).Ofthese,194 participants(51.2%)worriedaboutfoodrunningoutbefore havingmoneytobuymore,207respondents(54.6%) reportedfoodnotlastinglongenoughandnothavingmoney formore,and175respondents(46.2%)reportedboth concerns.(Table2).

TheRAsdistributed224vouchers(98.2%ofparticipants whoscreenedpositive)andobserved86(38.4%)redemptions within30daysofdistribution(Figure2).Themediantimeto voucherredemptionwasninedays(IQR9–19days).Of participantsscreeningpositiveforfoodinsecurity,98(43.0%) hadprimarycarefollow-upwithin90daysoftheED visit.Themediantimetoprimarycarefollow-upwas41days (IQR21–67days).Ofnote,39participants(17.1%of thosewhoscreenedpositiveforfoodinsecurity)neither redeemedafoodvouchernorattendedprimarycare follow-up(Table3).

Demographicdescriptionsdidnotvarygreatlybetween participantswhoredeemedavouchercomparedtothosewho didnot.However,foodvoucherredemptionwasaffectedby dischargetime:participantsdischargedduringmarket operatinghourshadahigherproportionofvoucher redemption66,41.5%)comparedtoparticipantsdischarged whenthemarketwasclosed(20,30.8%)(Appendix1).

DISCUSSION

Inthisprospectivecohortstudyofadultpatientsseenata countyED,ahighproportionofrespondentsscreened positiveforfoodinsecurityandacceptedfoodvouchers; however,theredemptionrateoffoodvouchersatthe hospital’sco-locatedfoodmarketwaslowandoften occurredgreaterthanoneweekaftervoucherdistribution.

Theobservedproportionofparticipantsscreening positiveforfoodinsecurityinthisstudy(61%)isgreaterthan inpriorstudiesofEDpatients,whichhistoricallyranged from16–51%.6,9,15,18–21 Evenwhenaccountingfortheeffect oftheCOVID-19pandemic,theobservedproportionof participantsscreeningpositiveforfoodinsecurityinthis

StudyCONSORTdiagram.

Table1. Demographicdescriptionofenrolledparticipants,stratiﬁedbypresenceorabsenceoffoodinsecurity.

Variable

Gender, n(column%)

Allparticipants (N = 379)

Foodinsecuritypresent (n = 228)

Foodinsecurityabsent (n = 151)

Male188(49.6)106(46.5)82(54.3)

Female191(50.4)122(53.5)69(45.7)

Age, median(IQR) 53(30–58)45(32–57)40(29–59)

Race, n(column%)

Black242(63.9)149(65.4)93(61.6)

NativeHawaiian/PaciﬁcIslander1(0.2)1(0.4)0

White133(35.1)78(34.2)55(36.4)

Missing3(0.8)03

Ethnicity, n(column%)

HispanicorLatino/a34(9.0)20(8.8)14(9.4)

Non-Hispanic343(91.0)208(91.2)135(90.6)

Missing202

Accesstocare, n(column%)

PCPpriortostudy234(61.7)141(62.4)93(61.6)

IQR,interquartilerange; PCP,primarycarephysician.

studywashigherthanotherstudiesduringthisperiod, includinganidenticalscreeningprocessimplementedin outpatientclinicsatthestudysite(30–37%).4,16 Participant acceptanceofavoucherwasalsohigherthanthe65%

acceptancerateinAylmard’s2021studyandBottinoetal’ s 2017studyacceptancerateof17%.18,22 However,participant useofthefoodvoucherresourcewassimilartoprior acceptanceratesforsocialservicesreferralsandshows

Figure1.

Table2. HungerVitalSignsquestionresponsesandscreening results(N = 379).

Questionresponsen(%)

“Withinthepast12monthsweworriedwhetherour foodwouldrunoutbeforewegotmoneyto buymore” 194(51.2)

“Withinthepast12monthsthefoodweboughtjust didn’tlastandwedidn’thavemoneytogetmore

ScreeningResult

Screenedpositiveforfoodinsecurity228(60.2)

Answeredyestoonlyonequestion51(22.4)

Answeredyestobothquestions175(76.8)

Screenednegativeforfoodinsecurity151(39.8)

Table3. Foodinsecurityafterdistribution,voucherredemption,and primarycarephysicianfollow-up(n = 228).

Variablen(%)

Foodvoucher

Vouchersdistributed224(98.2) Vouchersredeemed86(38.4)

Time-to-redemption, median(IQR) 9days(9–19)

Follow-up

PCPappointmentwithin90days98(43.3)

Time-to-appointment, median(IQR) 41days(21–67)

FoodvoucherandPCPfollowup

Bothused12(5.3) Neitherused39(17.3) IQR,interquartilerange; PCP,primarycarephysician.

ofothercookingitems(eg,cookingsprayorutensils),ifthose wereneededmorethanfooditems.Additionally,thestudy protocolalloweddistributionofperishablefood,whichhas rarelybeenofferedinsimilarED-basedprograms.These differenceshighlightkeytradeoffsbetweentwodifferent approaches:theeaseofpre-made,ED-distributedfoodbags vsthecustomizabilityofapatientsshoppingforthemselves. Futureworkshouldfurthercharacterizepatientpreferences betweenthesestrategiestoprovidecriticalinsightsintothe circumstancesinwhichoneispreferredovertheotherby patientswithsocialneeds.

Figure2. Frequencyofscreeningtoolcompletion,positive screeningforfoodinsecurity,voucheracceptancebypatient, andvoucherredemption.

similarutilizationofsummerfoodprogramsthataddressed foodinsecurityatachildren’sED.18,23

ThisstudybuildsupontheworkofJahnesetal(2020), whichaddressedasimilarproblemthroughadifferent approach:theauthorsimplementedaprograminwhich patientsweregivenabagoffoodatthetimeofdischarge withoutexaminationofeligibilitycriteriaorfurther documentation.Bagsincludedshelf-stablefoodaswellas “no-cookbags” forindividualswithoutcooking infrastructure.The ﬁrstnotabledifferenceinresultsisthat 3,000foodbagsweredistributedbyJahnesetalasopposed to226foodvouchersinthisstudy.Whilethemonetaryvalue ofeachfoodbaginJahnesetalisnotknown,evenifone assumedacostof$10perbag,morefoodwasdistributedin anoperationalprogramfocusedonfooddistributionrather thaninthisresearchstudy.Thesecondnotabledifference wasthatthisstudy’sapproachallowedparticipantstochoose whatoptionsbestservedtheirneeds,includingthepurchase

Itwasunexpectedthatdespitehighacceptanceratesof foodvouchers,lessthanhalfofallparticipantsredeemeda voucherattheco-locatedhospitalfoodmarket.Aconceptual explanationofthisdiscrepancycouldbesocialriskvssocial need;whilethescreeningidentiﬁedasocialrisk(ie,an adversesocialconditionassociatedwithpoorhealth)and providedresourcesdirectedatreducingasocialrisk, participantsmaynothaveperceivedfoodasasocialneed (ie,anadversesocialdeterminantofhealthforwhichthey wouldhavelikedassistanceandviewedasapriority).24

However,thisisconsideredlesslikelyduetothemedian redemptiontimeoffoodvouchersofninedays,suggesting thatusingthevoucherwasimportantenoughtoreturnfor redemption.Apragmaticexplanationoftheobserved discrepancyisthatthefoodmarketwasinitiallydesignedfor afoodinsecurityscreeningandinterventionintheoutpatient primarycaresetting(ie,weekdays,daytimehours),rather thantheEDsetting(ie,alldaysandhours).The misalignmentbetweenEDscreeningtimesandmarkethours appearstohavemodiﬁedtheeffectoffoodvoucher distributiononredemptionrates.Theoverallredemption ratewas38.9%;redemptionrateforindividualsdischarged duringmarkethourswas41.5%,whiletheredemptionrate forindividualsdischargedaftermarkethourswas30.8%). Theseresultsshouldpromptfurtherconsiderationofunique

aspectsofEDoperationswhendesigningfuturefood insecurityinterventionswithinahospitalsystem.

Itwasalsosurprisingthatnearly60%ofparticipants experiencingfoodinsecurityhadanestablishedprimarycare doctorpriortotheirEDevaluation.This ﬁndingcontrasts withRobinsonetal’s2018study,whichfoundfoodinsecurity wasassociatedwithlackofprimarycare.3 Whilethisstudy wasnotdesignedtodeterminewhetherpreviousprimarycare appointmentshadscreenedfororaddressedfoodinsecurity, participantwillingnesstodiscloseasocialriskduringtheED encounteralignswithCullenetal’s2019study,whichfound thatfamiliesweremorecomfortablewithsocialdeterminants screeningintheEDratherthantheprimarycaresettingand supportsongoingeffortstoscreenforsocialneedsintheED.25 Theobservedfollow-uprateoflessthan50%andtimeto primarycareappointmentexceedingonemonthareconsistent withpriorobservationsbyLooetal(2013),Wallaceetal (2021),andZuetal(2006).11,12,19

These ﬁndingsalsohighlighttheimportantroleoftheED inaddressingsocialrisksthatareidentiﬁedduringED screening.Ifscreenedrisksarenotaddressed(andinstead referredtooutpatientphysicians),follow-upmaynotoccur foruptoonemonth.Oneconceptualquestionthatremains unanswerediswhatservices(eg,primarycare,socialwork, casemanagement,nutrition/dietetics,communityagencies, ormultidisciplinaryteamseitherinpersonorvirtually)are mostappropriatetoreferpatientstoaftertheyscreenpositive forasocialneed,suchasfoodinsecurity,intheED.The studyprotocoloptedtoreferparticipantsbacktotheir primarycarephysicianbecausefoodinsecuritywouldlikely requireamorecomprehensivesocialneedsassessment,but researchersinfuturestudiesmaywishtoconsideralternative strategiestoaddressthisquestion.

LIMITATIONS

Thestudydesignwasatriskofselectionbias,participation bias,andcontaminationbias.Selectionbiasoccurredduring inclusion/exclusionwhereinindividualsarrivingoutsidethe studyhours,individualswithpsychiatricillness,patients presentinginextremis,minors,andindividualswhocould notprovidewrittenconsentinEnglishwereexcluded. Althoughsubsequentqualityimprovementprojectshave addressedthesepopulations,theresearchresultspresented herearenotgeneralizabletopatientsoutsidethestudy population.Participationbiasmayhaveoccurredduetothe ethicsrequirementtodisclosetherisksandbeneﬁtsofstudy participation,includingafoodvoucherwithmonetaryvalue; thismayalsoexplainthehigher-than-expectedproportionof participantsscreeningpositiveforfoodinsecurityamongthe studypopulation.Additionally,participationbiasmayhave occurredusingawrittenscreeningtoolthatmayhavemade individualswithlowliteracylesslikelytoparticipate,even thoughtheRAprotocolincludedofferingthatthescreening questionsbereadaloud.

Thesingle-centerstudydesignwithoutlongitudinal contactwithstudyparticipantslimitsourabilitytocomment oncontaminationbias;itispossiblethatpatientsobtained connectiontocarefromexternalresourcesratherthanthe study’svoucherprogramandprintedresourcesanddidnot usetheprovidedvoucherforthisreason.Thelowredemption rateoffoodvoucherswasunexpected,andthestudydesign andinformedconsentdidnotallowfurtherinvestigationinto thereasonsforthelowproportionofvoucherredemption; thus,conclusionsaboutthecausesofthis ﬁndingarelimited. Finally,theuniqueaspectsofprojectfundingandmarket locationlimitthisstudy’sgeneralizabilitytosimilarhealth systemswithsimilaravailableresources.

CONCLUSION

Ahighproportionofstudyparticipantsscreenedpositive forfoodinsecurityandacceptedafoodvoucherforacolocatedresourceaddressingthissocialrisk;however,voucher redemptionrateswerelowandoccurredgreaterthanone weekfollowingdistribution.Theseresultsimplythatfood insecurityscreeningandvoucherdistributionarefeasible, butthatco-locationofresourcesalonemaynotcompletely addressthesocialriskandshouldpromptconsiderationof resourceaccessibility(bothlocationandhours), customizability,andpatientpreferencesintreatingsocial needsidentiﬁedintheemergencydepartment.

AddressforCorrespondence:AlexanderJ.Ulintz,MD,TheOhio StateUniversityCollegeofMedicine,DepartmentofEmergency Medicine,750PriorHall,376W10th Avenue,Columbus,Ohio 43210.Email: alexander.ulintz@osumc.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Thisstudywasfundedbyaninternal grantwithintheIndianaUniversitySchoolofMedicineDepartmentof EmergencyMedicine.Theauthorsreporttherearenocompeting intereststodeclare.Therearenootherconflictsofinterestorsources offundingtodeclare.

REFERENCES

1.AlderwickHandGottliebLM.Meaningsandmisunderstandings:a socialdeterminantsofhealthlexiconforhealthcaresystems. MilbankQ. 2019;97(2):407–19.

2.GordonJA.Thehospitalemergencydepartmentasasocialwelfare institution. AnnEmergMed.Mar1999;33(3):321–5.

3.RobinsonT,BryanL,JohnsonV,etal.Hunger:amissedopportunityfor screeninginthepediatricemergencydepartment. ClinPediatr(Phila). 2018;57(11):1318–25.

4.EnayatiD,ChanV,KoenigG,etal.ImpactofCOVID-19pandemicon foodinsecurityinanurbanemergencydepartmentpatientpopulation. WestJEmergMed. 2023;24(2):127–34.

5.KerseyMA,BeranMS,McGovernPG,etal.Theprevalenceandeffects ofhungerinanemergencydepartmentpatientpopulation. AcadEmerg Med. 1999;6(11):1109–14.

6.BirosMH,HoffmanPL,ReschK.Theprevalenceandperceivedhealth consequencesofhungerinemergencydepartmentpatientpopulations. AcadEmergMed. 2005;12(4):310–7.

7.NhoungHK,GoyalM,CacciapuotiM,etal.Foodinsecurityandinsulin useinhyperglycemicpatientspresentingtotheemergencydepartment. WestJEmergMed. 2020;21(4):959–63.

8.MenF,UrquiaML,TarasukV.Pain-drivenemergencydepartmentvisits andfoodinsecurity:across-sectionalstudylinkingCanadiansurveyand healthadministrativedata. CMAJOpen. 2022;10(1):E8–E18.

9.EstrellaA,ScheidellJ,KhanM,etal.Cross-sectionalanalysisoffood insecurityandfrequentemergencydepartmentuse. WestJEmergMed. 2021;22(4):911–8.

10.WalkerRJ,ChawlaA,GaracciE,etal.Assessingtherelationship betweenfoodinsecurityandmortalityamongU.S.adults. AnnEpidemiol. 2019;32:43–8.

11.WallaceAS,LutherBL,SislerSM,etal.Integratingsocialdeterminants ofhealthscreeningandreferralduringroutineemergencydepartment care:evaluationofreachandimplementationchallenges. ImplementSci Commun. 2021;2(1):114.

12.ZunLSandDowneyL.Adulthealthscreeningandreferralinthe emergencydepartment. SouthMedJ. 2006;99(9):940–8.

13.JahnesK,HopkinsK,FooteE,etal.(2020).Foodbagprogramto addresstheimmediatefoodneedsofpatientsduringtheCOVID-19 crisis.In: COVID-19:AnnFamMed[Preprint].Availableat: https://deepblue.lib.umich.edu/handle/2027.42/156010 AccessedAugust31,2023.

14.HagerE,QuiggA,BlackM,etal.Developmentandvalidationofa 2-itemscreentoidentifyfamiliesatriskforfoodinsecurity. Pediatrics. 2009;126(1):26–32.

15.GonzalezJV,HartfordEA,MooreJ,etal.Foodinsecurityinapediatric emergencydepartmentandthefeasibilityofuniversalscreening. WestJ EmergMed. 2021;22(6):1295–300.

16.ReinosoD,HautD,ClaffeyS,etal.Addressingfoodinsecurity:lessons learnedfromco-locatingafoodpantrywithafederallyqualiﬁedhealth center. IntJIntegrCare. 2022;22(3):24.

17.HealthE.Freshforyoumarketprovideseskenazihealthfrontline staffaccesstogroceryitems.2020.Availableat: https://www. eskenazihealth.edu/news/fresh-for-you-market-provides-eskenazihealth-frontline-staff-access-to-grocery-items AccessedAugust29,2022.

18.AylwardAF,EngelbergAndersonJ,MorrisA,etal.Usingmalnutrition andfoodinsecurityscreeningtoidentifybroaderhealth-relatedsocial needsamongstolderadultsreceivingemergencydepartmentcarein theSoutheasternUnitedStates:across-sectionalstudy. HealthSoc CareCommunity. 2021;29(6):e420–30.

19.LooS,AndersonE,LinJG,etal.Evaluatingasocialriskscreeningand referralprograminanurbansafety-nethospitalemergencydepartment. JAmCollEmergPhysiciansOpen. 2023;4(1):e12883.

20.MinerJR,WestgardB,OlivesTD,etal.Hungerandfoodinsecurity amongpatientsinanurbanemergencydepartment. WestJEmergMed. 2013;14(3):253–62.

21.ShararehN,WallaceAS,BrintzBJ,etal.Associatedfactorswithpatientreportedunmetfoodneedsamongemergencydepartmentadult patients-Asocialneedperspective. PrevMedRep. 2022;29:101974.

22.BottinoCJ,RhodesET,KreatsoulasC,etal.Foodinsecurityscreening inpediatricprimarycare:Canofferingreferralshelpidentifyfamiliesin need? AcadPediatr. 2017;17(5):497–503.

23.CullenD,BlauchA,MirthM,etal.Completeeats:Summermeals offeredbytheemergencydepartmentforfoodinsecurity. Pediatrics. 2019;144(4):e20190201.

24.MolinaMF,LiCN,ManchandaEC,etal.Prevalenceofemergency departmentsocialriskandsocialneeds. WestJEmergMed. 2020;21(6):152–61.

25.CullenD,WoodfordA,FeinJ.Foodforthought:aRandomizedtrialof foodinsecurityscreeningintheemergencydepartment. AcadPediatr. 2019;19(6):646–51.

TheCaliforniaManagedCareOrganizationTaxandMedi-Cal PatientsintheEmergencyDepartment

LaurenMurphy,MS*†

GitaGolonzka,BS‡

EllenShank,MD§ JorgeFernandez,MD†

*CaliforniaStateUniversity,Chico,Chico,California † CaliforniaChapteroftheAmericanCollegeofEmergencyPhysicians, Sacramento,California ‡ UniversityofCaliforniaSanDiego,SchoolofMedicine,LaJolla,California § UniversityofCaliforniaDavisHealth,DepartmentofEmergencyMedicine, Sacramento,California

SectionEditor:MarkI.Langdorf,MD,MHPE

Submissionhistory:SubmittedSeptember1,2024;RevisionreceivedSeptember4,2024;AcceptedSeptember6,2024

ElectronicallypublishedSeptember19,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.35257

Lastweek,Timothy*,a60-year-oldunhousedgentleman, presentedtotheEDrequestingmedicalclearance.When askedwhy,heexplainedthathewasstayingatawarming centerandhadbeenseparatedfromhiswifeandquarantined inaseparatesleepingspaceduetoconcernsthathewas infectious.Hethendoffedhisbeaniecaptoreveala10 × 10centimeterfungatingsquamouscellcarcinomathathadbeen thrivingonhisscalpforthepasttwoyears.Hehadmoved countiesafewyearsbackandsincethenhadhaddifﬁculty reestablishingandunderstandinghiscoveragewithina geographicallymanagedcaresystem.Hewasonthewaiting listtoseeadermatologistwhoacceptedMedi-Cal (California’sversionofUSMedicaidhealthinsuranceforthe poor).Furtherchartreviewrevealedthat,followingatleast threeEDvisits,hehadbeenreferredtohealthnavigatorsto trytosecureadermatologistappointment;buthehadbeen waitingoverayearandnowwassimplyrequestingmedical clearancetobereunitedwithhiswife.Thiscasewasastark exampleofthearduousbarrierssomepatientsinCalifornia mustovercometoreceivecareandthehurdlesthat emergencyphysicians(EP)mustsurmounttohelpthese coveredpatientsaccessthefollow-upcaretheyrequire.

HISTORYOFTHEMANAGEDCARE ORGANIZATION(MCO)TAX

Medi-Calisfundedthroughstateandfederaldollars.The federalgovernmentusestheFederalMedicaidMatching RatetocalculatehowtomatchstatespendingonMedicaid programs.ForeverydollarspentonMedicaid,astatecan receiveatleast$1inMedicaidfederal ﬁnancialparticipation (FFP)funding.Becausemanystatesstruggledtogenerate enoughrevenuetocovertheirshareofMedicaidcosts,in 1985theHealthCareFinancingAdministrationallowed statestoacceptdonationsfromprivatemedicalcare providersanddeemedthesedonationseligibleforFFP matching.1 In2006,Congressgavestatestheauthoritytotax

providers,includingmanagedcareorganizations(MCO),to meettheirshareofMedicaidspending.2 Accordingtothe MedicaidandChildren’sHealthInsuranceProgram PaymentandAccessCommission,49statesusedsomeform ofprovidertaxtofundtheirMedicaidprogramsin2019.3

CaliforniahasusedanMCOtaxforover20yearsto receivefederalFFPmatchingfunds.Thetaxmustbe authorizedbythelegislatureandissubjecttofederal approval.Theauthorizinglegislationincludesanenddate forthetax,requiringittobereauthorizedperiodically. Historically,Californiahasusedthemoniesgenerated byFFPdollarstobackﬁlltheGeneralFunddeﬁcitor tofundanarrayofpublicservicesandsystemsoutsidethe Medi-Calprogram.4

2023MANAGEDCARETAXAGREEMENT

WhentheauthorizinglegislationforthepreviousMCOtax wasnearingitssunsetdate,acoalitionofprovidersand healthcarefacilitiescametogethertonegotiateanewMCO taxagreement.Theproposalbythecoalitionwastodedicate theadditionalfundstoprovideaMedi-Calrateincreasefora varietyofhealthcareproviders.Theincreasewasproposedto rolloutoveraseriesofyears.Inthe ﬁrst ﬁscalyear2023–24,a subsetofprimarycareclinicians,reproductivehealthservices, andsomeoutpatientmentalhealthserviceswouldreceive increaseseffectiveJanuary1,2024.Italsoproposedto appropriate$1.28billionforprimarycarerateincreasesand $1.15billioninspecialtyrateincreaseseffectiveJanuary1, 2025.5 Therewereadditionalfundsallocatedforfacilitiesand transport,reproductivehealth,andmentalhealththatwould alsotakeeffectin2025.ThecoalitionarguedthatCalifornia’ s Medi-Calreimbursementrateswereobjectionablylow, forcingclinicianstolimitthenumberofMedi-Calpatients theysee,therebylimitingpatientaccesstocare.

AstheLegislaturewasconsideringthecoalitionproposal, theCaliforniaChapteroftheAmericanCollegeof

EmergencyPhysician(CalACEP)advocacyteamlobbied foraspeciﬁcpooloffundingexclusivelyforEPs,arguingthat thefundingwasnecessaryduetotheirexclusionfrom previousMedi-Calphysicianincreasesandtheir disproportionatecarefortheMedi-Calpopulationrelativeto otherclinicians.Asaresultofitsadvocacy,a$200million statebudgetlineitemwasincludedtobringEMMedi-Cal ratesto87.5%ofMedicare.6

MAYREVISIONANDFINALBUDGET

Theagreementbetweenthecoalition,CalACEP,andthe CaliforniaLegislaturewascodifiedinlegislationin2023and wasscheduledtotakeeffectJanuary1,2025.However, becausethestatebudgetisonlyaone-yeardocumentandthe Legislaturedoesnothavethesolepowertocreatethestate’ s budget,theMCOtaxagreementforthe2024–25 fiscalyear wassubjecttonegotiationonceagainbetweenthe Legislatureandthegovernor.GovernorNewsomincluded theagreedrateincreasesinhisproposedbudgetreleasedin January2024,buthisrevisedbudgetreleasedinMay removedtherateincreaseentirelyandusedtheMCOtax surplustobackfillCalifornia’sGeneralFunddeficit.7

TheCalACEPadvocacyteamusedavarietyofstrategiesto lobbylegislatorstorestorethe$200millionforEPrate increases.StafflobbiedmembersoftheBudgetSubcommittees inpersonforeachhouseoftheLegislaturetoexplainthe impactoflowreimbursementonemergencymedicinepractice andhowiteffectsaccesstocareforpatients.Staffalso coordinatedatargetedsocialmediacampaignthatfocusedon theSacramentoareatokeeptheissueinthemindsof stakeholders.MembersoftheCalACEPExecutiveCommittee wroteletterstotheeditorsofmajornewsorganizations throughoutthestate.Finally,CalACEPcoordinateda grassrootscampaigntoencourageEPstocontactlegislators directlyandtellpersonalstoriesabouttheirpatientsandMediCalaccess.Ultimately,the$200millionforEPrateincreases were includedinthebudgetthatGovernorNewsomsignedon June29,2024.However,noneoftheotherpreviouslypromised rateincreasesforotherspecialties,whichhadbeenscheduledto takeeffectJanuary1,2025,wereincluded.

PROPOSITION35

WhileEPswereabletogettheirincreaserestored,other physicianswerenotsofortunate.TheCaliforniaMedical Association,withthesupportoforganizationsthatwerea partofthe2023MCOcoalition,qualiﬁedaninitiativeforthe 2024CaliforniaStateBallotthat,ifpassedbyCalifornia voters,wouldpermanentlyenshrinetheagreementreached bytheCoalitionandtheLegislaturein2023intolaw. Proposition35wouldrequirethestatetousetheMCOtax revenuestofundincreasesfortheMedi-Calprogramforthe durationofthecurrentMCOtaxauthorization.Thiswould restoretherevokedMedi-Calrateincreasesforother cliniciansandprotecttheEPincreaseforyearstocome,

withouthavingtorenegotiatewiththeLegislatureand governoreveryyear,asistypicallyrequiredbytheannual budgetprocess.AsofAugust2024,thepropositionseemsto havebroadbipartisansupport.Numeroushealthcare organizationsincludingtheCaliforniaMedicalAssociation, PlannedParenthood,theCaliforniaHospitalAssociation, andCalACEPformallysupporttheproposition.Other supportersincludetheCaliforniaHawaiiStateConference NationalAssociationfortheAdvancementofColored People,theInsuretheUninsuredProject,theCalifornia DemocraticParty,andtheCaliforniaRepublicanParty.8

Thepropositionhasnoregisteredopposition,but GovernorNewsomhasexpressedseriousconcernsinthe pressabouttheimpactthepropositionwillhaveonthe abilityoftheLegislatureandgovernortodeliverabalanced budgetinthefuture.9

IMPACTONPATIENTSANDCLINICIANS

Emergencydepartments(ED)areuniquelypoisedto sufferthestrainofinsufficientMedi-Calfunding,aspatients coveredbyMedi-CalseekcareintheEDathigherratesthan patientswithothertypesofinsurance.Since2019,oneof threeCaliforniansarecoveredbyMedi-Cal,10 but42%of visitstotheEDarebyMedi-Calpatients.11 Thisdiscrepancy islargelydrivenbylowphysicianparticipationinMedi-Cal. California’slowMedi-Calreimbursementratesfor physiciansresultinveryfewprimarycarephysiciansand specialistswhoacceptpatientswithMedi-Calandlongwait times.12 Thisoftenleavesthesickestandmostvulnerable patientswithnowheretogoexcepttheED.Emergency physiciansoperateundertheEmergencyMedicalTreatment andLaborAct,enactedin1986,whichensuresthatall patientsarrivingintheEDwillreceivetreatmentregardless ofinsurancestatusorabilitytopay.Giventhecurrentstate ofMedi-Calfunding,EPsareforcedtonavigatenotonly direlysickandurgentcases,butto fillgapsforpreventiveand specialtycare.Thisleavesthemscramblingto findand coordinatefollow-upcare.Italsoleavesthemwiththemoral injuryofshowinguptoshiftsdayinanddayoutwithout beingabletogettheirpatientsthetransferandfollow-upcare theyneed.ThisistakingatollonpatientsandEPsalike. WhileEPsareproudtohelppatientswhentheyneedit most,thecurrentfundingconditionsareunsustainable. DespitethenumberofEDpatientsincreasing,underfunded departmentsaredecreasingthenumberofEPsworking,and insomecasesareemployingmorephysicianassistantsand nursepractitioners.13 Inturn,Californiafacesdifficultyin hiringandretainingwell-qualifiedandexperiencedEPs, particularlyinhistoricallyunder-resourcedareas.14 These changesimpactallEDpatientsintheformoflongerwait timesand,intheworstcases,poorpatientoutcomes. TheimpactofMedi-Calunderfundingreverberatesfar beyondemergencyservices,especiallyasMedi-Caleligibility criteriaexpands.Underthecurrentfundingscheme,an

increasingnumberofpatientscoveredbyMedi-Calin California’shealthcaresystemwillresultindeeperinequity, asmoreMedi-Calpatientsarecompetingforthesamefew spotsfromalimitednumberofphysicians.Aproportional increaseinaccesstophysiciansisneeded,andthismustcome intheformofincreasedfunding,suchastheMedi-Calrate increasesoutlinedinProposition35.

Emergencyphysiciansneedasufﬁcientnetworkof primarycareandspecialtyphysicianswillingtoacceptMediCaltoprovidecomprehensiveandtimelycaretopatients. AdequatenetworkswouldenablepatientslikeTimothyto receivethecaretheyneed,whentheyneedit,asopposedto thecurrentsystemofbacklogandwaiting.Increasing reimbursementtomorecloselymatchthecostofcare,and protectingdedicatedfunds,willimproveefﬁciencyand equityinthehealthcaresystem,ultimatelyimprovingthe qualityofcareforallCalifornians.

*Patientnamehasbeenchangedtoprotectconﬁdentiality.

AddressforCorrespondence:LaurenMurphy,MS,PublicHealth& HealthServicesAdministration,400W.FirstSt.,Chico,CA95929. Email: LAMurphy@csuchico.edu

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.LaurenMurphyisapaidemployeeof CalACEP.EllenShankisaHealthPolicyfellowatUCDavisandnonvotingmemberoftheCalACEPboardofdirectors.Therearenoother conflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.KuLandCoughlinTA.1995.Medicaiddisproportionateshareandother special ﬁnancingprograms. HealthCareFinancRev. 1995;16(3):27–54.

2.109thCongress.2006.S.1932-DeﬁcitReductionActof2005.

3.MACPAC.Healthcare-relatedtaxesinMedicaid.2021.Availableat: https://www.macpac.gov/wp-content/uploads/2020/01/HealthCare-Related-Taxes-in-Medicaid.pdf .AccessedAugust21,2024.

4.CaliforniaLegislativeAnalyst’sOfﬁce.TheMCOtaxpackageatMay revision.2024.Availableat: https://lao.ca.gov/Publications/Report/ 4905.AccessedAugust21,2024.

5.CaliforniaDepartmentofFinance.2024enactmentbudget summary-HealthandHumanServices.2024a.Availableat: https:// ebudget.ca.gov/2024-25/pdf/BudgetSummary/ HealthandHumanServices.pdf.AccessedAugust21,2024.

6.CaliforniaDepartmentofFinance.Budgetsummary.2024c. Availableat: https://ebudget.ca.gov/2024-25/pdf/Enacted/ BudgetSummary/HealthandHumanServices.pdf AccessedAugust21,2024.

7.CaliforniaDepartmentofFinance.2024–25Mayrevisionbudget summary.2024.Availableat: https://ebudget.ca.gov/2024-25/pdf/ Revised/BudgetSummary/FullBudgetSummary.pdf AccessedAugust21,2024.

8.CaliforniaSecretaryofState.Ofﬁcialvoterinformationguide. 2024.Availableat: https://voterguide.sos.ca.gov/ AccessedAugust21,2024.

9.LaL.NewsomsoundsoffonCAballotmeasuresoncrime,health insurance.2024.Availableat: https://calmatters.org/newsletter/ newsom-ballot-measures/.AccessedAugust21,2024.

10.McConvilleS,DanielsonC,HsiaR.Emergencydepartmentusein California:demographics,trends,andtheimpactoftheACA.2022. Availableat: https://www.ppic.org/publication/emergency-departmentuse-in-california-demographics-trends-and-the-impact-of-the-aca/ AccessedAugust21,2024.

11.CaliforniaHealthCareFoundation.Californiaemergencydepartments: acriticalsourceofcare.2021.Availableat: https://www.chcf.org/ wp-content/uploads/2021/12/EmergencyDepartmentsAlmanac2021. pdf.AccessedAugust21,2024.

12.IbarraA.CaliforniawantstoincreasepayforsomeMedi-Calproviders. Howitmighthelppatientsaccesscare.2023.Availableat: https://laist. com/news/health/california-wants-to-increase-pay-for-some-medi-calproviders-how-it-might-help-patients-access-care AccessedAugust21,2024.

13.WuFandDarracqMA.Comparingphysicianassistantandnurse practitionerpracticeinU.S.emergencydepartments,2010–2017. WestJEM. 2021;22(5):1150–5.

14.PreiksaitisC,KrzyzaniaketS,BowersK,etal.Characteristicsof emergencymedicineresidencyprogramswithunﬁlledpositionsinthe 2023match. AnnEmergMed. 2023;82(5):598–607.

AReviewoftheClinicalPresentation,Causes,andDiagnostic EvaluationofIncreasedIntracranialPressureinthe EmergencyDepartment

CristianaOlaru,MD,MSCE

SamLangberg,MD

NicoleStreiffMcCoin,MD

SectionEditors:RichardLucarelli,MD

OchsnerHealth,DepartmentofEmergencyMedicine,NewOrleans,Louisiana

Submissionhistory:SubmittedOctober8,2023;RevisionreceivedFebruary6,2024;AcceptedMay6,2024

ElectronicallypublishedSeptember6,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18500

Increasedintracranialpressure(ICP)isencounteredinnumeroustraumaticandnon-traumaticmedical situations,anditrequiresimmediaterecognitionandattention.Clinically,ICPtypicallypresentswitha headachethatismostsevereinthemorning,aggravatedbyValsalva-likemaneuvers,andassociated withnauseaorvomiting.Papilledemaisawell-recognizedsignofincreasedICP;however,emergency physiciansoften ﬁnditdifﬁculttovisualizetheopticdiscusingophthalmoscopyortoaccuratelyinterpret digitalfundusphotographswhenusinganon-mydriaticretinalcamera.

Emergencyultrasoundcanevaluatetheopticnervesheathdiameter(ONSD)andopticdiscelevation todeterminewhetherincreasedICPispresent,however,thestudieshavebeensmallwithdifferent deﬁnitionsandmeasurementsoftheONSD.TheONSDthresholdvaluesforincreasedICPhavebeen reportedanywherefrom4.8to6.3millimeters.

Neuroimagingisthenextstepintheevaluationofpatientswithpapilledemaorhighclinicalsuspicion ofincreasedICP,asitcanidentifymoststructuralcausesortypicalradiologicalpatternsofincreased ICP.NeuroradiographicsignsofincreasedICPcanbehelpfulinsuggestingidiopathicintracranial hypertension(IIH),especiallywhenpapilledemaisabsent.

Patientswithpapilledemaandnormalneuroimagingmayundergolumbarpunctureaspartoftheir clinicalworkup.Thecerebrospinal ﬂuid(CSF)openingpressureremainsoneofthemostimportant investigationstoestablishthediagnosisofIIH.ACSFevaluationisalsorequiredtoexcludeotheretiologies ofelevatedICPsuchasinfectious,inﬂammatory,andneoplasticmeningitis.InvasiveICPmeasurement remainsthestandardtomeasureandmonitorthiscondition.[WestJEmergMed.2024;25(6)1003–1010.]

INTRODUCTION

Increasedintracranialpressure(ICP),regardlessof etiology,isalife-threateningconditionthatrequiresprompt diagnosisandtreatment.Itcanleadtodecreasedcerebral perfusionpressurewithsubsequentcerebralischemiaand herniation,andthuspotentialdisabilityandincreased mortality.1 RecognitionofelevatedICPisofutmost importanceintheemergencydepartment(ED).Knowledge oftheclinicalpresentation(whichcanhelpdifferentiatenot onlybetweenmultiplecausesofheadacheandalteredmental statusbutalsobetweencausesofelevatedICP)and diagnosticoptionsandtheiraccuracyisparamountfor correctdiagnosisandrapidtreatment.

TheMonro-Kelliehypothesisstatesthatthesumofthe intracranialvolumeofblood,brain,cerebrospinal ﬂ uid (CSF),andothercomponents(eg,tumor,hematoma)is constant.Theskullisarigidcontainer;hence,anincreasein oneoftheintracranialcomponentswillcauseadecreasein thevolumeofoneormoreoftheothercomponents. Intracranialblood(especiallyinthevenouscompartment) andCSFarethetwocomponentswhosevolumecan adaptmosteasilytoaccommodateanincreaseinthe volumeofintracranialcontents.Whenthecompensatory capacityisexhausted,theICPbeginstorise,compromising cerebralperfusionandcausingcerebralischemiaor herniation.NormalICPis7–15millimetersofmercury

(mmHg)or10–20centimetersofwater(cmH 2O) inadults. 1, 2

CerebralperfusionpressureCPPðÞ = MeanarterialpressureMAPðÞ intracranialpressureICPðÞ

CAUSESANDCLINICALPRESENTATION

IncreasedICPiscausedbyavarietyofdiseaseprocesses suchasspace-occupyinglesions(eg,mass,hemorrhage); obstructivehydrocephalus;communicatinghydrocephalus (eg,inadequatereabsorptionofCSFsuchasseenin subarachnoidhemorrhagesecondarytohypersecretionof CSFand ﬁbrosisofarachnoidgranulations);venousoutﬂow obstruction(eg,cerebralvenoussinusthrombosis);diffuse cerebraledema(eg,vasogenic,asseenintumors;cytotoxic, asseenintraumaticbraininjuryorstroke;interstitial,asseen inhydrocephalusormeningitis;orosmotic,suchasseenin hyponatremia,diabeticketoacidosis);increasedCSF secretion(eg,choroidplexustumor);andidiopathiccauses (eg,idiopathicintracranialhypertension[IIH]).3,4,5

Thecombinationofheadache,papilledema,andvomiting isconsideredindicativeofincreasedICP,althoughthereisno consistentrelationbetweenseverityofsymptomsandthe degreeofelevatedICP.2 Headacheisacommoncomplaintin theED,representing2.6%ofEDvisits,andthesixthmost commonreasonforpresentationtotheED.6 Theheadache relatedtoincreasedICPistypicallyaglobalheadache commonlydescribedasthrobbingorbursting,ismostsevere inthemorning,oftenaggravatedbymaneuversthatincrease ICP(eg,Valsalva-likemaneuvers,coughing,sneezing, recumbency),andisassociatedwithnauseaorvomiting.3

OthersignsandsymptomsofincreasedICPincludechangesin behaviorsuchasirritabilityandrestlessness,visualchanges (eg,diplopia,visual fielddeficits),pupillarychanges(dilated unreactive,mid-position fixed,pinpointpupils),bilateral ptosis,impairedupwardgaze,focalneurologicdeficits, depressedconsciousness,seizures,andtheominous findingsof decorticateordecerebrateposturingandCushingtriad (bradycardia,hypertension,andrespiratorydepression).

Aprogressivedeteriorationinlevelofconsciousnesscan beseenwithworseningincreasedICP(exceptforIIH,which ischaracterizedbynormalmentalstatus).2 Brainherniation leadstofurtherbraininjuryandischemia,compressionof vesselsandcranialnerves,andobstructionofthenormal circulationofCSFproducinghydrocephalus.(See Figure1 fortypesofbrainherniation.)Owingtoitslocation, eachtypeofherniationisassociatedwithspeciﬁc neurologic ﬁndings.7

Subfalcineherniation,alsoknownasmidlineshiftor cingulateherniation,isthemostcommontypeofherniation. Itisusuallycausedbymasseffectthatpushestheipsilateral cingulategyrusdownandunderthefalxcerebri.The

Figure1. Typesofbrainherniation11:1)subfalcineherniation; 2)centraldescendingtranstentorialherniation;3)lateral descendingtranstentorialherniation;4)tonsillarherniation;and 5)ascendingcerebellartranstentorialherniation.

AdaptedfromUser:Delldot,CCBY-SA3.0 <http:// creativecommons.org/licenses/by-sa/3.0/>,version21:08, 5March2008viaWikimediaCommons.

quantiﬁcationofthemidlineshiftismadebymeasuringthe deviationoftheseptumpellucidumcomparedtomidlineat theleveloftheforamenofMonro.Thismeasurementisused forprognosiswithlessthan5millimeters(mm)ofdeviation associatedwithgoodprognosisandgreaterthan15mm associatedwithpoorprognosis.Itcanpresentwith hypobulia,apathy,andindifference.Iftheanteriorcerebral arteryiscompressed,itwillmanifestwithcontralateralor bilaterallegweaknessandacuteurinaryretention.7

Descendingtranstentorialherniationisthesecondmost commontypeofcerebralherniation.Itoccurswhenbrain tissueisdisplaceddownwardthroughthetentorialnotchand maybelateral(anteriorandposterior)orcentral.Lateral herniasinvolvethemedialtemporallobe;anteriorhernias involvetheuncus(alsocalleduncalherniation);and posteriorherniasinvolvetheparahippocampalgyrus.Inthe centralhernia,thereisdescentofthediencephalon, midbrain,andpons.7

Anteriordescendingtranstentorial(uncal)herniation leadstocompressionoftheparasympathetic fibersrunning withthethirdcranialnerve,causinganipsilateral fixedand dilatedpupil.Compressionoftheipsilateralcerebral pedunclewillcausecontralateralmotorparalysissincethe motortract fiberscrossbelowthislevel;however,the contralateralcerebralpedunclecanbecompressedagainst theedgeofthetentoriumcausingafalselocalizingsignwith ipsilateralhemiparesis.8 Posteriordescendingtranstentorial

herniationisduetoherniationoftheparahipoccampalgyrus, presentingwithsymptomsofParinaudsyndrome(vertical gazepalsy,lossofpupillaryreﬂextolightwithpreservation ofpupillaryconstrictionwithconvergence,uppereyelid retraction,convergence-retractionnystagmus).9,10 Central descendingtranstentorialherniationisduetoherniationof thethalamusandmidbrainthroughthetentorialnotchand themedullathoughtheforamenmagnum.Itcausesacute obstructivehydrocephalusandposteriorcerebralartery injury,clinicallypresentingwithagitationfollowedby obtundation,bilaterallypoorlyreactiveorpotentially midpoint ﬁxedpupils,andthendecorticatefollowedby decerebrateposturing,Cushingtriad,coma,anddeath.9

Ascendingcerebellartranstentorialherniationisdueto cerebellarherniationsuperiorlythroughthetentorialnotch. Itpresentswithsymptomsofpontomedullarycompression: obtundation;cardiorespiratoryinstability;severe bradycardia;arrhythmia;abdpinpointpupils.9 Tonsillar herniationinvolvesherniationofthetonsilsofthecerebellum throughtheforamenmagnumintotheupperspinalcanal, compressingthemedulla.Thismayresultin cardiorespiratoryimpairment,hypertension,highpulse pressure,andCheyne-Stokerespiration.Thecombinationof bradycardia,hypertension,andirregularrespirationsis knownasCushing’ sreﬂexandoccursinapproximatelyone thirdofcasesoftonsillarherniation.2 Itmayalsocause pinpointpupils, ﬂaccidparalysis,andsuddendeath.

Idiopathicintracranialhypertension,formallyknownas pseudotumorcerebriorbenignintracranialhypertension, deservesspecialattentionasithasauniquepresentation.Itis asyndromeofincreasedICPofunclearetiologythatoccurs mostofteninobesewomenofchildbearingage(averageage 28years).Itoccurslessofteninmen(approximately9%)who areusuallyobeseandaverage37yearsofageatdiagnosis.12

Itisusuallyadiagnosisofexclusionthatischaracterizedby signsandsymptomsofincreasedICP,normalmentalstatus, andabsenceoffocalneurologicsigns(althoughitcanbe associatedwithsixthandseventhnervepalsies).

NeuroimagingmightshowsignsofelevatedCSFpressure butwithoutobstructionordeformityoftheventriclesand withoutidentiﬁablecauseoftheincreasedICP.TheCSF evaluationwillhaveopeningpressuregreaterthan25 cmH2ObutwithnormalCSFcomposition.3

CommonsymptomsofIIHareheadache,visual disturbances,andpulsatiletinnitus.Themostcommon presentingsymptomofIIHisheadache,whichisfoundin 84%ofthepatients.Theheadacheisconstant,non-pulsating, exacerbatedbycoughingorValsalvamaneuver,andithasa progressivecourse.Thesecondmostfrequentsymptomof IIHisvisualdisturbancesuchasvariable,visual ﬁelddefects thatcommonlygounnoticedbythepatientuntilsevere; transientvisualobscurations(transientunilateralorbilateral visuallosslastinglessthanoneminuteandoftenprecipitated byposturalchanges,withfullrapidvisualrecoveryto

baseline);enlargingblindspots;diminishedvisualacuityin patientswithadvanceddisease;diplopia,especiallyinthe horizontalplanegenerallyduetosixthcranialnervepalsy;or blurryvisionduetoshorteningoftheglobesecondaryto increasedICP.Tinnitus,anothercommonsymptominIIH, ismoreoftenbilateral,pulsatile,synchronouswithheartrate, andcanoccurwithvariablefrequencyfromdailyto monthly.3,12 Idiopathicintracranialhypertensionis characterizedbynormalmentalstatus;however,itcancause disablingheadachesandblindness.

DIAGNOSTICEVALUATION

Papilledema

Thepresenceofopticnerveheadedema(ONHE)in patientswithheadachesignifiesasecondarycauseforthe headacheandtheneedforfurtherurgentevaluation.13 This conditioniscommonlyencounteredinpapilledema(optic discswellingduetoincreasedICP);opticneuropathy(optic neuritis,ischemicopticneuropathy);andpseudopapilledema (discelevationwithoutnerve fiberlayeredema).The distinctionbetweenthethreemajorcausesofdiscswellingis basedonhistory,eyeexaminationincludingfundoscopy,and ancillarytesting.Othercausesofopticdiscswellingare centralretinalveinocclusion,diabeticpapillopathy,uveitis, opticdisctumors,malignanthypertension,andopticnerve infiltration(suchasseeninsarcoidosis,lymphoma,and leukemia).14 Opticneuropathiesleadtoamoreseverevisual lossandareusuallysudden,unilateral,andassociatedwith afferentpupillarydefectandimpairedcolorvision.14

Pseudopapilledemaisassociatedwithopticnervevariants thatmimicpapilledemaophthalmoscopically,suchas congenitalabnormalities;crowdedhyperopicdisc;opticdisc hamartomas;oropticnerveheaddrusen.Visuallossmay occur,butitismoreindolent,painless,andfrequently unnoticedbythepatient.Pseudopapilledemaisstableover timecomparedtountreatedpapilledema,whichwillchange andprogressintime.Therearealsoophthalmoscopic findingsthatwillhelpdifferentiatepapilledemafrom pseudopapilledema.14 Opticdiscdrusenareacellular depositslocatedintheopticnervehead.Inchildren,theoptic discdrusenarenotcalcified;theyresemblepapilledemawith opticnerveheadswellingandcanbedifficulttodiagnoseon opthalmoscopy.15 Withage,theopticdiscdrusenbecome calcifiedandeasiertodiagnoseonophthalmoscopyand ultrasound,asopticdiscdrusenarehyperechoicwith posterioracousticshadowing.15

TheFrisénclassiﬁcationisthemostfrequentlyused papilledemagradingsystemanddescribesstagesofopticdisc swelling(grades0–5);however,ithaspoorinter-rater reliability.Therefore,moredescriptiveterminologyisoften usedtodescribepapilledema(eg,mildvshighgrade).14,16,17 Papilledemaisusuallybilateralandsymmetrical;however,it canbeasymmetricaland,rarely,itcanbeunilateraloreven absent.12,18 Papilledemaisthoughttobesecondarytoeither

axoplasmicstasisthatcausesaxonalswellingordueto enlargementofthesubarachnoidspace.Usually,the developmentofpapilledemarequiresatleast1–5daysof persistentlyelevatedICP;however,ithasalsobeenfoundto developrapidly,inhours,insubarachnoidand intraparenchymalhemorrhages.IftheelevatedICPis treated,papilledemausuallyresolvesoverweeks tomonths.14

Non-expertcliniciansoften finditdifficulttoproperly viewtheopticdiscusingophthalmoscopy.19 PhaseIofthe FOTO-EDstudyfoundthatdirectophthalmoscopywas rarelyandinadequatelyperformedbyemergencyphysicians (EP)inalargeacademicmedicalcenterwhereEPsfailedto diagnoseanycasesofoptic-nerveheadedemausingdirect ophthalmoscopy.20 PhaseIIoftheFOTO-EDstudyfound thatnon-mydriaticretinalphotographyintheEDwas superiortodirectophthalmoscopyperformedbyEPs; however,EPsdonotcommonlyperformthisandidentified only16of35relevant findings(sensitivity46%).21 Sachdeva etal13 performedacross-sectionalanalysisofpatientswith ONHEintheprospectiveFOTO-EDstudyandfoundthat 2.6%ofpatientspresentingtotheEDwithachiefcomplaint ofheadache,acutevisionloss,focalneurologicdeficit,ora diastolicbloodpressure ≥120mmHghadONHE.Themost common finaldiagnoseswereIIH(19/37),CSFshunt malfunction/infection(3/37),andopticneuritis(3/37),thus reiteratingtheimportanceofocularfundusexaminationin thesepatients.13

EmergencyUltrasoundEvaluationforIncreased IntracranialPressure

Emergencyultrasoundisaneasy-to-use,noninvasive methodofincreasedICPassessmentbyevaluatingtheoptic nervesheathdiameter(ONSD)andtheopticdiscelevation (ODE).Theopticnervecanbethoughtofasanoutpouching ofintactbraintissuewiththeintraorbitalcomponentCSF and ﬂuctuatesinsizebasedonchangesinICP.IncreasedICP causesenlargementofthesubarachnoidspaceandincrease oftheONSD.Thebulbousportionoftheopticnerve, approximately3mmposteriortotheglobe,appearstobethe mostdistensibleandsensitivetochangesinICP.22 On ultrasound,theglobeappearsasaround,anechoicstructure. Theopticnervepresentsasahypoechoicstructureposterior totheglobe(see Figure2).

Opticdiscelevation

Opticdiscelevationreferstotheheightofelevationofthe opticdiscfromthelaminacribrosa(theareathroughthe sclerawheretheopticnerveaxonspass).23 Themeasurement isperformedwiththeopticnerveinthehorizontalplane,and theviewwiththemaximumdiscelevationisselected.Disc elevationismeasuredfromtheuppermostpartoftheswollen disctothestronglyreﬂectinglinerepresentingthelamina cribrosa(see Figure3).24 Teismannetal25 determinedthata

Figure3. Opticdiscelevation(ODE)measurementbetweenthe topoftheswollenopticdiscandthelaminacribrosa.Inthis ﬁgure, ODEof0.13cmsuggestsopticdiscedema.

cutoffvalueof0.6mmforopticdiscelevation,asmeasured byultrasound,predictedthepresenceofopticdiscedema notedonfundoscopicexamwithasensitivityof82%(95% conﬁdenceinterval[CI]48–98%)andaspeciﬁcityof76%(95% CI50–93%).Acutoffvalueof1.0mmyieldedasensitivityof

Figure2. Opticnerveappearanceonultrasound.

73%(95%CI39–94%)andspeciﬁcityof100%(95%CI 81–100%).Inthisstudy,mostpatientshadIIHcausingdisc swellingduetoelevatedICP;however,discswellingcanalso befoundinpatientswithmultiplesclerosis,inﬁltrative processessuchassarcoidosisorlymphoma,infectionsdirectly affectingtheopticnerve,andmicrovascularinfarctioncaused bymalignantsystemichypertension.

Opticnervesheathdiameter

TheONSDismeasured3mmdeeptotheglobewhereit appearstobethemostdistensibleandsensitivetochangesin ICP.22 Threepossiblepositionsfordepthmarkershavebeen describedinstudies:1)locationwhereimaginarynerve midlineintersectsthecontouroftheretina;2)hyperechoic reﬂectioncorrespondingtothelaminacribrosa;and3)topof thehypoechoicstructurecorrespondingtotheopticnerve(see Figure4).However,thesediscrepanciesdidnotaffectONSD values,mostlikelybecausethedistancebetweenthedifferent anatomicallandmarksusedislessthan1mmresultingin comparableONSDvalues.Stevensetal26 recommendsusing thepapillaasreferenceforthe3-mmdepthassessment.

Theopticnerveitselfhasadiameterof3mm,andtheoptic nervesheathhasathicknessofapproximately1mm.From theinsideout,thesheathconsistsofpiamater,the subarachnoidspace,thearachnoidmater,andthedura mater.Whenstudieswerereviewed,twotypesofimageswith differentechoiccharacteristicsoftheopticnervesheathwere described.Onegroupofimagesshowedtwohyperechoic, stripedbandsbetweenthehyperechoicretrobulbarfat(see Figure5a),whiletheothergroupshowedasingledarklinear structuresurroundedbyhyperechoicretrobulbarfat (see Figure5b).26

Figure5. (a)Opticnerve(betweenCcalipers)asoneanechoic bandsurroundedbytwohyperechoicstripedbands(betweenC andBcalipersoneachsideoftheopticnerve)betweenthe hyperechoicretrobulbarfat(Acalipers).(b)Opticnervewithone singleanechoicbandrepresentingtheopticnerveanditssheath, Dcalipersattheborderbetweentheopticnervesheathandthe retrobulbarfat.

Thestripedbandshavebeeninterpretedtorepresenteither thepia,bothpiaandduramater,orthesubarachnoidspace. Stevensetal26 foundthatCcaliperpositionwastheleast sensitivetochangesintheICPandcorrespondstotheoutline oftheopticnerveitself.Bcalipermeasurementincreasesin patientswithelevatedICP;hence,itincorporatesthe arachnoidspace,anditlikelycorrespondstotheouteredge ofthesubarachnoidspace.Bcaliperpositionandcalipers measuringasingledarklinearstructure(Dcalipers)were foundtobeequallysensitivetochangesinICPandwere recommendedasreliableONSDmeasurements.26

TheONSDthresholdvaluesthatoptimizedsensitivityand speciﬁcityforelevatedICPvariedinstudiesfrom4.8mm (Rajajeeetal27)to6.3mm(delSaz-Saucedoetal28).The studiesonONSDhavebeensmallandheterogeneous,and thesestudieshavedonethefollowing:

• useddifferentdeﬁnitionsandmeasurements ofONSD26

• useddifferentplanestomeasureONSD:only transverseplanes27–32;transverseandsagittal planes33–35;transverseandcoronal36

Figure4.

• predominantlyusedmeanvaluesoftheONSD measurementsobtainedfrombotheyes,withAgrawal etal36 identifyingthehighesttransversemeasurement andRajajeeetal27 studyingeachindividual transversemeasurement

• enrolleddifferentpatientpopulations,withmultiple studiesenrollingneurocriticalcarepatientswith traumaticandnon-traumaticbraininjury,andsome studiesenrollingclinicpatientswithIIH28,32

• useddifferentconﬁrmationsofelevatedICPincluding computedtomography(CT),37,38 lumbarpuncture (LP),28,32 orexternalventriculardrain(EVD)and intraparenchymalcatheter33,36

• deﬁnedelevatedICPaseither20mmHgor25cmH2O (20mmHg = 27cmH2O39).

Inameta-analysisincluding18prospectivestudiesthat hadICPmeasuredbyEVDorintraparenchymalcatheter (LPmeasurementsexcluded),Aletrebyetal40 demonstrated thatONSDshowedreasonableaccuracyindiagnosingraised ICP.ThehighestsensitivitywasachievedusinganONSD cutoffofmorethan6mm.40 WhileONSDof <5mm correlateswithnormalICP,thecutoffONSDforelevated ICPvariesfromstudytostudywith >6mmdemonstratinga highsensitivityforelevatedICP.AnODE >0.6mmcanalso behelpfulindeterminingpapilledema;however,ODEcanbe seeninmultipleothermedicalconditions.WhileONSDand ODEcanbehelpfulinscreeningforpapilledema,larger studieswithstandardizedONSDmeasurementsanda consistentICPcutoffarestillneeded.Itisalsoknownthat papilledemacanbeasymmetrical;hence,eacheyeONSD measurementsshouldbestudiedindependently.

Neuroimaging

IntheevaluationofapatientwithincreasedICP, neuroimaginghastwomainpurposes:toexcludestructural causes;andtoidentifytypicalradiographicpatternsof elevatedICP.Non-enhancedCTiscommonlyusedasthe ﬁrsttestforevaluationofsecondaryheadachesandaltered mentalstatusintheemergencysetting,anditisthestandard imagingmodalityinacuteheadtrauma.Computed tomographyiswidelyavailable,hasshorteracquisitiontime, andislessexpensivethanbrainmagneticresonanceimaging (MRI).Intheemergencysetting,CTisregularlyperformed toreadilyidentifyconditionsthatrequiresurgical intervention.7 ACTcanidentifybrainedema,tumors, hydrocephalus,intracranialhemorrhage,andsignsof cerebralherniation.ACTmayalsodetectsignsofincreased ICP,suchasventricularorsulcaleffacement,compressionof thebasalCSFcisterns,herniation,ormidlineshift.However, ICPcanbeelevatedeveninthesettingofanormalCT. ComparedtoCT,MRIprovidesbettertissue characterization,especiallyforposteriorfossadisease,andis requiredforevaluationofunderlyingbrainlesions.Brain

MRIwithcontrastcanidentifymoststructuralcausesof increasedICP,andmagneticresonancevenography(MRV) withcontrastwillexcludecerebralvenoussinusthrombosis.3 InIIH,brainMRIandMRVareimportanttoexclude secondarycausesofelevatedICP.Additionally,emptysella turcica,posterior flatteningoftheglobe,enlargementofthe periopticsubarachnoidspace,opticnervemidportion tortuosity,hyperintensityoftheopticnerveandopticdisc, andbilateraltransversesinusstenosishavebeenfoundtobe associatedwithIIH;andthesensitivityandspecificity improvewithacombinationoftheseneuroimaging signs.12,41 Slit-likeventricles,tightsubarachnoidspace,and inferiorpositionofthecerebellartonsilsshowslowsensitivity butgoodspecificityof90–97%forIIH.12

IntracranialPressureMeasurement

Patientswithpapilledemaandnormalneuroimaging couldundergoLPaspartoftheirclinicalworkuptomeasure theICPinsearchforetiologiessuchasIIH.TheCSFopening pressureremainsoneofthemostimportant ﬁndingsto establishadiagnosisofIIH.TheCSFpressuremayvary considerablywithtime;thus,thepossibilitytorepeatLPor performamoreinvasiveICPmonitoringshouldbe consideredifclinicalsuspicionishigh.12 ACSFevaluationis alsorequiredtoexcludeotheretiologiesofelevatedICPsuch asinfectious,inﬂammatory,orneoplasticmeningitis.3 The LPisobtainedinlateralrecumbentposition,whiletheCSF openingpressureshouldbemeasuredwiththelegsextended, headinneutralposition,andthepatientbreathingnormally. ThenormalCSFopeningpressureinadultsis10–20cmH2O andisconsideredhighifgreaterthan25cmH2O.ACSF openingpressureof20–25cmH2Oisconsideredborderline, anditisinterpretedintheclinicalcontext.3

Idiopathicintracranialhemorrhagewithoutpapilledema iswellrecognized.Thesepatientstendtohaveloweropening pressurelevelsthanthosewhopresentwithpapilledema.In theabsenceofpapilledema,theothercriteriashouldbemet, withtheadditionalsymptomofunilateralorbilateral abducensnervepalsies.Ifthereisnoabducensnervepalsy, threeofthefollowingneuroimagingcriteriashouldbemet:1) emptysella;2) ﬂatteningoftheposterioraspectoftheglobe; 3)distentionofthesubarachnoidspacewithorwithouta tortuousopticnerve;and4)transversesinusstenosis.A normalopeningpressureleveldoesnotexcludethediagnosis ofIIHwhenthepatienthasothertypicalsymptoms. Conversely,anincreasedopeningpressurelevelwithout appropriatesymptomsshouldnotbeinterpretedasIIH.42

ALPcarriesnoriskofherniationifthereisnobrainshift, whetherCSFpressureisraisedornotandwhether papilledemaispresentornot.ACTisusedtodiagnosebrain shiftseeninspace-occupyinglesionsanddiffusebrain swelling.FindingsofbrainshiftonCTmaydemonstratethe following:displacementofbrainstructures;lossof differentiationbetweengrayandwhitematter; ﬂattenedgyri;

effacementofCSFspaces(suchassulci,Sylvian ﬁssures, ventriclesandbasalcisterns);dilatedventricles (inobstructivepathology);andinadvancedstagesherniation (displacementofbrainfromoneintracranial compartmenttoanother).43

AninvasiveICPmeasurementdevicesuchasexternal ventriculardrainorintraparenchymaldeviceremainsthe standardtodeterminethepressureinthecranialvault.The necessityforICPmeasurementiseitherdeducedbya pathologicalCToraconsciousnessimpairmentscoreof ≤8ontheGlasgowComaScale.44

CONCLUSION

Increasedintracranialpressurerepresentsalifethreateningdiagnosis.Idiopathicintracranialhemorrhage, whilenotlife-threatening,cancauseirreversiblevisualloss anddisablingheadaches.Clinicalpresentationrequires immediaterecognitionandinvestigation.Whilepapilledema isimportanttodiagnoseintheED,emergencyphysicians havefounditdifficulttodiagnosebydirect ophthalmoscopy.20 Ocularultrasound(opticnervesheath diameterandopticdiscelevation)iscommonlyusedto screenforincreasedICP;however,thereisneedforfurther researchusingstandardizedONSDmeasurements. Neuroimagingremainsthe firststepininvestigatingelevated ICPasitexcludesstructuralcausesandidentifiestypical radiologicalpatternsofelevatedICP.InvasiveICP measurementremainsthestandardtomeasureand monitorICP.

ACKNOWLEDGMENTS

Drs.OlaruandMcCoinwereaffiliatedwithOchsner Healthduringthecreationofthispaper.Currently,Dr. OlaruisaffiliatedwithUniversityofColoradoinAurora, Colorado,andDr.McCoiniscurrentlyaffiliatedwith UniversityofKentuckyCollegeofMedicinein Lexington,Kentucky.

AddressforCorrespondence:SamLangberg,MD,Divisionof EmergencyUltrasound,OchsnerHealth,DepartmentofEmergency Medicine,1514JeffersonHwy,NewOrleans,LA70121.Email: samlangberg@gmail.com

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.CanacN,JalaleddiniK,ThorpeSG,etal.Review:pathophysiologyof intracranialhypertensionandnoninvasiveintracranialpressure monitoring. FluidsBarriersCNS. 2020;17(1):40.

2.DunnLT.Raisedintracranialpressure. JNeurolNeurosurgPsychiatry. 2002;73(suppl1):i23–7.

3.ThurtellMJ.Idiopathicintracranialhypertension. Continuum(Minneap Minn). 2019;25(5):1289–1309.

4.ChenS,LuoJ,ReisC,etal.Hydrocephalusaftersubarachnoid hemorrhage:pathophysiology,diagnosis,andtreatment. BiomedRes Int. 2017;2017:8584753.

5.NehringSM,TadiP,TennyS.Cerebraledema.In: StatPearls[Internet] TreasureIsland,FL:StatPearlsPublishing,2022.Availableat: https:// www.ncbi.nlm.nih.gov/books/NBK537272/.AccessedSeptember8, 2023.

6.CairnsCandKangK.NationalHospitalAmbulatoryMedicalCare Survey:2019emergencydepartmentsummarytables.2022. Availableat: https://www.cdc.gov/nchs/data/nhamcs/web_ tables/2019-nhamcs-ed-web-tables-508.pdf.AccessedSeptember8, 2023.

7.RiverosGilardiB,Mu ˜ nozL´opezJI,HernándezVillegasAC,etal.Types ofcerebralherniationandtheirimagingfeatures. RadioGraphics. 2019;39(6):1598–610.

8.NayakLandDeangelisLM.Clinicalfeaturesofbraintumors.In: Blue BooksofNeurology,vol36.(54–70).Amsterdam,TheNetherlands: Elsevier,2010.

9.TadevosyanAandKornbluthJ.Brainherniationandintracranial hypertension. NeurolClin. 2021;39(2):293–318.

10.PrakashE,KuhtHJ,HarieaswarS,etal.Dorsalmidbrain(Parinaud) syndrome. PractNeurol. 2021;21(6):550–1.

11.Delldot.Brainherniationtypes.2008.Availableat: https://upload. wikimedia.org/wikipedia/commons/archive/c/cb/20090916112024% 21Brain_herniation_types.svg.AccessedDecember29,2023.

12.ToscanoS,LoFermoS,ReggioE,etal.Anupdateonidiopathic intracranialhypertensioninadults:alookatpathophysiology, diagnosticapproachandmanagement. JNeurol. 2021;268(9):3249–68.

13.SachdevaV,VasseneixC,HageR,etal.Opticnerveheadedema amongpatientspresentingtotheemergencydepartment. Neurology. 2018;90(5):e373–9.

14.LiuGT,VolpeNJ,GalettaSL. Liu,Volpe,andGaletta’s Neuro-Ophthalmology:DiagnosisandManagement, 3rded. Amsterdam,TheNetherlands:Elsevier,2019.

15.HamannS,MalmqvistL,CostelloF.Opticdiscdrusen:understanding anoldproblemfromanewperspective. ActaOphthalmol(Copenh). 2018;96(7):673–84.

16.SinclairAJ,BurdonMA,NightingalePG,etal.Ratingpapilloedema:an evaluationoftheFrisénclassiﬁcationinidiopathicintracranial hypertension. JNeurol. 2012;259(7):1406–12.

17.FrisénL.Swellingoftheopticnervehead:astagingscheme. J Neurol NeurosurgPsychiatry. 1982;45(1):13–8.

18.SwinkinE,JabehdarMaralaniP,SundaramAN.Unilateralpapilledema inidiopathicintracranialhypertension:acaseseries. CanJNeurolSci. 2022;49(2):278–81.

19.KardonR.Opticalcoherencetomographyinpapilledema:whatamI missing? JNeuroophthalmol. 2014;34Suppl:S10–7.

20.BruceBB,LamirelC,BiousseV,etal.Feasibilityofnonmydriaticocular fundusphotographyintheemergencydepartment:phaseIoftheFOTOEDstudy:feasibilityofnonmydriaticocularfundusphotography. Acad EmergMed. 2011;18(9):928–33.

21.BruceBB,ThulasiP,FraserCL,etal.Diagnosticaccuracyand useofnonmydriaticocularfundusphotographybyemergency physicians:phaseIIoftheFOTO-EDstudy. AnnEmergMed. 2013;62(1):28–33.e1.

22.RichardsE,MunakomiS,MathewD.Opticnervesheathultrasound.In: StatPearls[Internet].TreasureIsland,FL:StatPearlsPublishing,2022. Availableat: https://www.ncbi.nlm.nih.gov/books/NBK554479/ AccessedSeptember8,2023.

23.ElkingtonAR,InmanCBE,SteartPV,etal.Thestructureofthelamina cribrosaofthehumaneye:animmunocytochemicalandelectron microscopicalstudy. Eye. 1990;4(1):42–57.

24.ErtlM,GamulescuMA,SchlachetzkiF.(2012).Applicationoforbital sonographyinneurology.In:ThoirsK(Ed), Sonography.InTechOpen, 2012.Availableat: http://www.intechopen.com/books/sonography/ application-of-orbital-sonography-in-neurology AccessedSeptember8,2023.

25.TeismannN,LenaghanP,NolanR,etal.Point-of-careocular ultrasoundtodetectopticdiscswelling. AcadEmergMed. 2013;20(9):920–5.

26.StevensRRF,GommerED,AriesMJH,etal.Opticnervesheath diameterassessmentbyneurosonology:areviewofmethodologic discrepancies. JNeuroimaging. 2021;31(5):814–25.

27.RajajeeV,VanamanM,FletcherJJ,etal.Opticnerveultrasound forthedetectionofraisedintracranialpressure. NeurocritCare. 2011;15(3):506–15.

28.delSaz-SaucedoP,Redondo-GonzálezO,Mateu-MateuÁ,etal. Sonographicassessmentoftheopticnervesheathdiameterinthe diagnosisofidiopathicintracranialhypertension. JNeurolSci. 2016;361:122–7.

29.KimberlyHH,ShahS,MarillK,etal.Correlationofopticnervesheath diameterwithdirectmeasurementofintracranialpressure. AcadEmerg Med. 2008;15(2):201–4.

30.JeonJP,LeeSU,KimSE,etal.Correlationofopticnervesheath diameterwithdirectlymeasuredintracranialpressureinKoreanadults usingbedsideultrasonography. PLoSOne. 2017;12(9):e0183170.

31.RafﬁzMandAbdullahJM.Opticnervesheathdiametermeasurement:a meansofdetectingraisedICPinadulttraumaticandnon-traumatic neurosurgicalpatients. AmJEmergMed. 2017;35(1):150–3.

32.KorsbækJJ,HagenSM,SchytzHW,etal.Transorbitalsonography:a non-invasivebedsidescreeningtoolfordetectionofpseudotumor cerebrisyndrome. Cephalalgia. 2022;42(11-12):1116–26.

33.GeeraertsT,MerceronS,BenhamouD,etal.Non-invasiveassessment ofintracranialpressureusingocularsonographyinneurocriticalcare patients. IntensiveCareMed. 2008;34(11):2062–7.

34.MorettiRandPizziB.Opticnerveultrasoundfordetectionofintracranial hypertensioninintracranialhemorrhagepatients:conﬁrmationof previous ﬁndingsinadifferentpatientpopulation. JNeurosurg Anesthesiol. 2009;21(1):16–20.

35.RobbaC,CardimD,TajsicT,etal.Ultrasoundnon-invasive measurementofintracranialpressureinneurointensivecare:a prospectiveobservationalstudy. PLoSMed. 2017;14(7):e1002356.

36.AgrawalA,ChengR,TangJ,etal.Comparisonoftwotechniquesto measureopticnervesheathdiameterinpatientsatriskforincreased intracranialpressure. CritCareMed. 2019;47(6):e495–501.

37.BlaivasM,TheodoroD,SierzenskiPR.Elevatedintracranialpressure detectedbybedsideemergencyultrasonographyoftheopticnerve sheath. AcadEmergMed. 2003;10(4):376–81.

38.TayalVS,NeulanderM,NortonHJ,etal.Emergencydepartment sonographicmeasurementofopticnervesheathdiametertodetect ﬁndingsofincreasedintracranialpressureinadultheadinjurypatients. AnnEmergMed. 2007;49(4):508–14.

39.HeifermanDM,JackMM,JohnsonKD,etal.Mercuryvs.water:an analysisofmulti-institutionalsurveydataassessingintracranial pressureunitofmeasureawareness. WorldNeurosurg. 2019;129:e657–62.

40.AletrebyW,AlharthyA,BrindleyPG,etal.Opticnervesheathdiameter ultrasoundforraisedintracranialpressure:aliteraturereviewand meta-analysisofitsdiagnosticaccuracy. JUltrasoundMed. 2022;41(3):585–95.

41.BarkatullahAF,LeishangthemL,MossHE.MRI ﬁndingsasmarkers ofidiopathicintracranialhypertension. CurrOpinNeurol. 2021;34(1):75–83.

42.GuryildirimM,KontzialisM,OzenM,etal.Acuteheadacheinthe emergencysetting. Radiographics. 2019;39(6):1739–59.

43.vanCrevelH,HijdraA,deGansJ.Lumbarpunctureandtherisk ofherniation:whenshouldwe ﬁrstperformCT? JNeurol. 2002;249(2):129–37.

44.SchürerL,WolfS,LumentaCB.1074208Intracranialhypertension.In: LumentaCB,DiRoccoC,HaaseJ,MooijJJA(Eds.), Neurosurgery (605–10).BerlinHeidelberg,Germany:Springer,2010.

BarrierstoAdoptionofaChild-AbuseClinicalDecisionSupport SysteminEmergencyDepartments

AlannaC.Peterson,MD*

DonaldM.Yealy,MD†

EmilyHeineman,MA‡

RachelP.Berger,MD,MPH‡§

*UniversityofPittsburghMedicalCenter,Pittsburgh,Pennsylvania

† UniversityofPittsburgh,DepartmentofEmergencyMedicine, Pittsburgh,Pennsylvania

‡ UniversityofPittsburgh,UPMCChildren’sHospitalofPittsburgh,DivisionofChild Advocacy,DepartmentofPediatrics,Pittsburgh,Pennsylvania

§ PennsylvaniaOfﬁceofChildren,YouthandFamilies,Harrisburg,Pennsylvania

SectionEditor:NikhilGoyal,MD

Submissionhistory:SubmittedOctober17,2023;RevisionreceivedJuly15,2024;AcceptedAugust12,2024

ElectronicallypublishedOctober14,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18501

Introduction: Childabuseisaleadingcauseofmorbidityandmortalityinchildren.Therateofmissed childabuseingeneralemergencydepartments(ED),where85%ofchildrenareevaluated,ishigherthan inpediatricEDs.Wesoughttoevaluatetheimpactofanelectronichealthrecord(EHR)-embeddedchildabuseclinicaldecisionsupportsystem(CA-CDSS)intheidentiﬁcationandevaluationofchild maltreatmentinanetworkofEDsthreeyearsafterimplementation.

Methods: Weanonymouslysurveyedall196EDattendingphysiciansandadvancedpractice practitioners(APP)intheUniversityofPittsburghMedicalCenternetwork.Thesurveyevaluated practitionerawarenessof,attitudestoward,andchangesinclinicalpracticepromptedbytheCA-CDSS. Wealsoassessedpractitionerrecognitionandevaluationofsentinelinjuries.

Results: Ofthe71practitioners(36%)whorespondedtothesurvey,75%feltthetoolraisedchildabuse awareness,and72%hadaface-to-facediscussionwiththechild’snurseafterreceivingaCA-CDSS alert.AmongAPPs,72%consultedwiththeattendingphysicianafterreceivinganalert.Many practitionerswereunawareofatleastonefunctionoftheCA-CDSS;38%didnotknowwhocompleted thechildabusescreen(CAS);54%wereunawarethattheycouldviewtheresultsoftheCASintheEHR, and69%didnotrecognizetheclinicaldecisionsupportdashboardicon.Slightlyover20%of respondentsfeltthattheCA-CDSSlimitedautonomy;and4.5%disagreedwiththerecommendationsin thephysicalabuseorderset,whichreﬂectsAmericanAcademyofPediatrics(AAP)guidelines.Greater than90%ofrespondentscorrectlyidentiﬁedanintraoralinjuryandtorsobruiseinaninfantassentinel injuriesrequiringanevaluationforabuse.

Conclusion: Achild-abuseclinicaldecisionsupportsystemembeddedintheelectronichealthrecord wasassociatedwithcommunicationamongpractitionersandwasoverallperceivedasimprovingchild abuseawarenessinoursystem.Practitionerscorrectlyrecognizedinjuriesconcerningforabuse. BarrierstoimprovingidentiﬁcationandevaluationofabuseincludegapsinknowledgeabouttheCACDSSandthepresenceofpractitionerswhodisagreewiththeAAPrecommendationsforphysicalabuse evaluationand/orfeltthatclinicaldecisionsupportingenerallimitedtheirclinicalautonomy.[WestJ EmergMed.2024;25(6)1011–1019.]

INTRODUCTION

Childmaltreatmentcausessigniﬁcantmorbidityand mortalityintheUnitedStates,especiallyforchildrenunder fouryearsofage.Overthreemillionchildreneachyearare reportedtochildprotectiveservices(CPS),and1,600 childrendieatthehandsofcaregiversduetomaltreatment,1 anumbergreaterthanchildrenwhodiedfromCOVID-19 duringthe ﬁrst2½yearsofthepandemic.2 Thenumberof childrendyingfrommaltreatmenthasbeenincreasing;there wasa10%increaseinfatalitiesfrom2016–2021.1,3 Between 20–25%ofchildrenwhoareultimatelydiagnosedwith physicalchildabusehavebeenpreviouslyseenbyamedical practitionerwhofailedtoidentifytheabuse.4–7 Appropriate recognitionandevaluationofphysicalchildabuseingeneral emergencydepartments(ED),wheremostchildrenreceive emergencycare,iscrucial.

Toimprovethequalityofidentiﬁcation,evaluation,and reportingofchildmaltreatmentintheUniversityof PittsburghMedicalCenter(UPMC)generalEDs,we developedanddeployedachild-abuseclinicaldecision supportsystem(CA-CDSS)inourelectronichealthrecord (EHR)startingin2016.8–10 Atthattimetherewere13 generalEDswiththehospitalmanagementsoftwareCerner (CernerCorp,KansasCity,MO)intheUPMCsystem;they wentlivewiththeCA-CDSSsystembetween January–March2016.Asthehospitalsystemacquired additionalEDs(oneeachinJune2017andApril2019,and fourinSeptember2019),theywereaddedtotheCA-CDSS.

Priortothego-liveateachsite,trainingoccurredforboth EDnursesandpractitioners.Nursescompletedan interactiveonlinelearningmodule,whichremainspartofthe onboardingprocessforEDnursesandhasbecomean educationrequirementeverytwoyears.Practitioner educationwasdonethroughtheEDmedicaldirectors.Prior tothego-liveateachhospital,eachEDmedicaldirector receivedanonboardingpacketthatincludedgeneralchild abuseeducation,screenshotsofallpartsoftheCA-CDSS, caseexamples,andawayforpractitionerstoreachoutwith questions.EachEDmedicaldirectoralsometwithoneofthe authors(RB)whoreviewedtheonboardingpacketand answeredquestions.Themedicaldirectorswere,and continuetobe,responsiblefordisseminatingeducationto theirpractitioners.Inadditiontotheinitialtraining,ongoing trainingincludesfeedbacktopractitionersaboutcasesfrom themedicaldirectorofeachEDsite-speciﬁctrainingsat standingpractitionermeetingsinindividualEDsandbimonthlysystemwideconferencecalls,whichusecase examplesasaspringboardfordiscussingspeciﬁcchildabuserelatedtopics.Thesecallsprovidecontinuingmedical educationcreditforpractitioners.

Thefeaturesofthistoolincludeasetoftriggersincludinga childabusescreen(CAS)completedbytheprimarynurse,an alertthatpractitionersreceivewhenapatienthastriggered theCA-CDSS,andaphysicalabuseordersettoassist

PopulationHealthResearchCapsule

Whatdowealreadyknowaboutthisissue? Earlydiagnosisofchildabuseiscriticalto decreasemorbidityandmortality.Abuse identi ﬁ cationingeneralEDsmaybeassisted byclinicaldecisionsupport.

Whatwastheresearchquestion?

Whatarethebene ﬁ tsofandchallengesto sustainabilityofachildabuseclinicaldecision supportsystem(CA-CDSS)embeddedinthe electronichealthrecord(EHR)?

Whatwasthemajor ﬁndingofthestudy?

Three-quartersofpractitionersreportedthe CA-CDSSincreasedchildabuseawareness andpromptedinterdisciplinaryinteractions.

Howdoesthisimprovepopulationhealth?

UsinganEHR-embeddedCA-CDSSmaybe oneapproachtoimprovingchildabuse awarenessingeneralEDs,therebydecreasing abuse-relatedmorbidityandmortality.

practitionersinorderingthecorrecttestingbasedonpatient ageandinjury.8,9 Inaddition,triggeringtheCA-CDSS resultsinaniconappearingonthemainEDdashboardnext tothepatient’sname.Whenprovidingfeedbackto practitionersaboutcasesinwhichthephysicalabuseorder setwasn’tusedwhenitwasindicated,oneoftheco-authors (AP),whoisalsothedirectorofoneofthegeneralEDs, notedthatsomepractitionersreportedthattheydidnotagree withtherecommendationsintheordersetandpreferredto useclinicaljudgment.

Wesoughttounderstandthebarrierstocompliancewith theordersetrecommendations,assesstheimpactofthe CA-CDSS,andidentifyopportunitiestoimprovethe CA-CDSSwiththegoalofincreasingengagementwiththe CA-CDSSoverall.

METHODS Setting

The19generalEDsintheUPMChospitalsystemoperate inurbanandruralsettingsandincludecommunityhospitals andacademiccenters.Theprimaryacademiccentersarein thecityofPittsburgh,PA.TheremainderareEDsafﬁliated withcommunityhospitalsacrossmuchofPennsylvaniaand withindividualsitesinNewYork.Annualpractitioner turnoveratthese19generalEDsforfull-timeemployees averages6.8%forAPPsand3.6%forphysicians.Thereisa

totalof~30,000EDvisitsforchildren <13yearsofage(the ageincludedintheCA-CDSS)atthe19EDsannually;the proportionofallEDvisitsinvolvingchildrenrangesfrom 1–3%attheacademicsitesandupto12%inthe communitysites.

Survey

InFebruary2020,ourteam(ageneralemergency physicianwhoisthedirectorofoneofthehospitalsystem’ s EDs[AP]andachildabusepediatrician[RB]fromthe afﬁliatedchildren’shospitalthatwasnotoneofthe19 includedhospitals)emailedasurveytoall196attending physiciansandadvancedpracticepractitioners(APP)atthe 19UPMCEDs.Theemailprovidedananonymouslinktoa 25-question,web-basedsurvey(QualtricsLLC,Seattle,WA) thatusedskiplogic,meaningcliniciansreceivedonly questionsthatwererelevantbasedonpreviousresponses (AppendixA).Self-reporteddemographicdataincluded yearsinpractice,hospitalafﬁliation(s),andpractitionertype (physicianorAPP).Thesurveyaimedtoassessthe practitioner’s1)knowledgeabouttheCA-CDSSandits associatedfunctionality;2)engagementwithandattitudes towardtheCA-CDSS;3)recognitionofsentinelinjuries minorinjuriesthatnecessitateanevaluationforphysical abuse;and4)reasonsfornotusingthephysicalabuseorder setevenwhenitwasindicated.Thesurveywasdesignedso thatpractitionerswouldlearnabouttheCA-CDSSasthey completedthequestions.

StatisticalAnalysis/Measures

Weuseddescriptiveanalysestomeasuretheproportionof surveyscompleted,knowledgeofpractitionersaboutthe CA-CDSS,attitudestowardtheCA-CDSS,recognitionof

Demographiccharacteristics ofsurveyrespondents

Practitionertype

injuriesthatshouldraiseconcernforphysicalabuse,and barrierstoevaluatingandreportingsuspectedabuse.

EthicalConsideration/Approval

TheUPMCQualityImprovementCommitteeapproved thisproject.Therewasnoformalethicsreview,andno potentialconﬂictsofinterestwereidentiﬁed.

RESULTS

ResponseRateandPractitionerCharacteristics

Therewasa43%(84/196)initialresponserate,with13 surveysexcludedforlackofcompleteness,leaving71surveys (36%)foranalysis.Ofthe13incompletesurveys,one practitionerwasn’teligibleand10oftheremaining12 incompletesurveyshadfewerthan35%ofthequestions answered.Asaresult,wechosetoexcludethementirely. Mostrespondentswerephysicianswhoworkedin communityEDsandhadmorethan15yearsof experience(Table1).

PractitionerKnowledgeAbouttheCA-CDSSandIts AssociatedFunctionality

Thechildabusescreen(CAS)

Ofthe71respondents,27(38%)didnotknowwho completestheCAS,and54%(38/71)wereunawarethatthey couldseethecompletedCAS(vssimplybeingalertedwhenit waspositive).

Thealert

Thesameproportionofpractitioners(27/71)didnotrecall everseeinganalert,and69%(49/71)ofpractitionersdidnot knowthatthelightbulbiconontheEDdashboardmeant thatthepatienthadtriggeredtheCA-CDSS.

#(%)ofrespondentswithcompleted surveys(N = 71)

Characteristicsofsurveyed population(n = 196)

Yearsofexperience >15yearsinpractice30(42.3%)Notavailable^ 6–15yearsinpractice27(38.0%)Notavailable 0–5yearsinpractice14(19.7%)Notavailable

*Twodeclinedtostate.

^Yearsofexperiencewascollectedaspartofthesurveyand,therefore,wasnotavailableforpractitionerswhodidnotcompletethesurvey. APP,advancedpracticepractitioner.

Table1. Demographiccharacteristicsofsurveyrespondents.

PractitionerEngagementwithandAttitudesToward theCA-CDSS

Thealert

Ofthepractitionerswhorememberedseeingthe alert,68%(30/44)reportedthattheyalwaysapproachedthe child’snurseforfurtherdetails,and86%(12/14)ofthe remainingpractitionersreportedthattheysometimes approachedthenurse;twocouldnotrecallwhether theyhaddoneso.Ofthe11practitionerswhowere APPsandrecalledseeingthealert,78%(8/11)reportedthat theyalwaysapproachedtheattendingphysician todiscussthecase,andtheotherthreereportedthey sometimesdid.

Theemergencydepartmentchildphysicalabuseorderset

Forty-twopercent(30/71)ofrespondentsreportedhaving usedthephysicalabuseorderset.Ofthe58%whodidnot reportusingit,34%(14/41)indicatedtheywereunawareofit, 54%(22/41)believedtheordersetwasnotrelevantforthe patient(s)theyweretreating,and5%(2/41)wereunableto ﬁndit.Onepractitionermadeabroadcommentaboutnot usinganyordersetsbecausehewantedto “usemybrain” and notfollowrecommendations.Half(33/66)ofrespondents indicatedtheyagreedwiththerecommendationscontained intheorderset,45%(30/66)wereneutral,and4.5%(3/66) disagreedwiththerecommendations.

Attitudes

OverallattitudesabouttheCA-CDSSwerepositive,with 79%ofpractitionersagreeingorstronglyagreeingwiththe statement, “TheCA-CDSSincreasesmyawarenessofthe potentialriskforchildabuse” (Figure1a).Twenty-two percentofrespondentsfeltthattheCA-CDSSalertand physicalabuseordersetlimitedtheirabilitytomake independentdecisions(Figure1b).Morethan75%of respondentsfeltthatthealertwasclear(Figure2a), thatthealertandorderset ﬁtwellintopractitioner workﬂow(Figure2b),andthatitsaved time(Figure2c).

Recognitionofinjuriesthatdoanddonotnecessitatea physicalabuseevaluation

Case1describeda13-day-oldwithasubconjunctival hemorrhageclearlydocumentedatbirth.Thiscasewasan exampleofaninfantwhodidnotneedachildabuse evaluation.Almost40%ofrespondents(26/71)incorrectly statedthatthisrequiredachildabuseevaluation.

Cases2and3describeda4-month-oldand2-month-old withanintraorallacerationandtorsobruise,respectively. Thesecaseswereusedtodemonstrateinfantswhorequirea childabuseevaluation.Inthesescenarios,91%(56/61)of practitionersincase2and97%(69/71)incase3correctly notedtheneedforachildabuseevaluation.

Figure1. Practitionerattitudesaboutthealert’seffectonclinical decision-making:a)Practitionerresponsestothestatement, “TheCA-CDSSincreasesmyawarenessofthepotentialriskfor childabuse,” (n = 70);b)Practitionerresponsestothestatement, “TheCA-CDSSpop-upalertandEDphysicalabuseordersetlimit myabilitytomakeindependentdecisions,” (n = 68)andc) Practitionerresponsestothestatement, “Iagreewiththe suggestedevaluations/workupintheEDphysicalabuseorder set.” (n = 66).

CA-CDSS,child-abuseclinicaldecisionsupportsystem; ED,emergencydepartment.

Barrierstoevaluatingandreportingsuspected physicalabuse

Ofrespondents,89%(60/71)expresseduncertainty regardingtheirabilitytorecognizechildabuse,with52% reportingatleastmoderateamountofuncertaintyin recognition(Figure3a).Whenaskedaboutpursuingthe appropriateevaluationforchildabuse,65%(46/71) expresseduncertaintyastowhichtestswereindicated (Figure3b).Lackofasocialworkerorancillarysupportin theEDwasidentifiedasabarrierby54%(38/70)of respondents.Toomuchtimeneededfortheworkupwas identifiedasabarrierfor16%(11/71)ofrespondents.Lastly, concernaboutbeingcalledtocourttotestifywasidentifiedas abarrierby10%(7/71)ofrespondents.

Free-textresponsesregardingthechildabuse-clinical decisionsupportsystem

The ﬁnalquestionallowedforrespondentstoprovidefreetextcommentsand/orsuggestionsabouttheCA-CDSS.

Figure2. Practitionerattitudesaboutthealertandorderset:a) Likertscalepractitionerresponsestothestatement “Thepop-up alertIreceivefromtheCA-CDSSisclearlyworded,” reportedasa percentageof70respondents(1declinedtostate);b)Likertscale practitionerresponsestothestatement “UsingtheEDphysical abuseorderset ﬁtswellinmyclinicalworkﬂow,” Reportedasa percentageof69respondents(twodeclinedtostate);andc) Likertscalepractitionerresponsestothestatement “TheED physicalabuseordersetsavestimewhenevaluatingpatients,” reportedasapercentageof68respondents.(Threedeclined tostate.)

CA-CDSS,Child-abuseclinicaldecisionsupportsystem; ED,emergencydepartment.

Atotalof18(25%)respondentsenteredafree-textresponse. Acommonresponserelatedtothesubtletiesinthesetypesof casesandthedesiretohavetheabilitytoconsultapediatric and/orchildabusespecialistwhilethepatientisintheED.As discussedbelow,practitionersdohavethisaccessbutwere unawareofit.

SometimesIhavetroublechoosinganordersetbecause thesituationdoesnot ﬁtinthelistprovided .

[Practitoner]uncertaintyaroundthespecifics/subtleties relatedtoaspecificcasecanbeabigfactorinchoosing tofullyutilizetheordersetand/orto fileareport.I think[practitioners]/patients/familieswouldgreatly benefitbymoredirectinvolvementwiththechild advocacyteam[while]thepatientisintheemergency department.

Figure3. Practitionerself-reportingaboutbarriersinchildabuse recognitionandevaluation:a)Lackofcertaintyaboutwhentobe concernedforabuse;b)Lackofcertaintyaboutwhattestsare indicated;c)Lackofsocialworkerorancillarysupport;andd) concernaboutbeingcalledtocourt.

Raiseawarenessto[practitioners]sothattheycan consultpediatricspecialistsincludingpeds[emergency] physiciansforfurtherassistancewithsuspectedchild abusecases.

AsecondthemewasnotrelatedspeciﬁcallytotheCACDSSbuttothemoregeneralissuesofpop-upfatigue” (ie, alertfatigue):

Morepop-upscauseclickfatigue.Thishasdestroyed medicine.

Wesufferfrompop-upfatigueandthiscanjustbe onemore.

Athirdsetofcommentswererelatedtothedesiretolearn moreabouttheCA-CDSSitself:

Iwasnotawareofthistool,sopleasesendoutanemail describingit.

Makecliniciansawareofthistool.

morewebinarsorcollaborationwith expertswould begreat.

Ifthereareroutinefreerecordedlecturesforusonclassic regionalcases ::: .Iwouldbeinterestedintakingthem :::

DISCUSSION

Thisisthe ﬁrstpapertoevaluatethelong-termeffectofan EHR-embeddedCA-CDSSonpatientcareinanetworkof generalEDs.Themostencouragingobservationisthatthe CA-CDSSappearstofacilitateinterdisciplinary communicationbetweennurses,physicians,andAPPs.One oftheconcernsabouttransitioningtheCASfromapaper form thewayinwhichitwasoriginallystudiedand validated11 toanelectronicformatwasthat multidisciplinarycommunicationwouldnolongerbe requiredand,therefore,wouldnotoccur.

Thefactthat91%and97%ofpractitioners,respectively, correctlyidentiﬁedanintraoralinjuryandabruiseassentinel injuriesandrecognizedtheneedforachildabuseevaluation isencouraginggiventheliteraturesuggestingthat practitionersoftendonotrecognizethesemoresubtleforms ofabuse.12–14 Itisnotpossibletoknowwhetherthis knowledgeisrelatedtothepresenceoftheCA-CDSSorthe practitionereducationthathassupplementedtheCA-CDSS.

Over90%ofpractitionersstatedthatconcernaboutbeing calledtocourthadlittleornoimpactontheirdecisionto report/evaluateforabuse.Thisisencouraging;inalandmark studybyFlahertyandcolleagues,15 “Clinicianspentmany hoursincourttestifying” wasoneofthepractitioner characteristicsidentiﬁedinasigniﬁcantproportionofthe practitionerswhodidnotreporttoCPSdespitehaving concernformaltreatment.Eachoftheauthorsofthecurrent manuscripthaspersonallyspokenwithpractitionerswho havesaidthattheydonotwanttogetinvolvedinthelegal system.Ourdatasuggeststhat,overall,thisconcernisnot drivingdecision-makingaboutwhetherornottomakea reporttoCPSinourgeneralEDs.

Ourresultsalsounderscoretheneedforusabilityofthe CA-CDSS;practitionersfeltthealertandorderssuggested wereclearanduseful, fitwellintoworkflow,andsavedtime. Theimportanceofusabilitycannotbeoveremphasized:if thisispoorthenitisunlikelythatpractitionerswillusethe tool.16 However,asotherresponsestooursurveyseemto demonstrate,usabilityisnecessary,butnotsufficient,for

practitionerengagementwiththeCA-CDSS.Thenumberof practitionersunawareofdifferentaspectsoftheCA-CDSS includingtheicon,thealert,andtheordersetdemonstrates thechallengesineducationaboutalow-frequencyeventfor anygivenpractitionerinalargehospitalsystem,evenwitha relativelylowpractitioner-turnoverrate.

ImprovingtheknowledgeofpractitionersabouttheCACDSSitselfislikelytobechallenging.Therearemultiple studiesevaluatingtheeffectivenessofdifferenttraining modalities,includinghands-ontraininginalaboratory setting,requiredtraining,useof “ superusers ” toassistothers inlearning,anduseof “just-in-time” training.17–20 Butall theseapproachesarechallengingwhentheyaimtoaddressa relativelyrareevent.Whilechildabuseisacommoncauseof morbidityandmortalityoverall,itisnotcommonlyseenor recognizedbymostindividualpractitioners.Multiple trainingapproachesaswellasindividualizedfollow-upand ongoingtraininghaveallbeenshowntosustainengagement; theseareapproacheswehaveusedandcontinuetouse.17

Despitetheseapproaches,manyofthefree-textresponses notedanunawarenessoftheexistingresources(eg,accessto pediatricexpertsandongoingavailableeducation).We hypothesizethatsomeofthislackofawarenessmaybe relatedtotheincreaseintheuseoflocumtenensand temporarilyorcasuallyemployedpractitioners,whichwas rarepriortoCOVID-19butincreasedsigniﬁcantlyduring thepandemicandhascontinued.Thesepractitionersrarelyif everattendtheeducationsessionsanddonotreceive onboardinglikeotherpractitioners,bothofwhichcanlimit therealutilityofresourcespositionedtoaidwithknowledge andcare.

Itisconcerningthatoneofthemostcommonfree-text responsesrelatedtothedesiretohavereal-timeaccesstoa pediatricand/orchildabusespecialistwhilethepatientisin theED.EmergencycliniciansinallthegeneralEDsinthe hospitalsystemalreadyhavethisaccess.Thereisaphone numbertheycancall24/7,whichprovidesconsultationwith apediatricianatthetertiary-carepediatrichospitalinthe hospitalsystem.Ifthepediatricpractitionerisunableto answerachildabuse-specificquestion,thechildabuse specialistoncallispaged.Thelackofknowledgeaboutthis phonenumberreflectsamoregenerallackofknowledge abouthospitalresources.Asaresultofthissurvey,anemail waswidelydistributedtomakepractitionersawareofthis phonenumber,anditisnowincludedon flowsheetsin multipleEDs.

Manyofthefree-textresponsestargetedCDStoolsin generalratherthanspeciﬁcallytheCA-CDSS.Alertfatigue wasmentionedinthefree-textresponsesregardingtheCDS toolandareconsistentwithpriorstudies,which demonstratedthatinterruptivealerts(eg,pop-ups)adversely affectpractitioneruseofrecommendationsfromCDS.21–24 Bothfree-textcommentsrelatedtothealert(“Morepop-ups causeclickfatigue.Thishasdestroyedmedicine,” and “We

sufferfrompop-upfatigueandthiscanjustbeonemore”)are consistentwiththebroaderissueofalertfatigue.Whilethe lowtriggerratefortheCA-CDSSishighlyunlikelytobea majorcontributingfactortoalertfatigue,itmayexacerbate pre-existingfrustrationwiththealerts/pop-upsingeneral.

Reportedsuccessfulinterventionstocombatalertfatigue includethepersistentpresenceofthepop-uponthechart untilitisacknowledged,asopposedtohavingmultiplealerts forthesameclinicalconcern.24 Whileapersistentalert wasn ’tapossibilityinourEHRatthetimetheCA-CDSS wasdeveloped,wedesignedthesystemsothateach practitioneronlyreceivesonealertinresponsetothespeciﬁc concernoftryingtoalleviatealertfatigue.Interestingly, someoftheearlyfeedbackaboutthesystemwasthatsome practitionerswantedtobealertedrepeatedlyuntilthey decidedwhethertoevaluateforabuse,atwhichpointthey wantedtobeabletosilencethealert.Itisnotpossibleforthe currentEHRsystemtocustomizealertingrulesfordifferent practitioners;whendesigninganygivenCDSsystem,a decisionmustbemadeaboutthetimingandfrequencyof alertsthatappliestoallpractitioners.Addressingtheissueof alertfatigueischallengingbecauseitisthesumofallalerts ratherthanthealertfromanysinglesystemthatimpacts adoptionratesforallCDS.Asaresult,developingasolution requiresahospital/hospitalsystemtoholisticallyevaluateall theCDSsystemsinuse.

Multipleconversationswithpractitionersdisplayedlittle agreementaboutwhenintheworkﬂowandhowoftenthe alertsshouldbeprovided.Somepractitionerswantedto knowat ﬁrstchartopenthattheyshouldbeconcernedabout abuse,sothattheyhavethisinmindwhentheyexaminethe childandspeakwiththefamily,whileotherpractitioners wanttheinformationlaterinthevisitwhentheyare formulatingadifferentialdiagnosis.Somepractitioners wantedtobealertedonlyonce,whileotherspreferredtobe alertedrepeatedlyuntiltheydecidedwhethertoevaluate forabuse.

The findingthatover20%ofpractitionersperceiveCDS asathreattophysicianautonomylikelyhasasignificant impactonengagementandacceptanceofanyCDS.The concernaboutCDSimpactingpractitionerautonomyisnot specifictochildabuseCDS;ratheritisamajorbarrierto CDSingeneral.21,25–27 Interestingly,andperhaps surprisingly,thecharacteristicsofthe15clinicianswhofelt thatCDSwasathreattophysicianautonomywerenot differentfromtheclinicianswhodidnothavethissentiment intermsoftheirresponsetothequestionsabouthowoften lackofcertaintyaboutwhentobeconcernedaboutabuseor lackofcertaintyaboutwhatteststodoinfluencedtheir decisionaboutwhethertodoaphysicalchildabuseworkup. Neitherdidtheydifferfromotherpractitionersintermsof theproportionwhocorrectlyansweredthethreescenarios relatedtosentinelinjuries.Whileitisdifficulttointerpretthis datagiventhesmallnumbers,itsuggeststhatthereisaneed

tobetterunderstandhowthesepractitionerswouldpreferto receiveassistancemakinghigh-quality,evidence-based decisionssincetheyrecognizethattheyareuncertainand thatthisimpactstheirclinicaldecision-making,butthey don’tseeCDSasasolution.

Behavioraleconomics(BE),anevolving fieldrootedin economicandpsychology,maybeoneapproachtoenhance physicianengagementwiththeCA-CDSS.Behavioral economicsisbasedonrecognitionthathumansarenot rationaldecisionmakersandrarelybehaveasthe conventionaleconomicstheorywouldpredict.Interventions informedbyBEattempttochangephysicianpracticeusinga “nudge,” aninterventionthatpredictablychangeshuman behaviorwithoutsignificantlylimitingfreechoiceor changing financialincentives.Changingdefaultsettingsand providingsocialreferencepoints(eg,peercomparison)are mostconsistentlyeffectiveinterventionsinimproving physicianpracticeasitrelatestofollowingevidence-based practice.28–30 AsitrelatestoaCA-CDSS,BEmaybeableto beusedtonudgepractitionerstofollowAAP recommendationsforphysicalabuse,forexample,by providingfeedbackabouttheirperformancecomparedto theirpeers.28–30 Importantly,BEfocusesonthesubsetof practitionerswhoareamenableto “nudging,”31 whichis generallyonlyasubsetofpractitioners.Onewouldnot expectapractitionerwhohasanegativeviewofCDS(ie, feelsthatitlimitsautonomyanddoesnotimprovepatient care)tobeamenabletonudging,butitmaystilloffersome abilitytoimproveengagement.

Perhapsthemostconcerningandconfusing ﬁndingwas howmanypractitionerseitherhaveaneutralornegative opinionoftheevidence-basedAAPrecommendationsfor childphysicalabuseevaluation.Thisobservationoccurred alongside65%expressinguncertaintyabouthowtoevaluate forphysicalabuseandmorethan90%recognizingthat sentinelinjuriesrequireaphysicalabuseevaluation. Requiringausertoselectareasonfornotfollowing recommendationsprovidedwithinthealert(eg,tofollowthe recommendationsinthephysicalabuseorderset)can increasecompliancewithrecommendations32 andmay provideinsightintotheseseeminglyinconsistentresponses. Implementationofarequiredresponseneedstobeweighed againstthepossibilityofgeneratingnegativeattitudes towardthetoolandperceivedimpairmentofworkﬂow. AnotherwaytopotentiallyincreaseadherencetoAAP recommendationsistoensurethatpractitionersunderstand howtheCA-CDSSfunctionsandensurethatpractitioners understandthesourcefortheelectronictool recommendation27;wecurrentlydothisduringbi-monthly educationsessionswithpractitioners.

LIMITATIONS

Thereareseverallimitationstothestudy.Theresponse ratewasrelativelylow,andthosewhochosetorespondmay

notberepresentativeofallpractitioners.Theonly demographicdataavailableforallpractitionerswaswhether theywereaphysicianorAPPandthetypeofhospitalthey practicedin;forthesecharacteristics,respondentsandnonrespondentslookedsimilar.Wefeltitwasimportantforthe surveytobeanonymoussothatrespondentswouldbe comfortableprovidinghonestfeedback;thisapproach meanswecouldnottargetnon-respondentstoimprovethe responserate.Itisalsopossiblethattherespondentshaving particularlystrongfeelings positiveornegative aboutthe CA-CDSSweremostlikelytorespondtothesurvey.Because ofthelownumberofpractitionerswhorespondedtothe surveyfromanysinglehospitalandbecausepractitionerscan workatmorethanonehospitalwithinthe19-hospital network,itwasnotpossibletodeterminewhethertherewere site-specificvariationsintheopinionsofpractitionersabout theCA-CDSS.Finally,theresponsestothesurveymaynot reflectactualpracticeandinsteadmayreflectwhat practitionersknowtheyshoulddo.Forexample, practitionersmaysaytheyspeakwiththenursewhenachild hasapositiveCAS,butitisnotpossibletoknowifthis actuallyoccurs.

CONCLUSION

Insummary,ourdatasuggeststhatachild-abuseclinical decisionsupportsystemembeddedintheelectronichealth recordhasyieldedpositiveresultsinbothinterdisciplinary communicationandpractitionerattitudestowardthetool, includingperceivingthetoolasincreasingchildabuse awareness.However,thereremaingapsinknowledgeofthe CA-CDSSfunctionalityandincompliancewiththe recommendations.Commentssuggestthatpractitioner dislikeofCDStoolsingeneral,andspecificallyalerts deliveredinpop-upform,maycontributetopooradherence. WhilelimitationsoftheEHRlimitthetypeofalert, CA-CDSSeducationaleffortscouldbeaugmentedto specificallyaddressperceivedbarrierstoautonomyand possiblytoincludebehavioraleconomicstechniques, suchaspeercomparisonortestimonials,33–35 toimprove compliancewithAmericanAcademyofPediatrics recommendations.Furtherresearchshouldfocusonthe effectivenessoftheseinterventionsaswecontinuetoimprove careforarareeventthatcarriessignificantmorbidity andmortality.

AddressforCorrespondence:RachelBergerMD,MPH,UPMC Children’sHospitalofPittsburgh,DivisionofChildAdvocacy, DepartmentofPediatrics,4427PennAve.,Pittsburgh,PA15224. Email: rachel.berger@chp.edu

relationshipswithanycompaniesthatarerelevanttothisstudy.The statementsinthisarticlearesolelytheresponsibilityoftheauthors anddonotnecessarilyrepresenttheviewsofthePennsylvania OfﬁceofChildren,YouthandFamilies.Therearenoconﬂictsof interestorsourcesoffundingtodeclare.

REFERENCES

1.U.S.DepartmentofHealthandHumanServices,Administrationfor ChildrenandFamilies,AdministrationonChildren,YouthandFamilies, Children’sBureau.2023.Childmaltreatment2021.Availableat: https:// www.acf.hhs.gov/cb/report/child-maltreatment-2021 AccessedAugust5,2024.

2.CentersforDiseaseControlandPrevention.2021.Deathsinvolving coronavirusdisease2019(COVID-19)amongages0–18yearsinthe UnitedStates.Secondarydeathsinvolvingcoronavirusdisease2019 (COVID-19)amongages0–18yearsintheUnitedStates.Availableat: http://data.cdc.gov/browse.AccessedAugust5,2024.

3.U.S.DepartmentofHealthandHumanServices,Administrationfor ChildrenandFamilies,AdministrationonChildren,YouthandFamilies, Children’sBureau.2018.Childmaltreatment2016.Availableat: https:// www.acf.hhs.gov/cb/report/child-maltreatment-2016 AccessedAugust5,2024.

4.RavichandiranN,SchuhS,BejukM,etal.Delayedidentiﬁcationof pediatricabuse-relatedfractures. Pediatrics. 2010;125(1):60–6.

5.JennyC,HymelKP,RitzenA,etal.Analysisofmissedcasesofabusive headtrauma. JAMA. 1999;281(7):621–6.

6.ThorpeEL,ZuckerbraunNS,WolfordJE,etal.Missedopportunitiesto diagnosechildphysicalabuse. PediatrEmergCare. 2014;30(11):771–6.

7.LetsonMM,CooperJN,DeansKJ,etal.Prioropportunitiestoidentify abuseinchildrenwithabusiveheadtrauma. ChildAbuseNegl. 2016;60:36–45.

8.Rumball-SmithJ,FromkinJ,RosenthalB,etal.Implementation ofroutineelectronichealthrecord-basedchildabusescreening ingeneralemergencydepartments. ChildAbuseNegl. 2018;85:58–67.

9.RosenthalB,SkrbinJ,FromkinJ,etal.Integrationofphysicalabuse clinicaldecisionsupportat2generalemergencydepartments. JAm MedInformAssoc. 2019;26(10):1020–9.

10.Patient-CenteredOutcomesResearchInstitute(PCORI).2020. Expandingtheuseofachildabusealertsystemtoimprovedetection andreporting.Availableat: https://www.pcori.org/research-results/ 2017/expanding-use-child-abuse-alert-system-improve-detectionand-reporting.AccessedAugust5,2024.

11.LouwersEC,AffourtitMJ,MollHA,etal.Screeningforchildabuse atemergencydepartments:asystematicreview. ArchDisChild. 2010;95(3):214–8.

12.TrokelM,WadimmbaA,GrifﬁthJ,etal.Variationinthediagnosisofchild abuseinseverelyinjuredinfants. Pediatrics. 2006;117(3):722–8.

13.PierceMC,KaczorK,AckerD,etal.History,injury,andpsychosocial riskfactorcommonalitiesamongcasesoffatalandnear-fatalphysical childabuse. ChildAbuseNegl. 2017;69:263–77.

14.TiyyaguraG,GawelM,KozielJR,etal.Barriersandfacilitatorsto detectingchildabuseandneglectingeneralemergencydepartments. AnnEmergMed. 2015;66(5):447–54.

15.FlahertyEG,SegeRD,GrifﬁthJ,etal.Fromsuspicionofphysicalchild abusetoreporting:primarycarecliniciandecision-making. Pediatrics. 2008;122(3):611–9.

16.BatesDW,KupermanGJ,WangS,etal.Tencommandmentsfor effectiveclinicaldecisionsupport:makingthepracticeofevidencebasedmedicineareality. JAmMedInformAssoc. 2003;10(6):523–30.

17.KealeyE,Leckman-WestinE,FinnertyMT.Impactoffourtraining conditionsonphysicianuseofaweb-basedclinicaldecisionsupport system. ArtifIntellMed. 2013;59(1):39–44.

18.KoskelaT,SandstromS,MakinenJ,etal.Userperspectivesonan electronicdecision-supporttoolperformingcomprehensivemedication reviews:afocusgroupstudywithphysiciansandnurses. BMCMed InformDecisMak. 2016;16:6.

19.FossumM,EhnforsM,FruhlingA,etal.Theexperiencesofusinga computerizeddecisionsupportsystem. NI2012(2012). 2012;113_5: eCollection2012.

20.FossumM,EhnforsM,FruhlingA,etal.Anevaluationoftheusabilityofa computerizeddecisionsupportsystemfornursinghomes. ApplClin Inform. 2011;2(4):420–36.

21.LugtenbergM,WeeninkJW,vanderWeijdenT,etal.Implementationof multiple-domaincoveringcomputerizeddecisionsupportsystemsin primarycare:afocusgroupstudyonperceivedbarriers. BMCMed InformDecisMak. 2015;15:82.

22.MillerA,MoonB,AndersS,etal.Integratingcomputerizedclinical decisionsupportsystemsintoclinicalwork:ameta-synthesisof qualitativeresearch. IntJMedInform. 2015;84(12):1009–18.

23.GurwitzJH,FieldTS,RochonP,etal.Effectofcomputerizedprovider orderentrywithclinicaldecisionsupportonadversedrugeventsinthe long-termcaresetting. JAmGeriatrSoc. 2008;56(12):2225–33.

24.Scheepers-HoeksAM,GroulsRJ,NeefC,etal.Physicians’ responses toclinicaldecisionsupportonanintensivecareunit–comparisonoffour differentalertingmethods. ArtifIntellMed. 2013;59(1):33–8.

25.VanDortBA,ZhengWY,BaysariMT.Prescriberperceptionsof medication-relatedcomputerizeddecisionsupportsystemsinhospitals: asynthesisofqualitativeresearch. IntJMedInform. 2019;129:285–95.

26.EsmaeilzadehP,SambasivanM,KumarN,etal.Adoptionofclinical decisionsupportsystemsinadevelopingcountry:antecedentsand outcomesofphysician’sthreattoperceivedprofessionalautonomy. IntJ MedInform. 2015;84(8):548–60.

27.LiberatiEG,RuggieroF,GaluppoL,etal.Whathinderstheuptakeof computerizeddecisionsupportsystemsinhospitals?Aqualitativestudy andframeworkforimplementation. ImplementSci. 2017;12(1):113.

28.StevensJ.Topicalreview:Behavioraleconomicsasapromising frameworkforpromotingtreatmentadherencetopediatricregimens. J PediatrPsychol 2014;39(10):1097–103.

29.StevensJandBerlanED.Applyingprinciplesfrombehavioral economicstopromotelong-actingreversiblecontraceptive(LARC) methods. PerspectSexualReprodHealth. 2014;46(3):165–70.

30.WangSYandGroeneO.Theeffectivenessofbehavioraleconomicsinformedinterventionsonphysicianbehavioralchange:asystematic literaturereview. PLoSOne. 2020;15(6):e0234149.

31.PatelMS,VolppKG,AschDA.Nudgeunitstoimprovethedeliveryof healthcare. NEnglJMed. 2018;378(3):214–6.

32.RoshanovPS,FernandesN,WilczynskiJM,etal.Featuresofeffective computerisedclinicaldecisionsupportsystems:meta-regressionof162 randomisedtrials. BMJ. 2013;346:f657.

33.JabbariJ,RollS,BufeS,etal. “Takemywordforit”:grouptextsand testimonialsenhancestateandfederalstudentaidapplications. JBehav ExpEcon. 2022;98:101864.

34.Apolinario-HagenJ,FritscheL,WoppererJ,etal.Investigatingthe persuasiveeffectsoftestimonialsontheacceptanceofdigitalstress managementtrainingsamonguniversitystudentsandunderlying mechanisms:arandomizedcontrolledtrial. FrontPsychol. 2021;12:738950.

35.ShenJ,PangS,SchwebelDC.Evaluationofadrowningprevention programbasedontestimonialvideos:arandomizedcontrolledtrial. J PediatrPsychol. 2016;41(5):555–65.

COMMENTARY

ReframingChildProtectioninEmergencyMedicine

JosephP.Shapiro,MD* GenevievePreer,MD†‡

CarolineJ.Kistin,MD,MSc§

*Children’sNationalHospital,DepartmentofEmergencyMedicineandTrauma, WashingtonDC

† BostonUniversityChobanian&AvedisianSchoolofMedicine, Boston,Massachusetts

‡ BostonMedicalCenter,Boston,Massachusetts

§ BrownUniversity,SchoolofPublicHealth,HassenfeldChildHealthandInnovation Institute,DepartmentofHealthServices,Policy,andPractice, Providence,RhodeIsland

SectionEditor:MuhammadWaseem,MD

Submissionhistory:SubmittedSeptember26,2023;RevisionreceivedApril11,2024;AcceptedMay10,2024

ElectronicallypublishedOctober1,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.18481

ChildmaltreatmentremainsaconcerningsourceofmorbidityandmortalityintheUnitedStates,where morethan600,000childrenarevictimsofabuseeachyear,withwell-described,long-term consequencesforphysicalandmentalhealth.

However,theUSchildwelfaresystemischaracterizedbysystemicracismandinequity.Blackand NativeAmericanchildrenaremorelikelytobeevaluatedandreportedforsuspectedabusedespite evidencethatracedoesnotindependentlychangetheirriskofbeingabused.Oncereportedtochild protectiveservices(CPS),thesechildrenaremorelikelytoberemovedfromtheirhomesandlesslikely tobereunitedwiththeirfamiliesthanWhitechildren.

Muchoftheinequityinthissystemstartsatthefrontdoor,whereagrowingbodyofresearch demonstratesthatbiasregularlyinﬁltratesdecision-makingintheinitialclinicalevaluationand managementofsuspectedabuse.Minoritychildrenpresentingtoemergencydepartments(ED)are morelikelytoreceivediagnostictestingandaremorelikelytobereferredtoCPS.

Inthiseditorial,wearguefortheapplicationofanequitylenstochildprotectionintheED.We discusshowemergencyphysicianscanbalanceeffortstoprotectchildrenfromabusewiththe imperativetoprotectchildrenandfamiliesfromtheharmsofaninequitablechildwelfaresystem.Our discussionconcludeswithconcreterecommendationsforemergencyclinicianstoparticipateinactive biasmitigationandthoughtfullynavigatetheirresponsibilitiesasmandatedreporters.[WestJEmerg Med.2024;25(6)1020–1024.]

Childmaltreatmentremainsaconcerningcauseof morbidityandmortalityintheUnitedStates,wheregreater than600,000childrenarevictimsofsubstantiatedabuseeach yearwithlong-termconsequencesforphysicalandmental health.1 TheUSchildwelfaresystem,however,is characterizedbymarkedinequity.2,3 BlackandNative Americanchildrenaremorelikelytobeevaluatedand reportedforsuspectedabusedespiteevidencethatracedoes notindependentlychangetheirriskofbeingabused.4,5

Inyearspast,typicalclinicaldecision-makingmighthave dictatedthatevenasmallsuspicionofchildabusemeantan automaticreporttochildprotectiveservices(CPS).However,

withthegrowingrecognitionoftheracialinequitywithinthe childwelfaresystemanditsadverseeffectsoncommunitiesof color,emergencyphysiciansnowmustreconsidertheir approach.6–8 Asfrontlinecarecliniciansforinjuredchildren, emergencyphysiciansshouldseektostrikeabalancebetween safeguardingchildrenfromabuseandshieldingfamiliesfrom unnecessaryexposuretothechildwelfaresystem.

Inasystematicreview,Cenatetalcompiled36studiesof proportionalityinchildwelfaresystemsandfoundthat27of thesestudiesshowBlackyouthareover-representedinthese systems.2 Thisdisproportionalityispresentateachphaseof thechildprotectionprocessbeginningwithreporting.

Nationally,Blackchildrenrepresent15%ofthegeneral populationbutaccountfor25%ofCPSinvestigations.9 Separatestate-levelanalysesofpublicrecordsin Washington,Colorado,Missouri,Wisconsin,Texas,and NorthCarolinashowthatBlackchildrenareatleasttwiceas likelyasWhitechildrentobereportedtoCPS.10–15

Onceaninvestigationisinitiated,BlackandNative Americanchildrenaremorelikelytoberemovedfromtheir homesandlesslikelytobereunitedwiththeirfamiliesthan Whitechildren.16 Inarecordreviewof4,000children referredtoCPSinCalifornia,Luetalfoundthatin comparisontoWhitechildren,Blackchildrenwere23.4% morelikelytobeplacedinout-of-homecareand11.7%less likelytobereuniﬁedwiththeirfamiliesifremovedfromthe home.17 AsimilarreviewbytheCenterfortheStudyof SocialPolicyfoundthatBlackvictimsofmaltreatmentwere 36%morelikelythanWhitevictimstoberemovedfrom theirhomes.18

TheburdenofchildwelfareinvolvementonBlackfamilies andcommunitiesissubstantial.ResearchbyKimetal indicatesthatover50%ofBlackchildrenwillundergoaCPS investigationwithintheirfamilybeforeturning18.19 SociologistDorothyRobertsandacoalitionofacademics andcommunityadvocatesleadanintellectualmovement thatcharacterizesthechildwelfaresystemasaformof “familypolicing.” Througharobustbodyofqualitativeand narrativeresearchtheyhighlighthowthesystemdisrupts parentalauthority,hinderschildren’srelationship development,fostersdistrustamongneighbors,andharms maternalmentalhealth.7,20–21 Thereislittleevidencethat contactwiththechildwelfaresystemcanimproveoutcomes forchildrenorfamilies.22

Theroleofemergencyphysiciansasmandatedreporters andprovidersoffrontlinecaretoinjuredchildrenmakesthem defactogatekeeperstothisunequalsystem.MostCPS referralsinhospitalsareforpatientswhosecarestartsinthe emergencydepartment(ED).23 Bothgeneralemergency physiciansandpediatricemergencyphysicianscarefor childrenwhomayhavebeenthevictimsofphysicalorsexual abuse,orwhomaybeexperiencingoneofmultipletypesof neglect(physical,supervisory,medical,educational,and emotional).Agrowingbodyofresearchdemonstratesthatbias regularlyinﬁltratesdecision-makingintheinitialclinical evaluationandmanagementofsuspectedabuse.Ina secondaryanalysisofCPSreportsconductedatanacademic tertiaryhospital,CortetalfoundBlackandHispanicpediatric patientswerefourtimesmore likelythanWhitepediatric patientstobereportedtoCPSbymedicalpersonnel.24

Laneetaldemonstratedinaretrospectivecohortstudy thatBlackandLatinxchildrenseenforfractureswereeight timesmorelikelythanWhitechildrentoreceiveaskeletal surveyandfourtimesmorelikelytobereportedtoCPS, regardlessofinsurancestatus(poverty)orlikelihoodofan abusiveinjury.5 Inastudyofpediatricpatientsadmittedfor

traumaticbraininjuryat39hospitals,Woodetalobserved thatfewerskeletalsurveysareperformedamongWhite infants(67.8%)thanBlackinfants(84%).25 Inacrosssectionalanalysisfrom18sitesinthePediatricBrainInjury ResearchNetwork,Hymeletalfoundthatamongacutely head-injuredpatientshospitalizedforintensivecare, minoritypatientsweretwiceaslikelytobeevaluatedand reportedforabusethanWhitepatients.Thisdisparitywas seenmostinpatientswhohadalowestimatedriskofabusive headtraumaattheoutsetoftheencounter.26 Mostrecently, Rebbeetalshowedthatamonghospitalizedpatients,Asian andNativeAmericanchildrenweremorelikelythanWhite childrentohavetheirencountercodedwithamaltreatmentrelateddiagnosticcodeandtobereportedtoCPS.27 Itisin partbecausebiasfactorsintochildmaltreatmentreporting thatemergencyphysicianshavesuchameaningfulrole toplay.

Itisnotourintentiontoencouragephysicianstoshirk mandatedreportingrequirements.Physicianswitha “reasonablesuspicion” ofchildmaltreatmentshouldreport theirconcernstotheappropriateauthorities.Thisremains thelegalstandardinmostjurisdictionsacrosstheUnited States.Thatsaid,itmustbeacknowledgedthatconsiderate, well-trainedpeopledisagreeonwhatconstitutes “reasonable suspicion.” Thetermisplaguedbyindeterminacyintheory andpractice.28 Surveystudiesconsistentlyshowwide variationamongphysiciansattemptingtodeﬁnereasonable suspicionandreachconsensusonthethresholdfor reporting.29 Inonestudy,clinicianschosetoreportonly73% ofchildrentheyconsidered “likely” or “verylikely” tohave beenabused,30 whileinaseparatestudynearlyhalfof respondentssaidtheywouldreportcaseswhereabusewas ﬁfthorlowerontheirlistofpotentialdifferential diagnoses.31 Inmanycases,itcannotbeassumedthatany twomandatedreportersagreeontheneedfora CPSreport.

Inthesettingofthisindeterminacy,theburdenfallstothe individualclinician,anditisarguablyanethicalonemore thanalegalone.UnnecessaryreportstoCPSstandindirect conflictwiththeprinciplesofnon-maleficenceandjustice. Reasonablepeoplemaydisagreeonthethresholdfor reportingpotentialabuse,butallcliniciansshouldconsider thefullscopeofharmsandbenefitsanytimethey contemplatea filing.

Optimistically,thischallengeisalsoanopportunity.There isagrowingenthusiasmforremedyingthechildwelfare systemamongpediatrichealthcarecliniciansinmany settings.4,6 Emergencyandpediatricemergencyphysicians areperfectlypositionedtojointhismovementandmake positivechangesinanunjustsystem.Inthe recommendationsthatfollow,wedrawontheworkof expertsinchildprotectionandimplicitbiastosuggest interventionsforemergencyphysiciansattheintrapersonal andinterpersonallevels.

1.Stayabreastofcurrentevidenceand recommendations.Theevidencebaseforchildabuse pediatricsisrapidlyevolving.Itisincreasingly understoodwhichmechanismsarefeasible explanationsforcertaininjuries,makingitpossible toaccuratelyidentifythe ﬁndingsmostindicativeof abuse.Agoodunderstandingofthisevidencebase allowsclinicianstoavoidunnecessaryworkupsand referrals.TheAmericanAcademyofPediatrics providesclinicalpracticeguidelinesthatsummarize currentrecommendationsfortheevaluationof physicalabuse.32,33 Goingastepfurther,physicians canusetheirknowledgeofevidence-basedpractice toimplementstandardizationwithintheir departments.Limitedstudiesshowthatclinical practiceguidelinescanimprovedisproportionalityin childmaltreatmentevaluation.34 Thisshouldbe donecarefully.Standardscreeningguidelinesmay improveproportionality,butstandardizedreferral guidelinesriskoverexposingchildrenandfamiliesof allracestothechildwelfaresystem.

2.Useamultidisciplinaryapproachtochildprotection. Athoroughprotectiveassessmentbyaskilledsocial workerisaninvaluabletoolfordeterminingwhether aCPS ﬁlingisnecessary.Inadditiontotheinjury historytypicallyobtainedbythemedicalteam,social workersoftenhavetheopportunitytolearnmore aboutfamilydynamics,housingstatus,caretaking responsibilities,protectivefactors,orpriorCPS involvement.Allthisinformationcanbeusedto makeaholisticassessmentofprotectiveconcerns andinformdialoguetomitigatebias.

3.Consultchildabusepediatricianswheneverpossible. Childabusepediatriciansaresubspecialistswith extensivetrainingandexpertiseinallaspectsofchild protection.Theycanrecommenddiagnosticworkups, commentonwhetherinjuriesareconsistentwithstated mechanisms,andhelpphysiciansnavigatethe reportingprocess.Thereisevidencethatconsultation withachildabusepediatriciancanreduceunnecessary referralstoCPS.35

4.UnderstandyourlocalCPSandmakereferralsin accordancewiththeircapabilities.Agoodemergency roomphysiciandoesnotorderatestortreatment withoutknowledgeofitslimitations.Thesameshould applytoareporttoCPS.Whattypesofconcernsdo investigatorshavetheexpertisetoinvestigate?What supportscantheyoffertofamilies?Cantheyhelpwith childcare,housing,transportation,oremployment?It isacommonmisconceptionamongcliniciansthata reporttoCPSbeneﬁtsfamiliesbyhelpingthemto accessadditionalresources.36,37 Inpractice,itis uncommonforCPStoprovidesocialsupportsthat couldnothavebeenmadeavailablebyahospital

socialworkerorprimarypediatrician.Ithasbeen questionedwhetherCPShasanimpactatallon modiﬁableriskfactorsforabuse.Inamulticenter cohortstudyof595childrenbetween4–8yearsold judgedtobeatriskofabuse,Campbelletalfoundno associationbetweenCPSinvestigationandsubsequent socialsupport,familyfunction,poverty,maternal education,orchildbehavioralproblems.38

5.Becarefulaboutlanguageandinformationsharing. Whencommunicatingaboutacasewithcolleaguesor CPS,avoiddiscussingunnecessarydemographic details(race,employmentstatus),orsharing subjectivejudgments(“lovelyfamily”).Themore communicationislimitedtoessentialinformation andneutrallanguage,themorewecanminimizethe effectofbias.

Emergencyclinicianscanplayanessentialrolein decreasingsystemicinequityinchildwelfarebymitigating implicitandexplicitbiasandtakingathoughtfulapproachto mandatedreporting.

AddressforCorrespondence:JosephP.Shapiro,MD,Children’s NationalHospital,DepartmentofEmergencyMedicineandTrauma, 111MichiganAvenueNW,WashingtonDC,20010.Email: Joseph. shapiro@childrensnational.org

ConflictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallaffiliations,fundingsources and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.U.S.DepartmentofHealth&HumanServices,Administrationfor ChildrenandFamilies,AdministrationonChildren,YouthandFamilies, Children’sBureau.Childmaltreatment2019.2021.Availableat: https:// www.acf.hhs.gov/cb/research-data-technology/statistics-research/ child-maltreatment.AccessedApril1,2023.

2.CénatJM,McInteeSE,MukunziJN,etal.Overrepresentation ofBlackchildreninthechildwelfaresystem:asystematic reviewtounderstandandbetteract. ChildrenYouthServRev. 2021;120:105714.

3.DettlaffAJ.(2014).Theevolvingunderstandingofdisproportionalityand disparitiesinchildwelfare.In:KorbinJEandKrugmanRD(Eds.), HandbookofChildMaltreatment (149–68).Dordrecht,TheNetherlands: Springer;2014.

4.PalusciVJandBotashAS.Raceandbiasinchildmaltreatment diagnosisandreporting. Pediatrics. 2021;148(1):e2020049625.

5.LaneWG,RubinDM,MonteithR,etal.Racialdifferencesinthe evaluationofpediatricfracturesforphysicalabuse. JAMA. 2002;288(13):1603–9.

6.RosenthalCM,ParkerDM,ThompsonLA.Racialdisparitiesinchild abusemedicine. JAMAPediatrics. 2021;176(2):119–20.

7.FongK.Gettingeyesinthehome:childprotectiveservices investigationsandstatesurveillanceoffamilylife. AmSociolRev. 2020;85(4):610–38.

8.FreisthlerB.Wehavemoretoolstosupportat-riskfamiliesthanjustchild protectiveserviceswithmandatedreporting. JAMAPediatrics. 2024;178(1):96.

9.TheAnnieE.CaseyFoundation. TheKidsCountDataBook. 2019. Availableat: https://www.aecf.org/resources/2019-kids-count-databook.AccessedApril1,2023.

10.WashingtonStateRacialDisproportionalityAdvisoryCommittee. (2010).RacialdisproportionalityinWashingtonStatereporttothe legislature.In: Lawsof2007(SHB1472).Olympia,WA:Washington StateDepartmentofSocialandHealthServices.

11.NeedellB,WebsterD,ArmijoM,etal.Childwelfareservicesreportsfor California.Availableat: http://cssr.berkeley.edu/ucb_childwelfare AccessedAugust31,2023.

12.DrakeB,LeeSM,Jonson-ReidM.Raceandchildmaltreatment reporting:areBlacksoverrepresented? ChildrenYouthServRev. 2009;31(3):309–16.

13.ReadL,O’RourkeCM,HoferL,etal.Racialdisproportionalityin Wisconsin’schildwelfaresystem.2009.Availableat: https://minds. wisconsin.edu/handle/1793/36539.AccessedApril1,2023.

14.TexasHealthandHumanServicesCommission,DepartmentofFamily andProtectiveServices.Disproportionalityinchildprotectiveservices: statewidereformeffortbeginswithexaminationoftheproblem.2006. Availableat: https://www.dfps.texas.gov/Child_Protection/ Disproportionality/documents/2006-01-02_Disproportionality.pdf AccessedApril1,2023.

15.HillRB.Synthesisofresearchondisproportionalityinchildwelfare:an update.2006.Availableat: https://www.aecf.org/resources/ synthesis-of-research-on-disproportionality-in-child-welfare-an-update AccessedApril1,2023.

16.YiY,EdwardsFR,WildemanC.Cumulativeprevalenceof conﬁrmedmaltreatmentandfostercareplacementforUS childrenbyrace/ethnicity,2011-2016. AmJPublicHealth. 2020;110(5):704–9.

17.LuYE,LandsverkJ,Ellis-MacleodM,etal.Race,ethnicity,and caseoutcomesinchildprotectiveservices. ChildrenYouthServRev. 2004;26(5):447–61.

18.HillRB.Ananalysisofracial/ethnicdisproportionalityanddisparityatthe national,state,andcountylevels.2007.Availableat: https://www.aecf. org/resources/an-analysis-of-racial-ethnic-disproportionality-anddisparity-at-the-nation.AccessedApril1,2023.

19.KimH,WildemanC,Jonson-ReidM,etal.Lifetimeprevalenceof investigatingchildmaltreatmentamongUSchildren. AmJPublic Health. 2017;107(2):274–80.

20.RobertsDE. ShatteredBonds:TheColorofChildWelfare.NewYork, NY:BasicBooks,2002.

21. LeeT. CatchingaCase:InequalityandFearinNewYorkCity’s ChildWelfareSystem.NewBrunswick,NJ:RutgersUniversity Press,2016.

22.EvangelistM,ThomasMMC,WaldfogelJ.Childprotectiveservices contactandyouthoutcomes. ChildAbuseNegl. 2023;136:105994.

23.RoviS,ChenPH,JohnsonMS.Theeconomicburdenof hospitalizationsassociatedwithchildabuseandneglect. AmJPublic Health. 2004;94(4):586–90.

24.CortNA,CerulliC,HeH.Investigatinghealthdisparitiesand disproportionalityinchildmaltreatmentreporting:2002–2006. JPublic HealthManagPract. 2010;16(4):329–36.

25.WoodJN,HallM,SchillingS,etal.Disparitiesintheevaluationand diagnosisofabuseamonginfantswithtraumaticbraininjury. Pediatrics. 2010;126(3):408–14.

26.HHymelKP,LaskeyAL,CrowellKR,etal.Racialandethnicdisparities andbiasintheevaluationandreportingofabusiveheadtrauma. J Pediatr. 2018;198:137–43.e1.

27.RebbeR,SattlerKM,MienkoJA.Theassociationofrace,ethnicity,and povertywithchildmaltreatmentreporting. Pediatrics. 2022;150(2):e2021053346.

28.LeviBHandLoebenG.Indexofsuspicion:feelingnotbelieving. Theor MedBioeth. 2004;25(4):277–310.

29.LeviBHandBrownG.Reasonablesuspicion:astudyof Pennsylvaniapediatriciansregardingchildabuse. Pediatrics. 2005;116(1):e5–12.

30.FlahertyEG,SegeRD,GrifﬁthJ,etal.Fromsuspicionofphysicalchild abusetoreporting:primarycarecliniciandecision-making. Pediatrics. 2008;122(3):611–9.

31.LeviBHandBrownG.Reasonablesuspicion:astudyof Pennsylvaniapediatriciansregardingchildabuse. Pediatrics. 2005;116(1):e5–12.

32.ChristianCW,CommitteeonChildAbuseandNeglect,American AcademyofPediatrics.Theevaluationofsuspectedchildphysical abuse. Pediatrics. 2015;135(5):e1337–54.

33.SectiononRadiology,AmericanAcademyofPediatrics.Diagnostic imagingofchildabuse. Pediatrics. 2009;123(5):1430–5.

34.RangelEL,CookBS,BennettBL,etal.Eliminatingdisparityin evaluationforabuseininfantswithheadinjury:useofascreening guideline. JournalofPediatricSurgery. 2009;44(6):1229–34; discussion1234–5.

35.RazM,Gupta-KaganJ,AsnesAG.Usingchildabusespecialiststo reduceunnecessarychildprotectiveservicesreportsand investigations. JAMAPediatr. 2023;177(12):1249–50.

36.JonesR,FlahertyEG,BinnsHJ,etal.Clinicians’ descriptionoffactors inﬂuencing theirreportingofsuspectedchildabuse:reportoftheChild

AbuseReportingExperienceStudyResearchGroup. Pediatrics. 2008;122(2):259–66.

37.FeldmanKW,MonasterskyC,FeldmanGK.Whenischildhood drowningneglect? ChildAbuseNeglt. 1993;17(3):329–36.

38.CampbellKA,CookLJ,LaFleurBJ,etal.Household,family,andchild riskfactorsafteraninvestigationforsuspectedchildmaltreatment:a missedopportunityforprevention. ArchPediatrAdolescMed. 2010;164(10):943–9.

LETTERSTOTHE EDITOR

Commentson “BicarbonateandSerumLabMarkersasPredictors ofMortalityintheTraumaPatient”

PatrickMcGinnis,BS*

SamanthaCamp,BS*

MinahilCheema,BS*

ShriyaJaddu*

QuincyTran,MD,PhD†

JessicaDowning,MD†

*UniversityofMarylandSchoolofMedicine,Baltimore,Maryland † UniversityofMarylandMedicalCenter,Baltimore,Maryland

SectionEditor:MarkI.Langdorf,MD,MHPE

Submissionhistory:SubmittedJuly17,2024;RevisionreceivedJuly25,2024;AcceptedJuly29,2024

ElectronicallypublishedSeptember25,2024

Fulltextavailablethroughopenaccessat http://escholarship.org/uc/uciem_westjem DOI: 10.5811/westjem.31021

DearEditor:

Wereadwithgreatenthusiasmthearticle “Bicarbonate andSerumLabMarkersasPredictorsofMortalityinthe TraumaPatient” byTalbottetal,1 whichwasrecently publishedinthe WesternJournalofEmergencyMedicine TalbottetalutilizedtheTriNetXdatabasetoexplore associationsbetweenserumlaboratorymarkers (bicarbonate,lactate,andbaseexcess)and30-daymortality. Despitethenoveltyoftheir ﬁndings,wehaveafewcomments aboutthemethodologyandresultsthatmaynotsupport theirconclusions.

Our ﬁrstcommentiswithregardtothelackof stratiﬁcationofanatomicallocationandtraumatype.While itiscommendabletoemploylargesamplesizestoimprove generalizability,itislesshelpfulforclinicianswhomanage varioustypesoftraumas.Forinstance,patientswhosustain gunshotwoundstothethighmaysufferhemorrhagicshock andmetabolicacidosisproducingelevatedlactatevalues,2 yettheyhaveahigherchanceofsurvivalifresuscitationis prompt.3 Severalstudieshaveshownlactatelevelsfrom patientswithblunttraumaticbraininjuryare indistinguishablefromhealthycontrols,yetthesepatients’ riskofmortalityisincreasedduetothelocationofinjury.4 Thesepatientsactivatedifferenttreatmentalgorithms, makingconclusionsabout “trauma” patientstoobroadfor clinicalutility.

Moreover,themethodologyforpatientselectioncouldbe morenarrowlydefined.TheauthorsusedtheInternational ClassificationofDiseases,10th Revision,Clinical Modification(ICD)codeICD10CM:T14Injurytoany unspecifiedbodyregion,whichincludesanyinjuryfroma superficialinjurytotraumaticamputation.Thisposesawide spectrumoftreatmentsandsurvivability,asinjurylocation impactsthetriageprocessandmobilizationofstaff andresources.

WereplicatedTalbott’sanalysiswithanarrower anatomicallocationusingICD10CM:S06Intracranial injuryincludingsubarachnoidandsubduralhemorrhage, bothofwhichraisesuspicionforelevatedlactate.Weagree higherlactatevaluesareassociatedwithincreasedriskof mortality;however,specifyingthelocationofinjury (intracranial)strengthenedthisassociationbeyondthe Talbottetalanalysis.Afterstratifyinglactatebyincrements of2wefoundthattheriskofmortalitywas1.06%,3.42%. 13.3%,27.4%,38.1%,and57.0%forlactaterangesof <2, 2–4,4–6,6–8,8–10,and10–12millimolesperliter, respectively.Theseresultsillustratehowspecifyingan injurytypeandanatomicallocationprovidesnarrower,more usefulassociationsbetweenserumlabmarkersand 30-daymortality.

Finally,Talbottetalused “sameday” lactatelevels. Althoughlactateatarrivalmayhaveprognosticvalue, persistentlyelevatedlevelafter20hoursofinjuryportends differentprognoses.5 Whenprovidingguidancetoclinicians treatingtraumapatients,thisgranularityaidsin understandingapatient’sclinicalcondition,asoutcomesfor patientswithelevatedarrivallactatethatisclearedafterfour hoursvsthosewithpersistentlyelevatedlactateat24hours arevastlydifferent.6 Nonetheless,wecommendthe workofTalbottetalandareexcitedaboutthemagnitudeof thisstudy.Werecommendthatcliniciansbe prudentininterpretingthestudy’sresultandinapplying themtotheirclinicalpractice.

AddressforCorrespondence:PatrickJohnMcGinnis,Universityof MarylandSchoolofMedicine,108NGreeneSt.,BaltimoreMD 21201.Email: pmcginnis@som.umaryland.edu

ConﬂictsofInterest:Bythe WestJEMarticlesubmissionagreement, allauthorsarerequiredtodiscloseallafﬁliations,fundingsources

and financialormanagementrelationshipsthatcouldbeperceived aspotentialsourcesofbias.Noauthorhasprofessionalor financial relationshipswithanycompaniesthatarerelevanttothisstudy. Therearenoconflictsofinterestorsourcesoffundingtodeclare.

REFERENCES

1.TalbottMM,WaguespackAN,ArmstrongPA,etal.Bicarbonateand serumlabmarkersaspredictorsofmortalityinthetraumapatient. West JEmergMed. 2024;25(4):661–7.

2.KrishnaU,JoshiSP,ModhM.Anevaluationofserialbloodlactate measurementasanearlypredictorofshockanditsoutcomeinpatients oftraumaorsepsis. IndianJCritCareMed. 2009;13(2):66–73.

3.KrauszMM.Initialresuscitationofhemorrhagicshock. WorldJEmerg Surg. 2006;1:14.

4.GlennTC,MartinNA,HorningMA,etal.Lactate:brainfuelinhuman traumaticbraininjury:acomparisonwithnormalhealthycontrol subjects. JNeurotrauma. 2015;32(11):820–32.

5.GewirtzRJ,DhillonHS,GoesSE,etal.Lactateandfreefatty acidsaftersubarachnoidhemorrhage. BrainRes. 1999;840(1–2):84–91.

6.DezmanZDW,ComerAC,SmithGS,etal.Repeatlactatelevel predictsmortalitybetterthanrateofclearance. AmJEmergMed. 2018;36(11):2005–9.