Journal of Managed Care Nursing Volume 7, Number 1

Vol. 7, No. 1, January 2020

Call for Articles

Journal of Managed Care Nursing Interested writers are requested to submit their editorial or research articles!

The JMCN publishes topics on managed care and related subjects, like quality & utilization management, patient advocacy, current trends, changing legislature, leadership tips, and more.

For more information on submitting an article, contact Jackie Beilhart at jbeilhart@aamcn.org or view the author guidelines at www.aamcn.org/jmcn.html

www.aamcn.org P (804) 747-9698 F (804)747-5316 2

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

JMCN JOURNAL OF MANAGED CARE NURSING 4435 Waterfront Drive, Suite 101 Glen Allen, VA 23060 EDITOR-IN-CHIEF Jacqueline Cole, RN-BSN, MS, CNOR, CPHQ, CMCN, CHC, CHPC, FNAHQ, FAHM, FHIAS PUBLISHER Jeremy Williams VICE PRESIDENT OF COMMUNICATIONS Jackie Beilhart JOURNAL MANAGEMENT American Association of Managed Care Nurses 4435 Waterfront Drive, Suite 101 Glen Allen, VA 23060 phone (804) 747-9698 fax (804) 747-5316 MANAGING EDITOR Jackie Beilhart jbeilhart@aamcn.org GRAPHIC DESIGN Jackie Beilhart jbeilhart@aamcn.org

ISSN: 2374-359X. The Journal of Managed Care Nursing is published by AAMCN. Corporate and Circulation offices: 4435 Waterfront Drive, Suite 101, Glen Allen, VA 23060; Tel (804) 747-9698; Fax (804) 747-5316.

Advertising Offices: Jackie Beilhart, 4435 Waterfront Drive, Suite 101, Glen Allen, VA 23060 jbeilhart@aamcn.org; Tel (804) 7479698. All rights reserved. Copyright 2020.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage or retrieval system, without written consent from the publisher. The publisher does not guarantee, either expressly or by implication, the factual accuracy of the articles and descriptions herein, nor does the publisher guarantee the accuracy of any views or opinions offered by the authors of said articles or descriptions.

Journal of Managed Care Nursing The Official Journal of the AMERICAN ASSOCIATION OF MANAGED CARE NURSES A Peer-Reviewed Publication

Vol. 7, No. 1, January 2020

TABLE OF CONTENTS Articles Comparison of the Quality of Hospitals That Admit Medicare Advantage Patients vs Traditional Medicare Patients David J. Meyers, Amal N. Trivedi, Vincent Mor, Momotazur Rahman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Association of Medicaid Expansion With Opioid Overdose Mortality in the United States Nicole Kravitz-Wirtz, Corey S. Davis, William R. Ponicki, Ariadne Rivera-Aguirre, Brandon D. L. Marshall, Silvia S. Martins, Magdalena Cerdรก. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Determining Levers of Cost-effectiveness for Screening Infants at High Risk for Peanut Sensitization Before Early Peanut Introduction Matthew Greenhawt, Marcus Shaker . . . . . . . . . . . . . . . . . . .28 Research Letter Evaluating Improvements and Shortcomings in Clinician Satisfaction With Electronic Health Record Usability Kylie M. Gomes, Raj M. Ratwani . . . . . . . . . . . . . . . . . . . . . .40 Managed Care Updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Newly Certified Managed Care Nurses (CMCNs) . . . . . . . 45 New AAMCN Members . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46 Author Submission Guidelines . . . . . . . . . . . . . . . . . . . . . .48

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

3

4

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Comparison of the Quality of Hospitals That Admit Medicare Advantage Patients vs Traditional Medicare Patients David J. Meyers, MPH(1); Amal N. Trivedi, MD, MPH(1,2); Vincent Mor, PhD(1,2); Momotazur Rahman, PhD(1) 1. Department of Health Services, Policy, and Practice, Brown University School of Public Health, Providence, Rhode Island; 2. Providence VA Medical Center, Providence, Rhode Island

Abstract Importance In the Medicare Advantage (MA) program, Medicare enrollees may be steered by their health plan to specific hospitals. Little is known about the quality of hospitals that serve MA enrollees.

Objective To compare the quality of hospitals that admit MA enrollees with the quality of those that admit traditional Medicare enrollees.

Design, Setting, and Participants This cross-sectional study used data from the 2012 to 2016 Medicare Provider Analysis and Review to compare quality of care, as measured by the star rating given by the Centers for Medicare and Medicaid Services and readmission rates, in hospitals that serve MA enrollees and traditional Medicare enrollees using multinomial logit models. Participants were 7 130 610 Medicare beneficiaries admitted to 2994 acute care hospitals across the United States in 2016. Data were analyzed between August 2018 and August 2019.

Exposures The exposure was MA enrollment. Adjusters included demographic and clinical characteristics and zip code fixed effects.

Main Outcomes and Measures Hospital Compare star ratings and quintiles of performance in 30-day readmission rates.

Results The sample included 7 130 610 Medicare beneficiaries in 2016 (54.3% female; mean [SD] age, 72.7 [13.2] years). Of 12 190 270 total hospitalizations, 1 211 293 traditional Medicare and 494 352 MA patients were admitted to 718 low-readmission hospitals and 1 205 586 traditional Medicare and 526 955 MA patients were admitted to 597 high-readmission hospitals. Accounting for observed patient characteristics, MA enrollees less often entered either low- or high-quality hospitals and were more often admitted to average-quality hospitals. For nonemergent hospitalizations, MA enrollees were 1.9 percentage points (95% CI, 1.5-2.2 percentage points) less likely to enter a low-readmissions hospital, 5.1 percentage points (95% CI, 4.6-5.6 percentage points) more likely to enter an average-readmissions hospital, and 3.2 percentage points (95% CI, 2.9-3.5 percentage points) less likely to enter a high-readmissions hospital compared with traditional Medicare enrollees. Patients with MA were also 2.6 percentage points (95% CI, 2.2-2.9 percentage points) less likely to enter a 1- to 2-star hospital, 5.5 percentage points (95% CI, 4.9-5.9 percentage points) more likely to enter a 3-star hospital, and 2.8 percentage points (95% CI, 2.5-3.2 percentage points) less likely to enter a 4- to 5-star hospital compared with traditional Medicare enrollees. The differences were less pronounced for emergency admissions. www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

5

Conclusions and Relevance This study found that enrollees in MA plans were more likely to be admitted to average-quality hospitals instead of either high- or low-quality hospitals, suggesting that MA plans may be steering their enrollees to specific hospitals.

Key Points Question How does the quality of hospitals that admit Medicare Advantage enrollees compare with the quality of those that admit traditional Medicare enrollees?

Findings In this cross-sectional study of 12 190 280 hospitalizations in the United States in 2016, Medicare Advantage enrollees had a 2.8–percentage point lower probability of entering a highly rated hospital, a 5.5–percentage point higher probability of entering a hospital with an average ranking, and a 2.6–percentage point lower probability of entering a poor-quality hospital compared with traditional Medicare enrollees. Results were consistent across measures of hospital quality.

Meaning These findings suggest that, owing either to differences in preference or steering by Medicare Advantage contracts, Medicare Advantage enrollees are less likely to enter the highest- or the lowest-quality hospitals.

INTRODUCTION More than one-third of all Medicare beneficiaries are now enrolled in the Medicare Advantage (MA)1 program, in which private insurance companies receive capitated payments to finance enrollees’ care needs.2 While MA plans are required to cover at least the same set of health care services as are covered by traditional Medicare (TM), MA plans can establish a preferred network of hospitals and health care practitioners for their enrollees. Very little is known about the differences in the quality of hospitals that serve MA enrollees compared with those serving TM enrollees.

In MA, the capitated payments that plans receive provide dual incentives to both reduce spending on unnecessary care and improve quality to prevent future expenditures. These incentives, and the flexibility MA plans have to steer their enrollees to specific practitioners, may lead to differences in the quality of care received by MA and TM enrollees. While a previous study3 found that MA enrollees are generally admitted to lower-quality nursing homes than TM enrollees, such differences in care quality cannot be generalized because MA enrollees historically use nursing home care at much lower rates compared with TM enrollees.4 Additionally, MA plans may emphasize primary care or hospital care that may reduce dependence on long-term care. Owing to a general lack of data on MA 6

enrollees, it is not currently known what other differences in quality might exist.

Patient access to hospitals has recently come under increased scrutiny in the press5 and the academic literature.6-8 Little is known about how MA enrollees use hospitals. Recent studies9,10 have found that the size of MA plans’ hospital networks varies widely, with approximately 23% of plans having broad hospital networks and 16% of plans having narrow or ultranarrow networks. Furthermore, MA plans may limit some types of specialty hospital access. Medicare Advantage plans have been found to pay 5.6% less for hospital services than TM,11 which may also influence the hospitals available to MA enrollees. Several other factors can result in differences in the quality of hospitals to which MA and TM enrollees are admitted. Medicare Advantage enrollees and TM enrollees may have different geographic access to hospitals depending on where they live. Furthermore, MA and TM enrollees are known to differ in important ways12 and may have different preferences when selecting which hospitals they use for their care needs.

In this cross-sectional study, we compare the quality of hospitals to which MA and TM enrollees are admitted, accounting for enrollees’ characteristics and their geographic access to hospitals.

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

METHODS Data and Measures This study was approved by the Brown University institutional review board and received a waiver of informed consent owing to the inability to contact enrollees in deidentified claims data. Reporting followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. Study data were analyzed between August 2018 and August 2019.

Our primary data source for this analysis was 100% Medicare Provider and Analysis Review (MedPAR) claims, which include hospitalization-level data on all MA and TM enrollees who were admitted to a MedPAR reporting hospital.13 We used MedPAR and enrollment data from 2012 to 2016 to assess trends in hospital entrance over time. We then focused on data from 2016 for the remainder of our analysis. Since 2008, MedPAR has included claims from MA enrollees who were admitted to hospitals that receive Medicaid Disproportionate Share Hospital payments or graduate medical education hospitals, which accounts for 90% of all MA hospitalizations annually.14,15 From our initial sample of 3234 acute care hospitals, we removed 240 hospitals that do not receive Medicaid Disproportionate Share Hospital payments or medical education credits determined by hospital cost reports, as these hospitals are not required to report completely for MA enrollees, for a final sample of 2994 acute care hospitals, which account for more than 95% of all hospital records in the 2016 MedPAR data set.

Among 13 560 748 hospitalizations in 2016, we excluded 851 640 hospitalizations (6.2%) that took place out of an enrollees’ home state, as MA policies may differ for hospital access when an enrollee is traveling.

We further excluded 518 838 hospitalizations (3.8%) that took place at non-MedPAR reporting hospitals, for a final sample of 12 190 270 hospitalizations among 7 130 610 Medicare enrollees. We included all Medicare enrollees (both those aged <65 years and those aged ≥65 years) in our study sample. In sensitivity analyses, we further excluded enrollees with a hospitalization in the prior 6 months, and only kept the first hospitalization of the year for each enrollee to check whether past hospitalizations are associated with future hospital choice. We stratified all hospitaliza-

tions by those that were admitted from the emergency department (8 608 120 hospitalizations) and those that were not (3 582 150 hospitalizations).

We used the Medicare Master Beneficiary Summary File to classify MA enrollees as those who were enrolled in the program for each month of 2012 to 2016 and assigned their MA status at the month they were admitted to the hospital. As of 2016, the Master Beneficiary Summary File includes monthly contract identification numbers for each enrollee that can be linked to publicly available MA star ratings and plan characteristics. For prior years, we linked enrollees to Healthcare Effectiveness Data and Information Set files, which have each enrollee’s contract number. For sensitivity analyses, we classified plans rated 4 or 5 stars as high quality and plans rated less than 4 stars as low to average quality to test whether there were further differences between different types of MA plans.

We linked each hospital admission to publicly available Centers for Medicare and Medicaid Services (CMS) 5-star hospital ratings for 2016, CMS adjusted 30-day readmission rates, and CMS adjusted 30-day mortality rates for acute myocardial infarction, stroke, heart failure, coronary artery bypass graft, and chronic obstructive pulmonary disease to classify admitted hospitals by quality. Our primary outcome of interest was whether an enrollee is admitted to a hospital in the lowest quintile of readmissions (low-readmissions hospitals), the second to fourth quintiles of readmissions (average-readmissions hospitals), or the highest quintile of readmissions (high-readmissions hospitals).

We also assessed whether enrollees are admitted to low–, medium–, or high–star rated hospitals and to hospitals with different quintiles of 30-day mortality.

For enrollees in MA, we linked each enrollee’s plan and contract to publicly reported CMS MA characteristic files that provide details on the types of plans, plan premiums, plan MA star ratings, and the parent companies of plans.

Statistical Analysis Our primary analysis is from 2016 when the star ratings were released, and we stratified all our analysis by emergency and nonemergency admissions. We present the unadjusted and adjusted percentages of MA and TM enrollees who were admitted to differ-

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

7

ent hospital quality categories in 2016. In sensitivity analysis, we also plotted admission trends from 2012 to 2016 to determine whether trends in admission changed over time.

To adjust for the observable patient characteristics that may be associated with the quality of the hospital beneficiaries enter, we fit separate multinomial logit models for admission to low-, average-, and highreadmissions hospitals. We also modeled entrance into hospitals of low (1-2 stars), average (3 stars), or high (4-5 stars) star rating quality and quintiles of 30-day mortality for several conditions. As geography is likely an important factor in determining the quality of hospitals available to beneficiaries, we used a Mundlak hybrid model to account for geographic differences.16,17 In the Mundlak model, we estimated multinomial logit regressions and included zip code–level means of all model covariates and clustered standard errors on the zip codes to approximate fixed effects.

In secondary analyses, we included in eTable 2 in the Supplement results using only standard multinomial logit models as well as linear probability models estimating each outcome separately using zip code fixed effects.

In each model, we adjusted for patient age, sex, race/ ethnicity, dual enrollment status, and a flag indicating whether patients were enrolled in MA or TM at the time of admission. We also included the distance each enrollee lived from the nearest high-, average-, or low-quality hospital. We stratified all our models by emergency and nonemergency admissions.

In sensitivity analyses, we further included covariates for intensive care unit use during the hospitalization and the Elixhauser Comorbidity Index score from the index hospitalization to evaluate whether patient acuity was associated with hospital selection (we cannot assess this fully, however, as both intensive care unit use and Elixhauser Comorbidity Index score assignment take place after hospital admission). We also stratified our results by less than 4 star– vs 4 or more star–quality MA plans, dual eligibility status, rural vs nonrural status, comorbidity status, and for the top principal diagnosis codes to determine whether the results are robust to different specifications. In additional specifications, we used inverse probability of treatment weights to balance observable differences in 8

patient characteristics between MA and TM. We also estimated the models using multinomial logit models.

The multinomial logit model is useful in that we could simultaneously assess the differences in the probability of entering a hospital of each rating category. The limitation, however, is that we were unable to account for the local neighborhood characteristics and availability of hospitals, as it is not appropriate to include fixed effects in nonlinear models.18,19 After assessing trends in admission between TM and MA enrollees, we conducted follow-up analyses on the MA population alone to evaluate what plan characteristics might be associated with the quality of admitting hospitals. We fit linear probability models adjusting for zip code fixed effects for each of the hospital quality outcomes described, including variables for MA plan contract star rating, whether the contract’s parent company is a national organization, tertiles of plan size, contract age, contract penetration in the enrollee’s county plan type (health maintenance organization [HMO], preferred provider organization, and other), and tertiles of plan premium. We clustered our standard errors in these models by the MA contract to account for multiple admissions from enrollees in the same contracts. All analysis was conducted using Stata software version 15 (StataCorp) and used 2-tailed significance tests with an α of .05.

RESULTS The sample included 7 130 610 Medicare beneficiaries in 2016 (54.3% female; mean [SD] age, 72.7 [13.2] years). There were a total of 12 190 270 hospitalizations in 2994 acute care hospitals. We found 1 211 293 TM and 494 352 MA patients were admitted to 718 low-readmission hospitals, 1 159 142 TM and 522 258 MA patients were admitted to 1679 average-admission hospitals, and 1 205 586 TM and 526 955 MA patients were admitted to 597 high-readmission hospitals.

Table 1 presents patient characteristics stratified by enrollment category and type of admission. Compared with TM enrollees, MA enrollees tended to be older (mean [SD] age for nonemergency admission, 70.6 [12.0] years vs 72.2 [10.3] years, respectively), were less likely to be white (for nonemergency admissions, 84.4% white vs 80.5% white, respectively), were less often dually enrolled in Medicaid (for nonemergency admissions, 17.6% vs 15.3%, respectively), and gener-

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

ally lived closer to higher-quality star-rated hospitals (for nonemergency admissions, median [interquartile range] distance to high-quality hospital, 5.4 [1.7-11.4] miles vs 4.4 [1.7-9.5] miles, respectively). The three most common nonemergency primary diagnoses were knee osteoarthritis, hip osteoarthritis, and coronary atherosclerosis. For the joint osteoarthritis diagnoses, hospitalizations were generally for joint replacement.

The three most common emergency diagnoses were septicemia, pneumonia, and acute myocardial infarction.

To understand the role of neighborhood as a driver of disparity in the quality of hospital chosen, in the Figure, we plotted the percentage of TM and MA patients entering a hospital rated less than 3 stars, a hospital rated 3 stars, or a hospital rated 4 or more stars for nonemergent admissions by the distance from their residential zip code centroid to the nearest hospital of that rating. As expected, for all categories, patients who lived closer to a hospital with a given rating category were more likely to be admitted to a hospital of that rating; however, MA enrollees were less often admitted to hospitals rated less than 3 stars, more often admitted to 3 star hospitals, and less often admitted to hospitals rated 4 or more stars than TM enrollees who

resided in the same zip codes. The same trend existed for emergency admissions (eFigure 2 in the Supplement); however, the magnitude of the differences was smaller.

In Table 2, we present the unadjusted percentages of patients in either TM or MA who were admitted to hospitals in each quality rating, readmission, and mortality category. We then present the adjusted difference based on the primary fixed-effects models. After adjusting for patient characteristics and accounting for geographic access, we found that MA enrollees were 1.9 percentage points (95% CI, 1.5-2.2 percentage points) less likely to be admitted to a hospital in the lowest quintile of readmissions, 5.1 percentage points (95% CI, 4.6-5.6 percentage points) more likely to be admitted to a hospital in the second to fourth quintiles of readmissions, and 3.2 percentage points (95% CI, 2.9-3.5 percentage points) less likely to be admitted to a hospital in the highest quintile of readmissions compared with TM enrollees. Medicare Advantage enrollees were also 2.6 percentage points (95% CI, 2.2-2.9 percentage points) less likely to be admitted to a 1- to 2-star hospital, 5.5 percentage points (95% CI, 4.9-5.9 percentage points) more likely to be admitted to a 3-star hospital, and 2.8 percentage points (95%

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

9

Figure. Share of Medicare Advantage and Traditional Medicare Patients Admitted to Hospitals in Neighborhoods With Different Proximity to Hospitals

10

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

CI, 2.5-3.2 percentage points) less likely to be admitted to a 4- to 5-star hospital compared with TM enrollees. Differences were similar, although substantially smaller, for emergency admissions. Full regression output is available in eTable 1 in the Supplement.

In Table 3, we present results from our analysis of plan characteristics associated with hospital quality. The probabilities of being admitted to a low-star hospital for enrollees in plans rated 3 to 3.5, 4 to 4.5, and 5 stars were 4.8 (95% CI, 0.0-9.3), 5.1 (95% CI, 0.09.6), and 22.8 (95% CI, 15.0-30.7) percentage points lower, respectively, than those enrolled in plans with 2 to 2.5 stars. Enrollees in contracts owned by national companies and non-HMOs also had lower probabilities of admission to lower-rated hospitals. Contract enrollment, age, and premium were all not statistically significantly associated with admission to a hospital with a high or low star rating. In eTable 2 in the Supplement, we also present results from the multino-

mial logit models, which yielded similar results.

In eFigure 1 in the Supplement, we present the unadjusted percentages of MA and TM enrollees who were admitted to hospitals rated 4 stars or higher (as defined in 2016) from 2012 to 2016, stratified by emergency admission status. Over time, there appears to be a persistent trend that MA enrollees had lower rates of unadjusted entry to highly rated hospitals for both emergency admissions (in quarter 1 2012, 25% of MA enrollees and 26% of TM enrollees were admitted to high-quality hospitals; in quarter 4 2016, 25% of MA enrollees and 26% of TM enrollees were admitted to high-quality hospitals) and nonemergency admissions (in quarter 1 2012, 25% of MA enrollees and 29% of TM enrollees were admitted to high-quality hospitals; in quarter 4 2016, 26% of MA enrollees and 31% of TM enrollees were admitted to high-quality hospitals).

These trends did not appear to change in 2016 when the star ratings were introduced.

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

11

In eTable 3 and eTable 4 in the Supplement, we present the results stratified by different patient subsamples. Admission trends were largely similar across patient subsamples. Notably, while differences were attenuated in rural settings, they were similar between dual-eligible and nonâ&#x20AC;&#x201C;dual eligible enrollees and across plan star ratings. The differences were also similar when further excluding enrollees with prior hospitalizations and only including each enrolleeâ&#x20AC;&#x2122;s first hospitalization of the year. In eTable 5 in the Supplement, we present our alternative model specification with linear probability models including zip code fixed effects, which yielded similar results to our primary models.

DISCUSSION We found that after adjusting for patient characteristics and geography, MA enrollees were systematically admitted to average-quality hospitals and less likely to be admitted to hospitals with high star ratings or to those with the lowest star ratings. We observed a similar trend for publicly reported outcome measures for 30-day readmissions and 5 different 30-day mortality measures. We cannot determine whether these differences are due to differences in networks that enrollees were actually able to use or differences in preference; however, the trends were consistent across quality outcomes.

Among enrollees in MA, plan quality had the strongest association with hospital quality. Enrollees in high-rated plans were much less likely to be admitted to lower-quality hospitals compared with enrollees in low-rated plans. We also did not find a consistent association between plan premiums and the quality of networks, indicating that enrollees may not be paying higher premiums in return for higher-quality networks.

Enrollment in higher-rated MA plans, plans that have a national presence, plans with high concentrations in a local market, and non-HMO plans all had lower probabilities of admission to lower-rated hospitals and higher probabilities of admission to higher-rated hospitals. National plans and those with a more robust local presence may be able to more effectively negotiate with hospitals to get better payment arrangements, allowing them to offer higher-quality hospitals in their 12

networks. It is unsurprising that HMOs appeared to be associated with lower-quality admissions, as narrow networks are often a key feature used by HMOs to control costs. As the market power of MA plans increases, they may be able to more effectively negotiate their preferred prices and have a greater ability to include higher-quality hospitals in their networks.

Future work that tests the association between market power and admitted hospitals may offer valuable further insight.

There are several possibilities that may have contributed to our findings. Medicare Advantage plans may restrict the hospitals available to patients in their networks,9 leading to both the highest- and lowest-quality hospitals being less likely to be included in networks.

A similar trend has been found for skilled nursing facilities.3 If higher-quality hospitals demand higher reimbursement rates, MA plans may avoid contracting with such hospitals, thereby excluding them from the network. On the other hand, if lower-quality hospitals have higher rehospitalization rates, they will necessarily be costlier to MA plans. Traditional Medicare enrollees living in the same neighborhoods do not face such restrictions on their choice of hospital, which may lead to their use of both higher- and lower-quality hospitals. It is unsurprising that the association was stronger for nonemergency admissions, as planned hospitalizations for surgical procedures and other treatments may be more subject to network design.

In the case of emergency hospitalizations, enrollees may be admitted to the nearest hospital; however, it is noteworthy that some differences persisted in selected quality measures. A previous study20 found generally broad MA networks in primary care settings; however, hospital care is a more expensive form of health care utilization and may be subject to greater restrictions.

It is notable that all the differences in hospital selection persisted when accounting for zip code in the model. We compared MA and TM enrollees who lived in the same neighborhood directly against each other in our estimates. As such, we do not believe the differences in admitted hospital quality can be explained by differential geographical proximity alone. We cannot rule out the possibility that enrollees in MA and TM have different preferences in the types of hospitals they select, which may explain some of the differences. Medicare Advantage enrollees may prefer to

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

enter average-rated hospitals over high- and low-rated hospitals if these hospitals are different in ways we cannot measure in this study.

There did not appear to be a substantial difference in the use of higher-quality hospitals between TM and MA enrollees in rural areas. These findings may be due to there being fewer hospitals available in rural areas, limiting the ability of MA plans to selectively contract with hospitals and to steer their enrollees to network hospitals.

While this study did not assess the health outcomes that beneficiaries experience, the limited selection of hospitals may lead to poorer health for patients if those patients admitted to average-quality hospitals have worse outcomes than those admitted to the highest-quality hospitals. At the same time, it may be beneficial to enrollees who are steered away from the lowest-quality hospitals. To address this concern, MA networks could be incorporated into annual calculation of MA star ratings and reported on each year. Currently, when enrollees choose MA plans, there is limited information about the breadth and quality of hospitals available in the plans’ networks that can be used when making enrollment decisions. The Centers for Medicare and Medicaid Services could take steps to make this more transparent.

Limitations This study has several limitations. Hospital star ratings are new and not necessarily a validated measure of quality. Early work has found that higher star ratings may be associated with higher quality of care for some conditions, but other work has found teaching hospitals and hospitals that treat low-income patients may be penalized.21,22 Despite these limitations, the star ratings are calculated on the basis of a variety of quality measures that may be important to patient outcomes, and the differences in admitted hospital quality persisted for mortality and readmissions rates, indicating that the differences in the rates of admission we detected may not be due to the idiosyncrasies of the star rating system alone. While this study included Medicare beneficiaries both older and younger than 65 years, the publicly reported measures we use are generally calculated only from those older than 65 years. As such, these quality measures may be less sensitive to the experience of younger beneficiaries. Another key

limitation of this study is that we must infer from MA plan member health care utilization patterns to discern which hospitals are in a given MA plan’s network, as accurate official hospital networks are not publicly available. Furthermore, while we adjusted for sociodemographic characteristics and comorbidities, it is possible that unobserved factors associated with both MA enrollment and hospital choice may have influenced these results. Nonetheless, the fact that we observed large differences in quality strongly suggests that MA plans restrict hospitals from membership in their networks. We are also limited in that we only have data on MA hospitalizations from hospitals that receive additional funding from Medicaid Disproportionate Share Hospital payments or medical education. While this covers most MA hospitalizations in the country, it may underrepresent MA enrollees living in rural areas.

CONCLUSIONS With its continued growth, MA has become a vital component of the health care system. This study found differences in the quality of hospitals admitting MA patients compared with TM patients, suggesting that policy makers should monitor the quality of hospitals available in MA plans’ network and make this information available to enrollees.

This is an open access article distributed under the terms of the CC-BY License. © 2020 Meyers DJ et al.

JAMA Network Open. Published: January 15, 2020. doi:10.1001/jamanetworkopen.2019.19310 The referenced supplemental content is available online: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2758745 REFERENCES 1. Jacobson G, Damico A, Neuman T, Gold M; Kaiser Family Foundation. Medicare Advantage 2017 spotlight: enrollment market update. https:// www.kff.org/medicare/issue-brief/medicare-advantage-2017-spotlight-enrollment-market-update/.

Published June 6, 2017. Accessed August 7, 2019.

2. Neuman P, Jacobson GA. Medicare Advantage checkup. N Engl J Med. 2018;379(22):2163-2172. doi:10.1056/NEJMhpr1804089 3. Meyers DJ, Mor V, Rahman M. Medicare

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

13

Advantage enrollees more likely to enter lowerquality nursing homes compared to fee-for-service enrollees. Health Aff (Millwood). 2018;37(1):7885. doi:10.1377/hlthaff.2017.0714 4. Goldberg EM, Trivedi AN, Mor V, Jung H-Y, Rahman M. Favorable risk selection in Medicare Advantage: trends in mortality and plan exits among nursing home beneficiaries. Med Care Res Rev. 2017;74(6):736-749. doi:10.1177/1077558716662565 5. Kliff S. I read 1,182 emergency room bills this year. here’s what I learned. Vox. https://www.vox. com/health-care/2018/12/18/18134825/emergency-room-bills-health-care-costs-america. Published December 18, 2018. Accessed August 7, 2019.

6. Chartock B, Garmon C, Schutz S. Consumers’ responses to surprise medical bills in elective situations. Health Aff (Millwood). 2019;38(3):425430. doi:10.1377/hlthaff.2018.05399 7. Cooper Z, Scott Morton F. Out-of-network emergency-physician bills—an unwelcome surprise. N Engl J Med. 2016;375(20):1915-1918. doi:10.1056/NEJMp1608571 8. Garmon C, Chartock B. One in five inpatient emergency department cases may lead to surprise bills. Health Aff (Millwood). 2017;36(1):177-181. doi:10.1377/hlthaff.2016.0970 9. Jacobson G, Trilling A, Neuman T; Kaiser Family Foundation. Medicare Advantage hospital networks: how much do they vary? https://insurance. maryland.gov/Consumer/Documents/agencyhearings/KaiserReport-Medicare-Advantage-HospitalNetworks-How-Much-Do-They-Vary.pdf. Published June 2016. Accessed August 7, 2019.

10. Colvin JD, Hall M, Thurm C, et al. Hypothetical network adequacy schemes for children fail to ensure patients’ access to in-network children’s hospital. Health Aff (Millwood). 2018;37(6):873880. doi:10.1377/hlthaff.2017.1339 11. Baker LC, Bundorf MK, Devlin AM, Kessler DP. Medicare Advantage plans pay hospitals less than traditional Medicare pays. Health Aff (Millwood). 2016;35(8):1444-1451. doi:10.1377/ hlthaff.2015.1553 12. Meyers DJ, Belanger E, Joyce N, McHugh J, Rahman M, Mor V. Analysis of drivers of disenrollment and plan switching among Medicare Advantage beneficiaries. JAMA Intern Med. 2019;179(4):524-532. doi:10.1001/jamainternmed.2018.7639 14

13. Research Data Assistance Center. Medicare Provider Analysis and Review RIF [data file search]. https://www.resdac.org/cms-data/files/medpar-rif.

Published 2017. Accessed December 22, 2017.

14. Huckfeldt PJ, Escarce JJ, Rabideau B, KaracaMandic P, Sood N. Less intense postacute care, better outcomes for enrollees in Medicare Advantage than those in fee-for-service. Health Aff (Millwood). 2017;36(1):91-100. doi:10.1377/ hlthaff.2016.1027 15. Kumar A, Rahman M, Trivedi AN, Resnik L, Gozalo P, Mor V. Comparing post-acute rehabilitation use, length of stay, and outcomes experienced by Medicare fee-for-service and Medicare Advantage beneficiaries with hip fracture in the United States: a secondary analysis of administrative data. PLoS Med. 2018;15(6):e1002592. doi:10.1371/journal.pmed.1002592 16. Bell A, Fairbrother M, Jones K. Fixed and random effects models: making an informed choice.

Qual Quant. 2019;53(2):1051-1074. doi:10.1007/ s11135-018-0802-x 17. Mundlak Y. On the pooling of time series and cross section data. Econometrica. 1978;46(1):6985. doi:10.2307/1913646 18. Greene W. Fixed effects and bias due to the incidental parameters problem in the tobit model.

Econom Rev. 2004;23(2):125-147. doi:10.1081/ ETC-120039606 19. Bester CA, Hansen CB. Grouped effects estimators in fixed effects models. J Econom.

2016;190(1):197-208. doi:10.1016/j.jeconom.2012.08.022 20. Feyman Y, Figueroa JF, Polsky DE, Adelberg M, Frakt A. Primary care physician networks in Medicare Advantage. Health Aff (Millwood). 2019;38(4):537-544. doi:10.1377/ hlthaff.2018.05501 21. Kaye DR, Norton EC, Ellimoottil C, et al. Understanding the relationship between the Centers for Medicare and Medicaid Services’ Hospital Compare star rating, surgical case volume, and short-term outcomes after major cancer surgery.

Cancer. 2017;123(21):4259-4267. doi:10.1002/ cncr.30866 22. Wan W, Liang CJ, Duszak R Jr, Lee CI. Impact of teaching intensity and sociodemographic characteristics on CMS Hospital Compare quality ratings. J Gen Intern Med. 2018;33(8):1221-1223. doi:10.1007/s11606-018-4442-6

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Article Information Corresponding Author: David J. Meyers, MPH, Department of Health Services, Policy, and Practice, Brown University School of Public Health, 121 South Main St, Providence, RI 02912 (david_meyers@brown.edu).

Author Contributions: Mr Meyers had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: Meyers, Trivedi, Rahman.

Drafting of the manuscript: Meyers, Rahman.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Meyers, Rahman.

Obtained funding: Mor, Rahman.

Administrative, technical, or material support: Meyers, Mor, Rahman.

Supervision: Trivedi, Mor, Rahman.

Conflict of Interest Disclosures: Dr Trivedi reported receiving grants from the National Institute on Aging during the conduct of the study. Dr Mor reported receiving personal fees from naviHealth outside the submitted work. Dr Rahman reported receiving grants from the National Institute on Aging during the conduct of the study. No other disclosures were reported.

Funding/Support: This work was supported by the National Institute on Aging of the National Institutes of Health under awards P01AG027296 and R21AG053712.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government.

www.AAMCN.org Spring Managed Care Forum 2020 • Visit the AAMCN website at www.aamcn.org to register for the Spring Managed Care Forum being held April 16-17, 2020 at the Gaylord Palms Resort in Orlando, FL. • Sit the CMCN exam the morning of April 15th Social Media • Members of AAMCN can join our new Facebook discussion group at www.facebook.com/ groups/AAMCN • LinkedIn

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

15

16

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Association of Medicaid Expansion With Opioid Overdose Mortality in the United States Nicole Kravitz-Wirtz, PhD, MPH(1); Corey S. Davis, JD, MSPH(2); William R. Ponicki, MA(3); Ariadne Rivera-Aguirre, MPP(4); Brandon D. L. Marshall, PhD(5); Silvia S. Martins, MD, PhD(6); Magdalena Cerdá, DrPH, MPH(4) 1. Violence Prevention Research Program, Department of Emergency Medicine, University of California Davis School of Medicine, Sacramento; 2. Network for Public Health Law, Los Angeles, California; 3. Prevention Research Center, Pacific Institute for Research and Evaluation, Berkeley, California; 4. Center for Opioid Epidemiology and Policy, Department of Population Health, New York University School of Medicine, New York; 5. Department of Epidemiology, Brown University School of Public Health, Providence, Rhode Island; 6. Mailman School of Public Health, Department of Epidemiology, Columbia University, New York, New York

Abstract Importance The Patient Protection and Affordable Care Act (ACA) permits states to expand Medicaid coverage for most low-income adults to 138% of the federal poverty level and requires the provision of mental health and substance use disorder services on parity with other medical and surgical services. Uptake of substance use disorder services with medications for opioid use disorder has increased more in Medicaid expansion states than in nonexpansion states, but whether ACArelated Medicaid expansion is associated with county-level opioid overdose mortality has not been examined.

Objective To examine whether Medicaid expansion is associated with county × year counts of opioid overdose deaths overall and by class of opioid.

Design, Setting, and Participants This serial cross-sectional study used data from 3109 counties within 49 states and the District of Columbia from January 1, 2001, to December 31, 2017 (N = 3109 counties × 17 years = 52 853 county-years). Overdose deaths were modeled using hierarchical Bayesian Poisson models. Analyses were performed from April 1, 2018, to July 31, 2019.

Exposures The primary exposure was state adoption of Medicaid expansion under the ACA, measured as the proportion of each calendar year during which a given state had Medicaid expansion in effect. By the end of study observation in 2017, a total of 32 states and the District of Columbia had expanded Medicaid eligibility.

Main Outcomes and Measures The outcomes of interest were annual county-level mortality from overdoses involving any opioid, natural and semisynthetic opioids, methadone, heroin, and synthetic opioids other than methadone, derived from the National Vital Statistics System multiple-cause-ofdeath files. A secondary analysis examined fatal overdoses involving all drugs.

Results There were 383 091 opioid overdose fatalities across observed US counties during the study period, with a mean (SD) of 7.25 (27.45) deaths per county (range, 0-1145 deaths per county).

Adoption of Medicaid expansion was associated with a 6% lower rate of total opioid overdose deaths compared with the rate in nonexpansion states (relative rate [RR], 0.94; 95% credible interval [CrI], 0.91-0.98). Counties in expansion states had an 11% lower rate of death involving heroin (RR, 0.89; 95% CrI, 0.84-0.94) and a 10% lower rate of death involving synthetic opioids other than methadone (RR, 0.90; 95% CrI, 0.84-0.96) compared with counties in nonexpansion states. An 11% increase was observed in methadone-related overdose mortality in expansion states (RR, 1.11; 95% www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

17

CrI, 1.04-1.19). An association between Medicaid expansion and deaths involving natural and semisynthetic opioids was not well supported (RR, 1.03; 95% CrI, 0.98-1.08).

Conclusions and Relevance Medicaid expansion was associated with reductions in total opioid overdose deaths, particularly deaths involving heroin and synthetic opioids other than methadone, but increases in methadone-related mortality. As states invest more resources in addressing the opioid overdose epidemic, attention should be paid to the role that Medicaid expansion may play in reducing opioid overdose mortality, in part through greater access to medications for opioid use disorder.

Key Points Question Is state Medicaid expansion associated with county-level opioid-involved overdose deaths in the United States?

Findings In this serial cross-sectional study of 3109 counties within 49 states and the District of Columbia from 2001 to 2017, Medicaid expansion was associated with reductions in total opioid overdose deaths and deaths involving heroin and synthetic opioids other than methadone. Expansion was associated with increased mortality involving methadone.

Meaning The findings suggest that expanding eligibility for Medicaid may help to mitigate the opioid overdose epidemic.

INTRODUCTION Drug overdose is a leading cause of injury-related death in the United States, responsible for more than 70 000 fatalities, or approximately 200 deaths per day, in 2017. Fatal drug overdoses have increased markedly during the past 2 decades in large part because of overdoses involving opioids, including prescription opioids and illegal opioids, such as heroin and illicitly manufactured fentanyl. Between 2001 and 2017, the age-adjusted mortality rate for opioid-related overdoses more than quadrupled, from 3.3 to 14.9 per 100 000 standard population. In 2017, more than two-thirds of all drug overdose fatalities (47 600 deaths) involved an opioid.1 Although overdose mortality may have stabilized in the past year, rates remain inordinately high.

The 2010 Patient Protection and Affordable Care Act (ACA) was signed into law during the rise in overdose deaths. Designed to increase access to and improve the quality of health insurance coverage, the ACA permits states to expand Medicaid coverage to essentially all non–Medicare-eligible people younger than 65 years with incomes at or below 138% of the federal poverty level ($16 643 for an individual in 2017).2 The law 18

also requires that individuals who receive coverage through the expansion be provided with mental health and substance use disorder (SUD) services on parity with other medical and surgical services.3 From the beginning of Medicaid expansion in 2014 to the end of study observation in 2017, a total of 32 states and the District of Columbia opted to expand Medicaid eligibility.4 Medicaid provides essential health care access to millions of low-income people and, by extension, greater access to low-cost prescription medications, including opioid pain relievers (OPRs). Such increased access to OPRs, particularly among a patient population with higher rates of chronic disease and disability compared with non-Medicaid recipients,5 has led some observers to question whether Medicaid expansion will contribute to additional opioid-related harms. To the contrary, recent studies6-8 have found that although Medicaid expansion was associated with an increased rate of overall Medicaid-reimbursed prescriptions, changes in prescriptions for OPRs before vs after the expansion were not significantly different in expansion vs nonexpansion states.

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Furthermore, Medicaid expansion has been an important source of coverage for SUD treatment, including for people with opioid use disorder (OUD). Previous research suggests that uptake of medications for opioid use disorder (MOUDs), including methadone, buprenorphine, and extended-release naltrexone, has increased more in expansion states compared with nonexpansion states.6-11 These medications (often in combination with counseling and behavioral therapies) have been linked to improvements in treatment retention and OUD remission as well as reductions, in some cases as high as 50%, in all-cause and overdoserelated mortality.12,13 Medicaid-reimbursed prescriptions for the opioid overdose reversal medication naloxone have also increased significantly more in expansion states compared with nonexpansion states.14 Early Medicaid expansions in Arizona, Maine, and New York in 2001 and 2002,15 along with more recent expansions in state Medicaid-eligibility thresholds for parents,16 have been associated with fewer drug overdose deaths. However, to our knowledge, with only 1 recent exception,17 no study has examined the association of ACA-related Medicaid expansion with opioidrelated overdose mortality more specifically.

Previous studies12,16,17 of the association of Medicaid expansion with fatal overdoses have been conducted at the state level. Although the most appropriate spatial scale for this association remains unclear, state-level analyses may not adequately reflect local (within-state) variation in the level and rate of growth of overdose deaths or differences in policy implementation, such as local disparities in the capacity for or accessibility of SUD treatment. Using overdose mortality and related covariates measured at the county rather than the state level, this study aimed to provide improved estimates of the association between Medicaid expansion under the ACA and fatal opioidinvolved overdoses from 2001 to 2017. We examined this association for county × year counts of total opioid overdose deaths and separately by class of opioid (ie, natural and semisynthetic opioids, methadone, heroin, and synthetic opioids other than methadone). For comparison with prior research, we also examined all drug overdose deaths as a secondary outcome.

METHODS This serial, cross-sectional study used data from 3109

counties in 49 states and the District of Columbia from January 1, 2001, to December 31, 2017. We organized this information into a series of space-time observations, with each observation referring to 1 year of data per county for a total of 52 853 county-years (3109 counties × 17 years). Analyses excluded Alaska because of substantial changes in the size and shape of counties within the state during the study period.

Individual data were aggregated to the county level.

This study and was approved by the institutional review board of the University of California, Davis.

No informed consent was required because this was a retrospective review of existing mortality data. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Outcome We determined annual, county-level counts of opioid overdose deaths from the restricted-use version of the National Vital Statistics System multiple-causeof-death files.18 Overdose deaths were identified based on the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) external cause of injury codes X40 to 44 (unintentional), X60 to 64 (suicide), X85 (homicide), and Y10 to 14 (undetermined). Among deaths with drug overdose as the underlying cause, we used the following ICD-10 specific drug codes to identify our outcomes: all opioids, T40.0-T40.4 and T40.6; natural and semisynthetic opioids, T40.2; methadone, T40.3; heroin, T40.1; and synthetic opioids other than methadone, T40.4. Deaths involving more than 1 class of opioid were included in the counts for each opioid subcategory; thus, opioid subcategories are not mutually exclusive.

Exposure Data on state Medicaid expansion status were obtained from the Kaiser Family Foundation.4 We created an indicator of the proportion of each calendar year during which a given state had Medicaid expansion in effect; states that expanded Medicaid were assigned a value of 0 in years before Medicaid expansion, a value between 0 and 1 in the year in which Medicaid expansion went into effect (according to the policy effective month), and a value of 1 in all subsequent years, whereas states that did not expand Medicaid by the end of the study period were assigned a value of

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

19

20

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

0 in all years. Of the 32 states (including the District of Columbia) in our study population that opted to expand Medicaid eligibility, 26 did so on January 1, 2014, then 2 additional states did so later that same year, followed by 2 states in 2015 and 2 states in 2016 (Table 1).

Covariates Annual, county-level estimates for a range of sociodemographic characteristics were obtained from GeoLytics Inc to be used as covariates, including age (percentage aged 0-19, 20-24, 25-44, and 45-64 years); percentage male; percentages non-Hispanic white, non-Hispanic Black, and Hispanic; percentage of families living in poverty; median household income (per $10â&#x20AC;Ż000); percentage unemployed; population density (1000 residents per square mile); and overall mortality rate (per 1000 people). We also considered the presence of co-occurring state policies, which have been associated in prior research19-21 with changes in opioidrelated harm, including prescription drug monitoring programs, overdose Good Samaritan laws, naloxone access laws, and medical marijuana laws. Information on these policies was derived from the Prescription Drug Abuse Policy System22 and from McClellan and colleagues19 and updated by us.

Statistical Analysis We examined the association between state Medicaid expansion status and county-level risk of fatal opioid overdoses overall and by class of opioid using Bayesian hierarchical Poisson models, with overdose deaths assumed to be distributed proportionally to the population of each county (aged â&#x2030;Ľ12 years). We introduced a 1-year lag between overdose rates and Medicaid expansion to address the possibility of temporal bias and to allow time for changes in Medicaid coverage, services, and related behaviors to materialize.

Analyses with Medicaid expansion instead measured concurrently with overdose rates produced similar results (eTable 2 in the Supplement). Furthermore, because drug-specific overdose rates may be variously underestimated or overestimated among states23 and for comparison with prior research, we conducted a secondary analysis with all drug overdose deaths as the outcome.

In practice, our models compared overdose trends in counties within states that expanded Medicaid before

vs after the expansion with trends in counties within nonexpansion states. Unlike conventional differencein-difference methods, the Bayesian approach does not assume that trends in overdose deaths before Medicaid expansion were the same among counties within expansion and nonexpansion states. Instead, by incorporating county-level random intercepts and trends, along with state-level fixed effects, growth mixtures among counties within states that occurred during the study period and could bias effect estimates were explicitly modeled. We also included conditional autoregressive spatial random effects, which account for the lack of independence in spatially contiguous counties (ie, spatial autocorrelation) and minimize the influence of large outlying rates in low-population counties by allowing each area to borrow strength from neighboring areas. All models also included fixed and random effects by county for Medicaid expansion to account for local variation in policy implementation across counties within states. We modeled secular trends in overdose using fixed linear and quadratic time trends and included annual, county-level sociodemographic covariates measured concurrently with overdose and co-occurring state policies with one-year time lags.

Analyses were implemented using the Integrated Nested Laplace Approximation method in R software, version 3.4.3 (R Project for Statistical Computing)24 from April 1, 2018, to July 31, 2019. Integrated nested Laplace approximation is an alternative to standard Markov chain Monte Carlo methods for estimating the integral of a posterior (probability) distribution.

Whereas Markov chain Monte Carlo samples from the posterior distribution of model parameters, integrated nested Laplace approximation returns comparable approximations to the posterior marginals in considerably less time.25,26 Results are reported as median relative rates (RRs) from the posterior marginal distribution and 95% credible intervals (CrIs) indicating a range of values that is expected to contain the true RR with 95% probability (a Bayesian analogue of a standard CI).

RESULTS There was a total of 383â&#x20AC;Ż091 opioid overdose fatalities across observed US counties for the study period of January 1, 2001, through December 31, 2017, with a mean (SD) of 7.25 (27.45) deaths per county (range,

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

21

0-1145 deaths per county) (Table 2). The overall opioid mortality rate increased over time, from 2.49 deaths per 100â&#x20AC;Ż000 people in 2001 to 11.41 deaths per 100â&#x20AC;Ż000 in 2017 (Figure 1). Rates were generally higher in expansion states than in nonexpansion states (eFigure in the Supplement). Overdoses involving natural and semisynthetic opioids accounted for the largest share of all county-year opioid overdose deaths (40.9%), followed by those involving heroin (25.3%), synthetic opioids other than methadone (24.0%), and methadone (17.1%). By 2017, most opioid overdose deaths (59.9%) involved synthetic opioids other than methadone (eg, illicitly manufactured fentanyl).

The estimated associations of 1-year lagged Medicaid expansion with RRs of opioid overdose deaths, overall and by class of opioid, are presented in Figure 2 22

(results for all model variables are in eTable 1 in the Supplement). Medicaid expansion was associated with lower risk of overdose mortality involving all opioids.

Specifically, counties within states that expanded Medicaid had a 6% decreased rate of opioid overdose deaths after expansion compared with counties within states that did not expand Medicaid eligibility (RR, 0.94; 95% CrI, 0.91-0.98). In drug-specific analyses, counties within states that expanded Medicaid had an 11% decreased rate of fatal heroin overdoses (RR, 0.89; 95% CrI, 0.84-0.94) and a 10% decreased rate of overdose deaths involving synthetic opioids other than methadone (RR, 0.90; 95% CrI, 0.84-0.96) after the expansion compared with counties in nonexpansion states. In contrast, the expansion was associated with an 11% increased rate of methadone-involved overdose deaths (RR, 1.11; 95% CrI, 1.04-1.19). An

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

of underreporting of specific drugs (ie, Alabama, Indiana, Louisiana, and Pennsylvania)23 produced substantively similar results as those in the primary analyses (eTable 2 in the Supplement).

Figure 1. Opioid Deaths per 100 000 Persons

DISCUSSION

Opioid Deaths per 100 000 Persons

association between Medicaid expansion and deaths involving natural and semisynthetic opioids was not well supported (RR, 1.03; 95% CrI, 0.98-1.08).

Consistent with previous research, our secondary analysis of overdose fatalities involving all drugs found that counties within states that expanded Medicaid had a 2% decreased rate of all drug overdose deaths after the expansion compared with those in nonexpansion states (RR, 0.98; 95% CrI, 0.96-1.00). Additional sensitivity analyses excluding 4 states with high levels

In this nationwide, population-based study of the association of Medicaid expansion under the ACA with county-level rates of opioid overdose mortality, we found empirical support for adopting and sustaining health coverage expansions as a potential tool for reducing opioid overdose deaths in the United States. Consistent with prior analyses16,27 examining Medicaid expansion and mortality from other causes, we found decreased rates of opioid overdose deaths associated with the adoption of Medicaid expansion.

In particular, given 82 228 opioid-related deaths from 2015 to 2017 in the 32 states that expanded Medicaid between 2014 and 2016, our findings suggest that these states would have had between 83 906 and 90 360 deaths in the absence of the expansion, implying that Medicaid expansion may have prevented between 1678 and 8132 deaths in these states during those years.

In analyses differentiated by class of opioid, we found a more substantial decreased risk associated with overdose deaths involving heroin and synthetic opioids other than methadone, which have been associ-

Figure 2. Estimated Associations of 1-Year Lagged Medicaid Expansion With Relative Rates of Opioid Overdose Deaths Overall and by Class of Opioid

Estimated Associations of 1-Year Lagged Medicaid Expansion With Relative Rates of Opioid Overdose Deaths Overall and by Class of Opioid CrI indicates credible interval. www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

23

ated with continued increases in opioid-related deaths in recent years. These findings align with previous research that indicates that implementation of the ACA was associated with 40% decreased odds of being uninsured among persons with heroin use disorders, primarily because of Medicaid expansion, whereas no changes in insurance coverage were detected among persons with prescription OUDs.28 We also did not find support for an association between ACA-related Medicaid expansion and natural and semisynthetic opioid overdose mortality.

The observed association between Medicaid expansion and decreased total opioid overdose deaths and deaths involving heroin and synthetic opioids other than methadone is likely in part attributable to the ACAâ&#x20AC;&#x2122;s inclusion of mental health and SUD services as essential health benefits. Expanded Medicaid eligibility has substantially increased access to these services among the low-income population.10,29 Recent evidence demonstrates that compared with nonexpansion states, Medicaid expansion states experienced increases in overall prescriptions for, Medicaid-covered prescriptions for, and Medicaid spending on both MOUDs, particularly buprenorphine and naltrexone, and the opioid overdose reversal medication naloxone.6-8,11,14,30,31,35 Two prior studies12,16 have found associations between income eligibility expansions for Medicaid and reductions in SUD-related deaths, and a recent study17 assessed changes in opioid-related deaths in Medicaid expansion vs nonexpansion states. Whereas the last study17 found that Medicaid expansion was associated with larger increases in opioid overdose mortality, particularly in 2015 and 2016, analyses were conducted only at the state level. This approach may have masked within-state variation in the level and rate of growth of opioid overdoses, as well as differences in local policy implementation. To our knowledge, ours is the first study to quantify the association between ACA-related Medicaid expansion and opioid-related deaths at the county level.

Although the rate of methadone-related mortality is relatively low compared with other opioid classes, our finding that Medicaid expansion was associated with increased methadone overdose deaths deserves further investigation. At the individual level, treatment of 24

OUD with methadone has been rigorously studied and found to be equally and, in some cases, more effective than other MOUDs in suppressing illicit opioid use, particularly heroin use, and retaining persons in treatment.31,32 On the basis of this evidence, in combination with our findings for heroin and synthetic opioids other than methadone, increased access to MOUDs likely not did not contribute to the observed increase in methadone mortality associated with Medicaid expansion. In contrast, past research has found high rates of methadone use to treat pain (rather than to treat OUD) among Medicaid beneficiaries and that the drug is disproportionately associated with overdose deaths among individuals in this population,33,34 underscoring the importance of ongoing local, state, and federal actions to address safety concerns associated with methadone for pain in tandem with Medicaid expansion.7,8 Limitations This study has limitations. First, we relied on ICD10 coding of death certificate data, which may not reliably identify the specific drugs involved in fatal overdoses and may lead to an underestimation or misclassification of opioid overdose mortality.23 However, a secondary analysis that examined overdose deaths involving all drugs and sensitivity analyses excluding states with high levels of underreporting of specific drugs produced similar results as those in our primary models. Second, we included deaths from opioid overdoses across the entire population, not just among Medicaid enrollees, which may understate the estimated outcomes of Medicaid expansion for those individuals most directly affected. Third, although we controlled for various county-level sociodemographic characteristics and state-level co-occurring policies, unmeasured confounding is still a possibility. Fourth, we did not examine the specific provisions of Medicaid expansion that may be associated with changes in opioid-related deaths (eg, state-level difference in Medicaidâ&#x20AC;&#x2122;s preferred drug lists). In addition, this study focused on the association of Medicaid expansion with fatal overdoses only. Future studies should consider the association of expansion with the spectrum of opioid-related harms, including prevention of SUD and nonfatal overdoses. Also, future studies should explicitly examine possible mediators and moderators of the association between Medicaid expansion and opioid overdose risk, including access to and use of

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

OPRs, MOUDs, and naloxone; local SUD treatment capacity; and the extent to which the association of Medicaid expansion with overdoses varies by individual sociodemographic characteristics and contextual conditions.

CONCLUSIONS This study found that Medicaid expansion was associated with reductions in opioid overdose deaths, particularly deaths involving heroin and synthetic opioids other than methadone, but with increases in methadone-related mortality. These findings add to the emerging body of evidence that Medicaid expansion under the ACA may be a critical component of state efforts to address the continuing opioid overdose epidemic in the United States. As states invest more resources in such efforts, attention should be paid to the role that health coverage expansions can play in reducing opioid overdose mortality, potentially through greater access to MOUDs. This is an open access article distributed under the terms of the CC-BY License. © 2020 Kravitz-Wirtz N et al. JAMA Network Open. Published: January 10, 2020. doi:10.1001/jamanetworkopen.2019.19066 Supplemental material is found at https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2758476 REFERENCES 1. Hedegaard H, Miniño AM, Warner M. Drug overdose deaths in the United States, 1999–2017.

NCHS Data Brief. 2018;(329):1-8.

2. 82 FR § 8831. 2017.

3. 42 CFR § 438, 456, and 457. 2016.

4. Henry J. Kaiser Family Foundation. Status of state action on the Medicaid expansion decision. https:// www.kff.org/health-reform/state-indicator/stateactivity-around-expanding-medicaid-under-theaffordable-care-act/?currentTimeframe=0&sortMo del=%7B%22colId%22:%22Location%22,%22so rt%22:%22asc%22%7D. Accessed April 20, 2018.

5. Chapel JM, Ritchey MD, Zhang D, Wang G. Prevalence and medical costs of chronic diseases among adult Medicaid beneficiaries. Am J Prev Med. 2017;53(6S2)(suppl 2):S143-S154.

doi:10.1016/j.amepre.2017.07.019 6. Sharp A, Jones A, Sherwood J, Kutsa O, Honermann B, Millett G. Impact of Medicaid expansion on access to opioid analgesic medications and medication-assisted treatment. Am J Public Health. 2018;108(5):642-648. doi:10.2105/ AJPH.2018.304338 7. Saloner B, Levin J, Chang HY, Jones C, Alexander GC. Changes in buprenorphine-naloxone and opioid pain reliever prescriptions after the Affordable Care Act Medicaid expansion. JAMA Netw Open. 2018;1(4):e181588-e181588. doi:10.1001/ jamanetworkopen.2018.1588 8. Cher BAY, Morden NE, Meara E. Medicaid expansion and prescription trends: opioids, addiction therapies, and other drugs.

Med Care. 2019;57(3):208-212. doi:10.1097/ MLR.0000000000001054 9. Meinhofer A, Witman AE. The role of health insurance on treatment for opioid use disorders: evidence from the Affordable Care Act Medicaid expansion. J Health Econ. 2018;60:177-197. doi:10.1016/j.jhealeco.2018.06.004 10. Zur J, Tolbert J. The Opioid Epidemic and Medicaid’s Role in Facilitating Access to Treatment.

San Francisco, CA: Henry J. Kaiser Family Foundation; 2018.

11. Wen H, Hockenberry JM, Borders TF, Druss BG. Impact of Medicaid expansion on Medicaidcovered utilization of buprenorphine for opioid use disorder treatment. Med Care. 2017;55(4):336341. doi:10.1097/MLR.0000000000000703 12. Sordo L, Barrio G, Bravo MJ, et al. Mortality risk during and after opioid substitution treatment: systematic review and meta-analysis of cohort studies. BMJ. 2017;357:j1550. doi:10.1136/bmj. j1550 13. Tanum L, Solli KK, Latif ZE, et al. Effectiveness of injectable extended-release naltrexone vs daily buprenorphine-naloxone for opioid dependence: a randomized clinical noninferiority trial. JAMA Psychiatry. 2017;74(12):1197-1205. doi:10.1001/jamapsychiatry.2017.3206 14. Frank RG, Fry CE. The impact of expanded Medicaid eligibility on access to naloxone. Addiction. 2019;114(9):1567-1574. doi:10.1111/ add.14634 15. Venkataramani AS, Chatterjee P. Early Medicaid expansions and drug overdose mortality in the

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

25

USA: a quasi-experimental analysis. J Gen Intern Med. 2019;34(1):23-25. doi:10.1007/s11606-0184664-7 16. Snider JT, Duncan ME, Gore MR, et al. Association between state Medicaid eligibility thresholds and deaths due to substance use disorders.

JAMA Netw Open. 2019;2(4):e193056-e193056. doi:10.1001/jamanetworkopen.2019.3056 17. Swartz JA, Beltran SJ. Prescription opioid availability and opioid overdose-related mortality rates in Medicaid expansion and non-expansion states.

Addiction. 2019;114(11):2016-2025. doi:10.1111/ add.14741 18. National Center for Health Statistics. Mortalityâ&#x20AC;&#x201D;All County, Micro-Data and Compressed, 2001-2017, for All States, as Compiled From Data Provided by the 57 Vital Statistics Jurisdictions Through the Vital Statistics Cooperative Program.

Hyattsville, MD: National Center for Health Statistics; 2017.

19. McClellan C, Lambdin BH, Ali MM, et al. Opioid-overdose laws association with opioid use and overdose mortality. Addict Behav. 2018;86:90-95. doi:10.1016/j.addbeh.2018.03.014 20. Bachhuber MA, Saloner B, Cunningham CO, Barry CL. Medical cannabis laws and opioid analgesic overdose mortality in the United States, 1999-2010. JAMA Intern Med. 2014;174(10):1668-1673. doi:10.1001/jamainternmed.2014.4005 21. Fink DS, Schleimer JP, Sarvet A, et al. Association between prescription drug monitoring programs and nonfatal and fatal drug overdoses: a systematic review. Ann Intern Med. 2018;168(11):783-790. doi:10.7326/M17-3074 22. Prescription Drug Abuse Policy System. http:// pdaps.org/. Accessed January 1, 2017.

23. Ruhm CJ. Geographic variation in opioid and heroin involved drug poisoning mortality rates. Am J Prev Med. 2017;53(6):745-753. doi:10.1016/j.amepre.2017.06.009 24. Blangiardo M, Cameletti M. Spatial and Spatial-Temporal Bayesian Models with R-INLA. Chichester, United Kingdom: Wiley; 2015. doi:10.1002/9781118950203 25. Beguin J, et al. Hierarchical analysis of spatially autocorrelated ecological data using integrated nested Laplace approximation. Methods Ecol Evol. 2012;3(5):921-929. doi:10.1111/j.204126

210X.2012.00211.x 26. Carroll R, Lawson AB, Faes C, Kirby RS, Aregay M, Watjou K. Comparing INLA and OpenBUGS for hierarchical Poisson modeling in disease mapping. Spat Spatiotemporal Epidemiol.

2015;14-15:45-54. doi:10.1016/j.sste.2015.08.001 27. Khatana SAM, Bhatla A, Nathan AS, et al. Association of Medicaid expansion with cardiovascular mortality. JAMA Cardiol. 2019;4(7):671-679. doi:10.1001/jamacardio.2019.1651 28. Feder KA, Mojtabai R, Krawczyk N, et al.

Trends in insurance coverage and treatment among persons with opioid use disorders following the Affordable Care Act. Drug Alcohol Depend. 2017;179:271-274. doi:10.1016/j.drugalcdep.2017.07.015 29. Antonisse L. The Effects of Medicaid Expansion Under the ACA: Updated Findings From a Literature Review. San Francisco, CA: Henry J. Kaiser Family Foundation; 2018.

30. Clemans-Cope L, Lynch V, Epstein M, Kenney GM. Medicaid Coverage of Effective Treatment for Opioid Use Disorder. Washington, DC: The Urban Institute; 2017.

31. Grogan CM, Andrews C, Abraham A, et al.

Survey highlights differences in medicaid coverage for substance use treatment and opioid use disorder medications. Health Aff (Millwood). 2016;35(12):2289-2296. doi:10.1377/ hlthaff.2016.0623 32. Mattick RP, Breen C, Kimber J, Davoli M. Buprenorphine maintenance versus placebo or methadone maintenance for opioid dependence.

Cochrane Database Syst Rev. 2014;(2):CD002207. doi:10.1002/14651858.CD002207.pub4 33. Faul M, Bohm M, Alexander C. Methadone prescribing and overdose and the association with medicaid preferred drug list policiesâ&#x20AC;&#x201D;United States, 2007-2014. MMWR Morb Mortal Wkly Rep. 2017;66(12):320-323. doi:10.15585/mmwr. mm6612a2 34. Urahn SK, Coukell A. The Use of Methadone for Pain by Medicaid Patients: an Examination of Prescribing Patterns and Drug Use Policies. Philadelphia, PA: The Pew Charitable Trusts; 2018.

35. Clemans-Cope L, Epstein M, Kenney G. Rapid Growth in Medicaid Spending on Medications to Treat Opioid Use Disorder and Overdose. Washington, DC: The Urban Institute; 2017.

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

ARTICLE INFORMATION Corresponding Author: Nicole Kravitz-Wirtz, PhD, MPH, Violence Prevention Research Program, Department of Emergency Medicine, University of California Davis School of Medicine, 2315 Stockton Blvd, Sacramento, CA 95817 (nkravitzwirtz@ ucdavis.edu).

Author Contributions: Dr Kravitz-Wirtz had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Kravitz-Wirtz, Marshall, Martins, Cerdá.

Acquisition, analysis, or interpretation of data: Kravitz-Wirtz, Davis, Ponicki, Rivera-Aguirre, Marshall, Cerdá.

Drafting of the manuscript: Kravitz-Wirtz, Cerdá.

Critical revision of the manuscript for important intellectual content: All authors.

Administrative, technical, or material support: Davis, RiveraAguirre, Marshall, Cerda.

Supervision: Davis, Marshall, Martins, Cerdá.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported by grant R01DA039962 from the National Institute on Drug Abuse (Dr Cerdá, primary investigator). Dr Kravitz-Wirtz was supported in part by the Violence Prevention Research Program, Department of Emergency Medicine, UC Davis School of Medicine. Dr Marshall was supported in part by grant P20-GM125507 from the National Institute of General Medical Sciences.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Statistical analysis: Kravitz-Wirtz, Ponicki, Rivera-Aguirre.

Obtained funding: Cerdá.

It’s our turn to pick up where you’ve left off.

Your care plan. Our CAREGivers.

Home Instead CAREGiversSM complete extensive training built on this experience and research, so they are prepared to offer the highest quality of care.

Our individualized approach allows us to tailor our care to a family’s unique needs. From personal care to hospice care support, you can trust we are in this together.

Visit HomeInstead.com

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

27

Each Home Instead Senior Care franchise is independently owned and operated.

Determining Levers of Cost-effectiveness for Screening Infants at High Risk for Peanut Sensitization Before Early Peanut Introduction Matthew Greenhawt, MD, MBA, MSc(1); Marcus Shaker, MD, MSc(2,3) 1. Section of Allergy and Immunology, Food Challenge and Research Unit, Children’s Hospital Colorado, University of Colorado School of Medicine, Aurora; 2. Section of Allergy and Immunology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; 3. Dartmouth Geisel School of Medicine, Hanover, New Hampshire

Abstract Importance Early peanut introduction reduces the risk of developing peanut allergy, especially in high-risk infants. Current US recommendations endorse screening but are not cost-effective relative to other international strategies.

Objective To identify scenarios in which current early peanut introduction guidelines would be costeffective.

Design, Setting, and Participants This simulation/cohort economic evaluation used microsimulations and cohort analyses in a Markov model to evaluate the cost-effectiveness of early peanut introduction with and without peanut skin prick test (SPT) screening in high-risk infants during an 80-year horizon from a societal perspective. Data were analyzed from April to May 2019.

Exposures High-risk infants with early-onset eczema and/or egg allergy underwent early peanut introduction with and without peanut SPT screening (100 000 infants per treatment strategy) using a dichotomous 8-mm SPT cutoff value (stipulated in the current US guideline).

Main Outcomes and Measures Cost, quality-adjusted life-years (QALYs), net monetary benefit, peanut allergic reactions, severe allergic reactions, and deaths due to peanut allergy.

Results In the simulated cohort of 200 000 infants and using the base case during the model horizon, a no-screening approach had lower mean (SD) costs ($13 449 [$38 163] vs $15 279 [$38 995]) and higher mean (SD) gain in QALYs (29.25 [3.28] vs 29.23 [3.30]) vs screening but resulted in more allergic reactions (mean [SD], 1.07 [3.15] vs 1.01 [3.02]), severe allergic reactions (mean [SD], 0.53 [1.66] vs 0.52 [1.62]), and anaphylaxis involving cardiorespiratory compromise (mean [SD], 0.50 [1.59] vs 0.49 [1.47]) per individual. In deterministic SPT sensitivity analyses at base-case sensitivity and specificity rates, screening could be cost-effective at a high disutility rate (the negative effect of a food allergic reaction) (76-148 days of life traded) for an at-home vs in-clinic reaction in combination with high baseline peanut allergy prevalence among infants at high risk for peanut allergy and not yet exposed to peanuts. If an equivalent rate and disutility of accidental and index anaphylaxis was assumed and the 8-mm SPT cutoff had 0.85 sensitivity and 0.98 specificity, screening was costeffective at a peanut allergy prevalence of 36%.

Conclusions and Relevance The results of this study suggest that the current screening approach to early peanut introduction could be cost-effective at a particular health utility for an in-clinic reaction, SPT sensitivity and specificity, and high baseline peanut allergy prevalence among high-risk infants.

However, such conditions are unlikely to be plausible to realistically achieve. Further research is needed to define the health state utility associated with reaction location.

28

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Key Points Question Although the current US policy of early peanut introduction is not cost-effective compared with universal introduction without screening, are there variables and assumptions under which this policy could be cost-effective?

Findings In this simulation/cohort economic evaluation, screening high-risk infants before peanut introduction was cost-effective at very high disutility (76-148 days of life traded) for having an inoffice index reaction or with greater than 36% baseline peanut allergy prevalence and peanut skin prick test sensitivity of 0.85 and specificity of 0.98.

Meaning These results suggest that cost-effectiveness of the current US early peanut introduction policy depends on a high disutility for having an index peanut allergic reaction at home, a high ambient prevalence of peanut allergy, and very high sensitivity and specificity of the skin prick test.

INTRODUCTION Peanut allergy affects 1% to 4.5% of children, can potentially be severe, and is not readily outgrown in most individuals.1 Moreover, although treatments are on the horizon, a cure remains elusive, and management involves strict avoidance and anaphylaxis preparedness.2,3 For these reasons, peanut allergy is associated with impaired quality of life and anxiety.4,5 Important advances have been made regarding prevention of peanut allergy through deliberate early introduction, in particular targeting populations of children at risk for developing peanut allergy, as demonstrated in the Learning Early About Peanut Allergy (LEAP) trial in which a dramatic risk reduction was noted against developing peanut allergy at 5 years of age through early peanut introduction at 4 to 11 months of life compared with delayed introduction.6 The strength of these findings helped to reverse prior recommendations to avoid peanut in infants and young children until 3 years of age and resulted in the recent National Institutes for Allergy and Infectious Disease (NIAID) addendum guidelines that recommend early introduction to prevent peanut allergy.7 This strategy was adopted in the United Kingdom, Canada, Australia, and New Zealand, although the wording of the policy and the implementation of this guidance vary among these nations. The United Kingdom, Australia, and New Zealand, and now the Canadian Pediatric Society all recommend early peanut introduction at approximately 6 months of age (but not before 4 months of age) without any prescreening and risk stratification.7-10 However, the NIAID guidelines strongly rec-

ommend that high-risk infants (eg, those with severe eczema and/or egg allergy) undergo peanut allergy testing at 4 to 6 months of age before having peanut introduced. Infants demonstrating peanut skin prick test (SPT) sensitization ranging from 3 to 7 mm are recommended to have in-clinic peanut introduction; those with sensitization of at least 8 mm are diagnosed as having preexisting peanut allergy, and introduction is withheld. For lower-risk infants (or infants not at risk), peanut introduction is advised as early as 6 months of age, without such medicalization or screening, in accordance with family values and preferences7 (eTable in the Supplement).

Although the NIAID strategy largely follows, with some extension, the parameters used in the LEAP study, the necessity of medical screening before peanut introduction was never specifically evaluated (because screening was not a randomized study variable) and can be questioned in light of the differing international strategies chosen.11 Indeed, applying the NIAID criteria to the HealthNuts population (an Australian population-level food allergy prevalence study in children aged one year),12 even if all infants with early-onset eczema and/or egg allergy were screened (approximately 16% of all infants born each year), 23% of all children in this cohort diagnosed with peanut allergy would have been missed. This raises questions about the sensitivity and specificity of these criteria as well as their necessity given that most initial reactions to peanut were mild and a fatal index peanut reaction has never been described.13 Moreover, in a recent cost-effectiveness model ex-

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

29

ploring the differing international recommendations for how to implement early peanut introduction, the screening strategy (irrespective of use of serologic evaluation or SPT) was dominated by a no-screen approach (eg, screening resulted in higher costs and lower benefits). Screening led to greater overdiagnosis, which attenuated the benefit of preventing a peanut index reaction.14 This overdiagnosis is a residual result of using a skin test cutoff based on a probability for allergy (whereby some will be diagnosed without ever eating and reacting to peanut), whereas if these individuals were allowed to ingest peanut, not all of them would have a reaction per se. Cost-effectiveness of the recommendations may not affect their implementation, because caregivers and physicians may still opt for care that is considered wasteful or low value based on such findings. A limitation of the previous analysis14 was that it strictly compared the approaches and did not explore all levers or pathways that may exist for the NIAID recommendations to potentially be cost-effective. One particular lever of cost-effectiveness may be the health utility surrounding the location where someone’s index reaction to peanut occurs. In the case of early introduction, some families and health care professionals may differentially value or more strongly prefer a particular setting for where a first potentially severe index reaction attributable to early peanut introduction occurs—in a clinic under medical supervision vs at home.15 Caregivers and health care professionals who place high value on avoiding an at-home index reaction to peanut, whether mild or severe, and who would rather this reaction occur in a medically supervised setting may strongly prefer the recommended screening approach with reflexive food challenge for modest positive screening results (3- to 7-mm wheal of a peanut SPT), whereas those who value this scenario less or have no preference may opt for at-home introduction. Such differing valuation could drastically affect the cost-effectiveness of screening. Therefore, we undertook this simulation and cost-effectiveness analysis to evaluate the optimal peanut introduction strategy for high-risk infants in the setting of differential potential health utility for medically supervised vs at-home index reactions to peanut.

METHODS This study was deemed exempt from institutional review board approval and informed consent by the 30

Colorado Multiple Institutional Review Board of the University of Colorado because it evaluated simulated cohorts of infants at risk for peanut allergy with the use of aggregate published data as model inputs and did not qualify as human research. The analysis conformed to the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) reporting guideline.16 Decision Model Microsimulations (100 000 per strategy) and cohort analyses were used to evaluate a Markov model of early peanut introduction with and without peanut SPT screening in infants deemed to be at high risk for peanut allergy development per the NIAID guidelines (those with early-onset eczema and/or egg allergy) during an extended 80-year horizon from a societal perspective. An extended time horizon was used to better understand the long-term societal outcomes of screening decisions made during infancy in an allergy considered to be lifelong for most patients.

Infants randomized to screening received a peanut SPT with an initial dichotomous outcome defined at 8 mm, the cutoff in the NIAID guidelines at which an infant is recommended to be diagnosed as allergic and not offered early introduction. A positive test result was considered to be a wheal of at least 8 mm; a negative test result, a wheal of less than 8 mm. Children with an SPT result of 3 to 7 mm underwent supervised peanut challenge in an allergy clinic, and those with an SPT result of less than 3 mm underwent home peanut introduction, per NIAID guidelines.7,14 Figure 1 depicts the following outcomes of screening at the 8-mm threshold: (1) true-positive (sensitized and allergic), (2) false-positive (sensitized but not truly allergic; however, not challenged to determine this), (3) truenegative (not sensitized and tolerant), and (4) falsenegative (skin test <8 mm but allergic). Sensitivity of peanut SPT (≥8 mm) was derived from the HealthNuts cohort and modeled at 0.54 with a specificity of 0.98, the population from which the NIAID guideline cutoff value was obtained.17 In the base-case model, a falsepositive test result (SPT ≥8 mm) led to a diagnosis of peanut allergy without challenge (as per NIAID guidelines)7; however, infants with a false-positive SPT result did not assume the risks of allergic reactions from accidental peanut ingestion. Children with true-positive test results (SPT ≥8 mm) avoided peanut

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Figure 1. Outcomes of Peanut Allergy Screening and Decision Trees of Diagnoses

and entered the natural history model of peanut allergy, whereas those with false-positive results avoiding peanut entered a natural history model that included avoidance under a peanut allergy health state utility but also included the possibility that natural tolerance would be inadvertently discovered. Children with false-negative SPT results were presumed to discover (through supervised or home challenge) that they were allergic within the initial year of the model cycle and

subsequently entered the peanut allergy health state.

Models were evaluated for cost, quality-adjusted lifeyears (QALYs), net monetary benefit, peanut allergic reactions, severe reactions, and deaths due to peanut allergy. Model trackers were used to evaluate episodes of severe allergic reactions and fatalities.12,18-21 A threshold for cost-effective care was set at $100â&#x20AC;Ż000/ QALY.22

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

31

Probabilities Model inputs12,13,17,19-21,23-30 shown in Table 1 included a 14% prevalence of peanut allergy in the high-risk infants during infancy.13,14 Age-adjusted all-cause mortality was incorporated with 2013 US life tables.23 The accidental rate of peanut exposure was 11.7% per year (range, 5.0%-45.0%), with severe allergic reactions occurring in 52.0% of accidental reactions (range, 1.0%-55.0%). Severe reactions on first exposure to peanut occurred in 30.5% of patients (range,

32

5.0%-55.0%), with 8.3% of index reactions involving respiratory or cardiovascular compromise.12,20 Hospitalization was required in 35.0% of patients experiencing severe allergic reactions (range, 5.0%-45.0%).21 Deaths due to food allergy were included (aged 0-19 years, 3.25 [95% CI, 1.73-6.10] per 1 million personyears; aged â&#x2030;Ľ20 years, 1.81 [95% CI, 0.94-3.45] per 1 million person-years).19 A 20% rate of discovery of overdiagnosis (range, 5%-80%) was modeled during the first 20 years of the simulation.

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Costs Costs of living with peanut allergy were expressed in 2018 dollars and discounted at 3% per annum. Jobrelated opportunity costs of caregivers were estimated at $2597 per year.24-26 Direct costs included allergist, primary health care professional, nutritionist, and alternative health care professional visits, self-injectable epinephrine, groceries, and anaphylaxis management (emergency department care and hospitalization), which were obtained from previously published analyses.27-29,31

of discovering false-positive diagnoses, and mortality rates increased to 10-fold in the base-case risk.17 Accidental annual peanut reaction rates were modeled to a lower limit of 1%, with as many as 55% of index reactions resulting in anaphylaxis. Sensitivity analyses excluding job-related opportunity costs and evaluating epinephrine autoinjector costs at $50 per year were also performed.

RESULTS

We used SDs to describe probabilistic determination of uncertainty associated with variation in event rates resulting from linked probabilities of individual outcomes during Monte Carlo simulation. In addition, for probabilistic sensitivity analysis, triangular distributions (minimal, maximal, and mode values specified) were evaluated simultaneously to evaluate certainty of findings.

The simulated population included 100 000 infants with and 100 000 infants without the SPT screening.

During the 80-year time horizon, a no-screening approach dominated SPT screening in high-risk infants for costs (mean [SD], $13 449 [$38 163] vs $15 279 [$38 995]) and QALYs (mean [SD], 29.25 [3.28] vs 29.23 [3.30]). As shown in Table 2, when compared with screening, a no-screening approach resulted in slightly higher rates of allergic reactions (mean [SD], 1.07 [3.15] vs 1.01 [3.02]), severe allergic reactions (mean [SD], 0.53 [1.66] vs 0.52 [1.62]), and accidental anaphylaxis together with index reactions that included respiratory or cardiovascular compromise (mean [SD], 0.50 [1.59] vs 0.49 [1.47]) per patient at risk. However, rates of deaths due to food allergy were similar (and rare). Skin testing led to peanut allergy diagnosis in 6.5% of the screening cohort vs 6.3% of participants in the nonscreening cohort as the model concluded. When modeling the possible protective benefit of screening against an index reaction–associated peanut fatality, a no-screening approach continued to dominate the analyses, even assuming as much as a 1000-fold protection against fatality on the index ingestion associated with screening (screening incremental cost, $1532; effectiveness, −0.018 in the cohort analysis).

Statistical Analysis Data were analyzed from April to May 2019. Univariate deterministic sensitivity analyses were performed on individual variables. Multivariate probabilistic sensitivity analyses (n = 1000) with triangular modal distributions were performed across upper and lower bounds of plausible ranges. Sensitivity analyses included SPT sensitivity ranges from 0.70 to 0.98 and specificities of 0.33 to 0.99, higher baseline prevalence rates of peanut allergy among high-risk infants undergoing early peanut introduction, lower chances

Sensitivity Analyses Because screening may incorporate variations in care that are sensitive to patient preference, additional analyses explored differential caregiver health utility and disutility. In deterministic sensitivity analyses at base-case sensitivity and specificity rates, SPT could be cost-effective (willingness to pay, $100 000/QALY) when applying a very high rate of disutility for a home reaction vs an in-clinic index reaction, in combination with a very high baseline peanut allergy prevalence in the high-risk infant with no peanut exposure (which

Health State Utilities Quality-adjusted life-years were derived from health state utilities for patients living with peanut allergies, discounted at 3% per annum. Carroll and Downs30 assessed health state utility values by standard gamble and time trade-off in 4016 parents or guardians of at least 1 child younger than 18 years recruited at random from multiple sources. Health state utility values for moderate and severe allergic reactions were 0.93 for standard gamble and 0.91 for time trade-off. The disutility of an allergic reaction was −0.09.30 Health state disutility represents a negative health detriment assigned for a particular event relative to the condition of interest. In this case, the disutility translates to approximately 33 days of life in a single year being traded to avoid having an allergic reaction.

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

33

exceeded the base-case utility difference for index in-clinic vs at-home anaphylaxis), and this interaction of these levers is demonstrated in Figure 2. We also considered the threshold for test sensitivity and specificity at which this analysis could be cost-effective.

If an equivalent rate and disutility of accidental and index anaphylaxis was assumed with an 8-mm SPT sensitivity of 0.85 and specificity of 0.98, the screening approach became cost-effective at a peanut allergy prevalence of 36% (eFigure 1 in the Supplement).

Additional deterministic analyses did not demonstrate cost-effective care for a screening approach (Figure 3).

In probabilistic sensitivity analysis during a 20-year time horizon (n = 10 000), a no-screen approach was the optimal strategy in 99.9% at a willingness to pay of $100 000/QALY (eFigure 2 in the Supplement).

Figure 2. Deterministic Sensitivity Analyses

Sensitivity analysis of the interaction between differential disutility of in-clinic or at-home index anaphylaxis and peanut allergy prevalence in high-risk infants at the threshold willingness to pay of $100 000/quality-adjusted life-year.

Health disutility is a negative detriment of an allergic reaction (every −0.1 = 36.5 days of life in a year traded to avoid a reaction).

34

DISCUSSION The previous analysis by Shaker et al14 comparing the US, UK, and Australia and New Zealand approaches calculated the cost to prevent a single case of a severe peanut allergic reaction under the pathway recommended by addendum 1 in the NIAID guidelines to be approximately US $101 963. Despite the lack of benefit, such a decision to screen may be preference sensitive. In a 2018 nationally representative survey of expecting parents (n = 1000) and new caregivers of infants (n = 1000),15 61% had no or minimal concern for their child developing a food allergy, 54% thought early introduction mattered in terms of preventing food allergy development, 31% were willing to introduce peanut before 6 months of age, and 51% were unwilling to allow in-office risk assessments of peanut allergy (eg, allergy testing or oral challenge) before 11 months of age. Although this study was not designed as a formal preference-elicitation study and was performed at the beginning of the life cycle of the NIAID addendum guidelines, this population was representative of caregivers and family potentially having to make such a choice and infers that differential preference may exist regarding early peanut introduction.

This present analysis confirms previous observations that screening high-risk infants for peanut allergy is not cost-effective through a different model incorporating a longer time horizon and the potential for spontaneous discovery of false-positive test results.

Given that change to this policy is unlikely, we took a novel approach of identifying key levers that could influence the cost-effectiveness of the existing NIAID policy, albeit with a narrow potential (and arguably an infeasible one given the specifications of the identi-

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Figure 3. Tornado Diagram of Incremental Net Monetary Benefit (NMB)

Deterministic sensitivity analyses across incremental NMB (a metric that synthesizes cost with monetization of the gain in quality-adjusted life-years [QALYs] so that higher NMB represents a higher-value intervention) for skin prick testing (SPT) vs no testing (willingness-to-pay threshold, $100 000/QALY). ED indicates emergency department; PCP, primary health care professional.

fied levers).14 These levers are a very high prevalence of preexisting peanut allergy in the infants undergoing early introduction, a high disutility for having an at-home index reaction, and enhanced SPT accuracy.

Under most assumptions, even when differential inclinic vs at-home anaphylaxis disutility is modeled, false-positive diagnoses are not subjected to ongoing risks of true peanut allergy, and spontaneous discovery of overdiagnosis is considered, the burden imposed

by screening SPT simply overshadows a no-screening approach in terms of cost and effectiveness assessed

by QALYs. In the present model, rates of anaphylaxis occurring with no screening being performed are marginally greater at a mean (SD) of 0.50 (1.59) vs 0.49 (1.47), and although this difference is not significant, this finding may reflect the current paradigm of advising that infants who are highly sensitized to peanut and have strong positive test results not be offered the

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

35

opportunity to establish oral tolerance to prevent the risk that they may react at ingestion.14,32 However, this approach has not actually proven to be a shared decision consistent with caregiver (as opposed to health care professional) values. In actuality, it may be more accurate to assume that without being given any other option, such children with strong peanut sensitization are highly likely to develop peanut allergy if not offered peanut in the first year of life, as evidenced by the data (and secondary analysis data) from the LEAP study.11 A large contributor to the poor cost-effectiveness of the screening approach relates to the poor accuracy of peanut allergy diagnostic test results in children who have not directly ingested peanut and developed symptoms of an allergic reaction.33,34 For this reason, multiple past food allergy guidelines in the United States and elsewhere have urged caution in testing individuals who have not ever eaten a food before, a situation representing exceptionally low pretest probability at worst and marginal pretest probability (which was elevated to marginal through severe eczema and/ or egg allergy as factors that may increase the odds of someone having peanut allergy compared with the general population) at best.1,35 An optimal screening test should maximize the ability to accurately identify disease that would lead to harm if not otherwise diagnosed early (eg, test sensitivity), because of adverse consequences due to the natural history of the disease or because early management resulting from identification can prevent likely complications of the disease, while minimizing potential false-positive diagnoses and the resulting harm from managing a condition that is not actually present (eg, test specificity). However, it is not clear that such situations exist in the realm of food allergy screening before introduction, given that the test precision is imperfect (the test is most interpretable when results are negative and is difficult to interpret when results are positive, particularly at a 3-mm cutoff) and carries significant risk of false-positive results. Even when a patient is accurately diagnosed with peanut allergy early, no treatment (and certainly no cure) is available yet (nor will be available until the child is 4 years of age).2 Thus, early diagnosis may prevent an index reaction (including a potentially severe one) under certain circumstances as the only tangible benefit and may 36

identify someone for future therapy, but otherwise, early diagnosis increases oneâ&#x20AC;&#x2122;s life span with peanut allergy and adds time under which poor quality of life may develop as a realistic detriment. The advent of an available peanut allergy therapy in younger children might increase the value of peanut allergy screening.

To our knowledge, no published or anecdotal reports exist of deaths due to peanut anaphylaxis during infancy related to early introduction of a small amount of peanut protein (even in an infant considered at high risk for peanut introduction). With an increase in early introduction, ongoing surveillance of fatal food reactions during infancy will be important to continue to evaluate the safety of this practice. In the HealthNuts study, although anaphylaxis rarely occurred, most reactions were mild and cutaneous.12 A similar finding was noted in LEAP, although the trial was right censored from challenge-proven outcomes at an SPT result of 5 mm, as well as in the other early introduction studies.6 Most importantly, no data are available to suggest that death due to food allergy is more likely in this age than any other age.36 One potential approach is to presume that this is a preference-sensitive choice, develop a decision aid, and provide families with a clear understanding of risks and benefits of each approach that allows them to make the best decision for themselves. Although this will not change the fact that, when viewed through an economic lens, peanut allergy screening for early introduction is not cost-effective, it may help offset some potentially lowvalue care if parents can be given more options than the current NIAID guidelines suggest.

Limitations This analysis is limited given that it is a simulation reliant on the quality of its inputs, which in this case come from the LEAP and HealthNuts cohorts.

Because those studies were not conducted in a US population, there could be difficulty with generalizability; nonetheless, they represent the most robust and reasonable inputs for this model.6,12,13 In addition, the precise health utilities for peanut allergy, early introduction, or preference for home vs supervised initial reactions are unknown and need to be established. We explored levers that we believed might be amenable to change, but there could be other levers that we did not explore. We did not include risks of motor vehicle collision fatality associated with transportation for allergy

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

evaluations, which could in fact greatly overshadow any screening-associated risk reduction for death due to anaphylaxis.36-38 Also, as was noted in the previous analysis by Shaker et al,14 we did not model wait lists for health care professionals and reduced access as factors that could affect decision-making and the ability to receive timely care.

CONCLUSIONS In this study, screening for peanut sensitization in high-risk infants and presumptively diagnosing the child with a peanut allergy based on large SPT result size or only providing the option for in-clinic introduction for those with small- to moderate-size SPT results was not found to be cost-effective compared with the general permissive strategy of recommending early introduction at home without any assessment. However, it appears that this strategy could be cost-effective if caregivers have a strong health utility for having an index reaction occur under medical supervision, with a very accurate test result, or at a very high rate of ambient preexisting peanut allergy before early introduction occurs. Further research is needed to better define these key attributes, and presuming differential health utility exists for where caregivers prefer index reactions to occur, a formal decision aid could be of considerable use to help caregivers and health care professionals engage in shared decision-making to facilitate early peanut introduction.

This is an open access article distributed under the terms of the CC-BY License. Š 2019 Greenhawt M et al. JAMA Network Open. Published: December 20, 2019. doi:10.1001/jamanetworkopen.2019.18041 Supplemental content is available here: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2757879 REFERENCES 1. Oria MP, Stallings VA, eds. Finding a Path to Safety in Food Allergy: Assessment of the Global Burden, Causes, Prevention, Management, and Public Policy. Washington, DC: National Academies Press; 2016.

2. Wood RA. Food allergen immunotherapy: current status and prospects for the future. J Allergy Clin

Immunol. 2016;137(4):973-982. doi:10.1016/j. jaci.2016.01.001 3. Sicherer SH, Sampson HA. Food allergy: a review and update on epidemiology, pathogenesis, diagnosis, prevention and management. J Allergy Clin Immunol. 2018;141(1):41-58. doi:10.1016/j. jaci.2017.11.003 4. Greenhawt M. Food allergy quality of life and living with food allergy. Curr Opin Allergy Clin Immunol. 2016;16(3):284-290. doi:10.1097/ ACI.0000000000000271 5. Greenhawt M. Food allergy quality of life. Ann Allergy Asthma Immunol. 2014;113(5):506-512. doi:10.1016/j.anai.2014.06.027 6. Du Toit G, Roberts G, Sayre PH, et al; LEAP Study Team. Randomized trial of peanut consumption in infants at risk for peanut allergy. N Engl J Med. 2015;372(9):803-813. doi:10.1056/ NEJMoa1414850 7. Togias A, Cooper SF, Acebal ML, et al. Addendum guidelines for the prevention of peanut allergy in the United States: report of the National Institute of Allergy and Infectious Diseasesâ&#x20AC;&#x201C;sponsored expert panel. J Allergy Clin Immunol.

2017;139(1):29-44. doi:10.1016/j.jaci.2016.10.010 8. Netting MJ, Campbell DE, Koplin JJ, et al; Centre for Food and Allergy Research, the Australasian Society of Clinical Immunology and Allergy, the National Allergy Strategy, and the Australian Infant Feeding Summit Consensus Group. An Australian consensus on infant feeding guidelines to prevent food allergy: outcomes from the Australian Infant Feeding Summit. J Allergy Clin Immunol Pract. 2017;5(6):1617-1624. doi:10.1016/j. jaip.2017.03.013 9. Turner PJ, Feeney M, Meyer R, Perkin MR, Fox AT. Implementing primary prevention of food allergy in infants: new BSACI guidance published. Clin Exp Allergy. 2018;48(8):912-915. doi:10.1111/cea.13218 10. Abrams EM, Hildebrand K, Blair B, Chang ES; Canadian Paediatric Society Allergy Section. Timing of introduction of allergenic solids for infants at high risk. https://www.cps.ca/en/documents/position/allergenic-solids. Posted January 24, 2019.

Accessed May 10, 2019.

11. Greenhawt M, Fleischer DM, Chan ES, et al.

LEAPing through the looking glass: secondary analysis of the effect of skin test size and age of

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

37

introduction on peanut tolerance after early peanut introduction. Allergy. 2017;72(8):1254-1260. doi:10.1111/all.13100 12. Osborne NJ, Koplin JJ, Martin PE, et al; HealthNuts Investigators. Prevalence of challenge-proven IgE-mediated food allergy using population-based sampling and predetermined challenge criteria in infants. J Allergy Clin Immunol. 2011;127(3):668-676.e2. doi:10.1016/j. jaci.2011.01.039 13. Koplin JJ, Peters RL, Dharmage SC, et al; HealthNuts study investigators. Understanding the feasibility and implications of implementing early peanut introduction for prevention of peanut allergy. J Allergy Clin Immunol. 2016;138(4):11311141.e2. doi:10.1016/j.jaci.2016.04.011 14. Shaker M, Stukus D, Chan ES, Fleischer DM, Spergel JM, Greenhawt M. â&#x20AC;&#x153;To screen or not to screenâ&#x20AC;?: comparing the health and economic benefits of early peanut introduction strategies in five countries. Allergy. 2018;73(8):1707-1714. doi:10.1111/all.13446 15. Greenhawt M, Chan ES, Fleischer DM, et al.

Caregiver and expecting caregiver support for early peanut introduction guidelines. Ann Allergy Asthma Immunol. 2018;120(6):620-625. doi:10.1016/j.anai.2018.03.001 16. Husereau D, Drummond M, Petrou S, et al; CHEERS Task Force. Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement. BMJ. 2013;346:f1049. doi:10.1136/ bmj.f1049 17. Peters RL, Allen KJ, Dharmage SC, et al; HealthNuts Study. Skin prick test responses and allergen-specific IgE levels as predictors of peanut, egg, and sesame allergy in infants. J Allergy Clin Immunol. 2013;132(4):874-880. doi:10.1016/j. jaci.2013.05.038 18. Neuman-Sunshine DL, Eckman JA, Keet CA, et al. The natural history of persistent peanut allergy.

Ann Allergy Asthma Immunol. 2012;108(5):326331.e3. doi:10.1016/j.anai.2011.11.010 19. Umasunthar T, Leonardi-Bee J, Hodes M, et al.

Incidence of fatal food anaphylaxis in people with food allergy: a systematic review and meta-analysis. Clin Exp Allergy. 2013;43(12):1333-1341. doi:10.1111/cea.12211 20. Vander Leek TK, Liu AH, Stefanski K, Blacker B, Bock SA. The natural history of peanut allergy 38

in young children and its association with serum peanut-specific IgE. J Pediatr. 2000;137(6):749755. doi:10.1067/mpd.2000.109376 21. Robinson M, Greenhawt M, Stukus DR. Factors associated with epinephrine administration for anaphylaxis in children before arrival to the emergency department. Ann Allergy Asthma Immunol. 2017;119(2):164-169. doi:10.1016/j. anai.2017.06.001 22. Winkelmayer WC, Weinstein MC, Mittleman MA, Glynn RJ, Pliskin JS. Health economic evaluations: the special case of end-stage renal disease treatment. Med Decis Making. 2002;22(5):417-430. doi:10.1177/027298902320556118 23. Arias E, Heron M, Xu J. United States life tables, 2013. Natl Vital Stat Rep. 2017;66(3):1-64.

24. US Department of Labor, Bureau of Labor Statistics. http://www.bls.gov. Accessed September 28, 2017.

25. Centers for Medicare & Medicaid Services. Physician fee schedule. http://www.cms.gov. Accessed October 3, 2017.

26. Dartmouth-Hitchcock. Doctors office visits. https://www.dartmouth-hitchcock.org/billingcharges/doctors_office_visits_dhmc.html. Accessed October 3, 2017.

27. Gupta R, Holdford D, Bilaver L, Dyer A, Holl JL, Meltzer D. The economic impact of childhood food allergy in the United States. JAMA Pediatr. 2013;167(11):1026-1031. doi:10.1001/ jamapediatrics.2013.2376 28. Patel DA, Holdford DA, Edwards E, Carroll NV. Estimating the economic burden of food-induced allergic reactions and anaphylaxis in the United States. J Allergy Clin Immunol. 2011;128(1):110-115.e5. doi:10.1016/j. jaci.2011.03.013 29. Shaker M, Bean K, Verdi M. Economic evaluation of epinephrine auto-injectors for peanut allergy. Ann Allergy Asthma Immunol. 2017;119(2):160-163. doi:10.1016/j. anai.2017.05.020 30. Carroll AE, Downs SM. Improving decision analyses: parent preferences (utility values) for pediatric health outcomes. J Pediatr.

2009;155(1):21-25, 25.e1-25.e5. doi:10.1016/j. jpeds.2009.01.040 31. Shaker M, Kanaoka T, Feenan L, Greenhawt M.

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Economic outcomes of immediate activation of emergency medical services after epinephrine use for peanut-induced anaphylaxis. Ann Allergy Asthma Immunol. 2019;122(1):79-85. doi:10.1016/j. anai.2018.06.035 32. Abrams EM, Soller L, Singer AG, Fleischer DM, Greenhawt M, Chan ES. Comparison of practice patterns among Canadian allergists before and after NIAID guideline recommendations. J Allergy Clin Immunol Pract. 2019;7(8):2901-2903. e3. doi:10.1016/j.jaip.2019.04.044 33. Flores Kim J, McCleary N, Nwaru BI, Stoddart A, Sheikh A. Diagnostic accuracy, risk assessment, and cost-effectiveness of componentresolved diagnostics for food allergy: a systematic review. Allergy. 2018;73(8):1609-1621. doi:10.1111/all.13399 34. Klemans RJ, van Os-Medendorp H, Blankestijn M, Bruijnzeel-Koomen CA, Knol EF, Knulst AC. Diagnostic accuracy of specific IgE to components in diagnosing peanut allergy: a systematic review. Clin Exp Allergy. 2015;45(4):720-730. doi:10.1111/cea.12412 35. Sampson HA, Aceves S, Bock SA, et al; Joint Task Force on Practice Parameters; Practice Parameter Workgroup. Food allergy: a practice parameter update-2014. J Allergy Clin Immunol. 2014;134(5):1016-25.e43. doi:10.1016/j. jaci.2014.05.013 36. Turner PJ, Jerschow E, Umasunthar T, Lin R, Campbell DE, Boyle RJ. Fatal anaphylaxis: mortality rate and risk factors. J Allergy Clin Immunol Pract. 2017;5(5):1169-1178. doi:10.1016/j. jaip.2017.06.031 37. US Department and Transportation. Traffic safety facts 2016 data. https://crashstats.nhtsa.dot.gov/ Api/Public/ViewPublication/812554. Published May 2018. Accessed June 15, 2019.

38. Shaker M, Briggs A, Dbouk A, Dutille E, Oppenheimer J, Greenhawt M. Estimation of health and economic benefits of clinic versus home administration of omalizumab and mepolizumab [published online October 15, 2019].

J Allergy Clin Immunol Pract. doi:10.1016/j. jaip.2019.09.037

ARTICLE INFORMATION Corresponding Author: Matthew Greenhawt, MD, MBA, MSc, Section of Allergy and Immunology, Food Challenge and Research Unit, Children’s Hospital Colorado, 13123 E 16th Ave, Aurora, CO 80045 (matthew.greenhawt@childrenscolorado.org).

Author Contributions: Dr Greenhawt had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Both authors.

Acquisition, analysis, or interpretation of data: Both authors.

Drafting of the manuscript: Both authors.

Critical revision of the manuscript for important intellectual content: Shaker.

Statistical analysis: Both authors.

Obtained funding: Greenhawt.

Administrative, technical, or material support: Both authors.

Supervision: Both authors.

Conflict of Interest Disclosures: Dr Greenhawt reported serving as an expert panel and coordinating committee member of the National Institutes for Allergy and Infectious Disease–sponsored Guidelines for Peanut Allergy Prevention; consulting for the Canadian Transportation Agency, Thermo Fisher Scientific, Intrommune Therapeutics, and Aimmune Therapeutics; serving on physician/medical advisory boards for Aimmune Therapeutics, DBV Technologies, Sanofi Genzyme, Genentech, Inc, Nutricia, Kaleo, Inc, Nestlé, Aquestive Therapeutics, Inc, Allergy Therapeutics, AllerGenis, Inc, Aravax, GlaxoSmithKline, Prota, and Monsanto; serving on the scientific advisory council for the National Peanut Board; receiving honoraria for lectures from Thermo Fisher Scientific, Aimmune Therapeutics, DBV Technologies, BEFORE Brands, Inc, multiple state allergy societies, the American College of Allergy Asthma and Immunology, and the European Academy of Allergy and Clinical Immunology; serving as an associate editor for the Annals of Allergy, Asthma, and Immunology; and serving as a member of the Joint Taskforce on Allergy Practice Parameters. Dr Shaker reported having a sibling who is chief executive officer of Altrix Medical, LLC; and serving as a member of the Joint Taskforce on Allergy Practice Parameters.

Funding/Support: This study was supported by grant 5K08HS024599-02 from the Agency for Healthcare Research and Quality (Dr Greenhawt).

Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

39

Evaluating Improvements and Shortcomings in Clinician Satisfaction With Electronic Health Record Usability Kylie M. Gomes, MS(1); Raj M. Ratwani, PhD(2) 1. University of Virginia, Charlottesville, Virginia; 2. MedStar Health National Center for Human Factors in Healthcare, MedStar Health, Georgetown University School of Medicine, Washington, DC

INTRODUCTION

proval was not required for this study because these are publicly available data sets that do not contain With the widespread adoption of electronic health protected human participant information. This report records (EHRs), there is increased focus on addressing followed the Strengthening the Reporting of Observathe challenges of EHR usability, ie, the extent to which tional Studies in Epidemiology (STROBE) reporting the technology enables users to achieve their goals guideline. effectively, efficiently, and satisfactorily.1 Poor usability is associated with clinician job dissatisfaction and We identified the 70 EHR vendors with the most 2-4 burnout and could have patient safety consequences. attestations to meaningful use from health care facilities between July 1, 2016, and April 30, 2018.

The US Department of Health and Human Services For inclusion in analysis the vendor must have had Office of the National Coordinator for Health Informa- an EHR product with computerized provider order tion Technology established safety-enhanced design entry functionality, certified according to the safetycertification requirements for EHRs to promote usenhanced design criterion, and a reported SUS score ability. These requirements stipulate that vendors must for the 2014 and 2015 certification requirements. For conduct and report results of formal usability testeach vendor, the usability report for the most recent ing, including measuring satisfaction with the EHR version of the product meeting the 2014 certification system.5 Results are publicly available. While some requirements (ie, before January 14, 2016, when the vendors use a 5-point, ease-of-use rating scale, most 2015 certification requirements became effective) and vendors use the system usability scale (SUS), which is the usability report for the most recent version of the a validated posttest questionnaire that measures user product meeting the 2015 certification requirements satisfaction with product usability.6 The questionnaire were retrieved, and the SUS scores were analyzed. A provides a score (range, 0-100) based on a participaired t test, with a 2-tailed P < .05 indicating statistipant’s rating of 10 statements regarding a product’s cal significance, was used to determine differences in usability.6 Higher scores indicate greater satisfaction SUS scores between 2014 and 2015, with means and with usability.6 Based on an analysis of more than 200 standard deviations reported. All statistical analyses studies of various products in various industries, an were performed with SPSS statistical software version SUS score of 68 is considered the average benchmark, 25 (IBM Corp). and an SUS of 80 is considered the above-average benchmark.6 Recognizing the importance of satisRESULTS faction with EHR usability to clinician burnout and patient safety, reported product 2015 SUS scores for A total of 27 vendors met the inclusion criteria. Mean EHR systems were compared with 2014 SUS scores (SD) SUS scores for 2014 and 2015 products were and with benchmarks to evaluate whether satisfaction not statistically different (73.2 [16.6] vs 75.0 [14.2]; is improving.2-4 t26 = 0.674; P = .51). Comparing 2014 products to benchmarks, 9 (33%) were below the average benchMETHODS mark SUS score of 68, 18 (67%) were at or above average, and 11 (41%) met or exceeded the abovePer Common Rule, institutional review board apaverage benchmark score of 80 (Figure). For 2015 40

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Figure. Comparison of System Usability Scale (SUS) Scores for 2014 and 2015 Certified Products by Vendor

Vendor-reported electronic health record (EHR) SUS scores for 2014 and 2015 certified products are compared with average benchmark (dotted line) and above-average benchmark (solid line) SUS scores.

products, 7 (26%) were below the average benchmark, 20 (74%) were at or above average, and 12 (44%) met or exceeded the above-average benchmark. Between 2014 and 2015, SUS scores for 12 products (44%) decreased, 13 (48%) increased, and 2 (7%) were unchanged.

DISCUSSION There was no statistical improvement in EHR SUS scores between products certified according to 2014 and 2015 standards. One-third of 2014 products and one-quarter of 2015 products fell below the average benchmark SUS score. Despite the implications of EHR dissatisfaction on clinician burnout and patient safety, SUS scores decreased for 44% of vendors from 2014 to 2015.2-4 This study has limitations. Vendor-reported SUS scores may not reflect satisfaction with implemented EHRs, and only a subset of vendors were analyzed because of differences in methods for measuring satisfaction.

Based on vendor-reported SUS scores, clinician satis-

faction with EHR usability is not improving for many widely used products. An increased focus on clinician end users during product design and development as well as optimized certification requirements are needed to improve usability.

This is an open access article distributed under the terms of the CC-BY-NC-ND License. Š 2019 Gomes KM et al. JAMA Network Open. Published: December 13, 2019. doi:10.1001/jamanetworkopen.2019.16651 REFERENCES 1. International Organization for Standardization.

Ergonomics of human-system interaction, part 11: usability: definitions and concepts. https://www. iso.org/obp/ui/#iso:std:iso:9241:-11:en. Accessed July 22, 2019.

2. Babbott S, Manwell LB, Brown R, et al. Electronic medical records and physician stress in primary care: results from the MEMO Study. J Am Med Inform Assoc. 2014;21(e1):e100-e106. doi:10.1136/amiajnl-2013-001875 3. Friedberg MW, Chen PG, Van Busum KR, et al.

Factors Affecting Physician Professional Satis-

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

41

faction and Their Implications for Patient Care, Health Systems, and Health Policy. Santa Monica, CA: The RAND Corporation; 2013.

4. Howe JL, Adams KT, Hettinger AZ, Ratwani RM. Electronic health record usability issues and potential contribution to patient harm.

JAMA. 2018;319(12):1276-1278. doi:10.1001/ jama.2018.1171 5. Office of the National Coordinator for Health Information Technology (ONC), Department of Health and Human Services (HHS). 2015 edition Health Information Technology (Health IT) certification criteria, 2015 edition base electronic health record (EHR) definition and ONC Health IT certification program modifications: final rule.

Fed Regist. 2015;80(200):62601-62759.

6. Lewis JR, Sauro J. Item benchmarks for the system usability scale. J Usability Stud.

2018;13(3):158-167. http://uxpajournal.org/wpcontent/uploads/sites/8/pdf/JUS_Lewis_May2018. pdf. Accessed October 25, 2019.

ARTICLE INFORMATION Corresponding Author: Raj M. Ratwani, PhD, MedStar Health National Center for Human Factors in Healthcare, MedStar Health, 3007 Tilden St, Ste 7M, Washington, DC 20008 (raj.m.ratwani@medstar.net).

Author Contributions: Ms Gomes had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Both authors.

Acquisition, analysis, or interpretation of data: Both authors.

Drafting of the manuscript: Both authors.

Critical revision of the manuscript for important intellectual content: Gomes.

Statistical analysis: Gomes.

Obtained funding: Ratwani.

Administrative, technical, or material support: Ratwani.

Supervision: Ratwani.

Conflict of Interest Disclosures: Dr Ratwani reported receiving grants from the Agency for Healthcare Research and Quality outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported by grant number R01 HS025136 from the Agency for Healthcare Research and Quality to Dr Ratwani.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

AAMCN Would Like to Recognize Our Corporate Partners Avedro Chamberlain University Exact Sciences Foundation Medicine, Inc. Gilead Sciences, Inc. Guardian Life Insurance Home Instead Senior CareÂŽ

42

Humana, Inc. Mallinckrodt Pharmaceuticals Morgan Consulting Resources, Inc. Novocure TCS Healthcare Woundtech

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Managed Care UPDATES

Health Outcomes Among Long-term Opioid Users With Testosterone Prescription in the Veterans Health Administration Key Points: Question - What are the health outcomes among long-term opioid users who receive testosterone treatment compared with opioid users who do not?

Findings - In this cohort study of 21 272 male long-term opioid users with testosterone deficiency, those who received opioids plus testosterone therapy had significantly lower all-cause mortality and lower incidence of major adverse cardiovascular events, anemia, and femoral or hip fractures than their counterparts who received opioids only in covariate-adjusted and propensity score–matched models.

Meaning - This study’s findings suggest that receiving opioids plus testosterone treatment is associated with lower all-cause mortality and a lower incidence of other adverse health outcomes among men with opioidinduced androgen deficiency. Read more at https://bit.ly/38ApwUm Association of Race and Socioeconomic Status With Colorectal Cancer Screening, Colorectal Cancer Risk, and Mortality in Southern US Adults Key Points: Question - What are the associations between colorectal cancer screening modalities with colorectal cancer risk and mortality among African American individuals and individuals with low socioeconomic status, who more often face barriers to screening and experience colorectal cancer health disparities?

Findings - In this cohort study of 47 596 adults, colorectal cancer screening was significantly associated with reduced colorectal cancer risk and mortality. Use of colonoscopy was associated with a 45% reduced incidence of colorectal cancer, use of sigmoidoscopy with a 34% reduced incidence, and stool-based tests with a 25% reduced incidence compared with no screening.

Meaning - The large colorectal cancer disparities experienced by individuals with low socioeconomic status and African American individuals may be lessened by improving access to and uptake of colorectal cancer screening. Read more at https://bit.ly/3aECdiM

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

43

Comparison of Machine Learning Methods With Traditional Models for Use of Administrative Claims With Electronic Medical Records to Predict Heart Failure Outcomes Key Points: Question - Can prediction of patient outcomes in heart failure based on routinely collected claims data be improved with machine learning methods and incorporating linked electronic medical records?

Findings - In this prognostic study including records on 9502 patients, machine learning methods offered only limited improvement over logistic regression in predicting key outcomes in heart failure based on administrative claims. Inclusion of additional predictors from electronic medical records improved prediction for mortality, heart failure hospitalization, and loss in home days but not for high cost.

Meaning - Models based on claims-only predictors may achieve modest discrimination and accuracy in prediction of key patient outcomes in heart failure, and machine learning approaches and incorporation of additional predictors from electronic medical records may offer some improvement in risk prediction of select outcomes. Read more at https://bit.ly/2RMUVfc Association of a Targeted Population Health Management Intervention with Hospital Admissions and Bed-Days for Medicaid-Enrolled Children Key Points: Question - Is a targeted population health management intervention developed for children enrolled in Medicaid and cared for in a large pediatric health system associated with changes in hospital admissions or bed-days?

Findings - In this quality improvement study using difference-in-differences analysis of Medicaid-enrolled children, children exposed to an integrated population health management program experienced a reduction of 0.39 monthly admissions and 2.20 monthly bed-days per 1000 children compared with similar children in the community who were not exposed to the program. Annualized, these differences could translate to a reduction of 3600 bed-days for a population of 93â&#x20AC;Ż000 children eligible for Medicaid.

Meaning - Mobilizing interdisciplinary care teams for targeted children with high risk and spreading registrybased information technology tools across a Medicaid population may provide a scalable strategy for other health systems that aim to improve the value of services provided to this population. Read more at https://bit. ly/30Ommtj Association of Mandatory Bundled Payments for Joint Replacement With Use of Postacute Care Among Medicare Advantage Enrollees Key Points: Question - Was the Comprehensive Care for Joint Replacement (CJR) program associated with changes among patients enrolled in Medicare Advantage plans, privately managed care plans that insure 34% of Medicare patients and were not participants in the CJR program?

Findings - In this cohort study of more than 1.5 million patients, the CJR program was associated with a 5.6% reduction in institutional postacute care days among traditional Medicare patients and a 2.5% reduction for Medicare Advantage patients, indicating that the policy change may have affected more patients than previously anticipated.

Meaning - Alternative payment models in traditional Medicare may affect care in the Medicare Advantage program; therefore, evaluations of the CJR program that exclude Medicare Advantage may not capture the full consequences of the policy. Read more at https://bit.ly/2TRGfyo 44

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Congratulations to the Newly Certified Managed Care Nurses (CMCNs)!

Linda Briscoe, RN, CMCN Kimberly Brookshear, RN, CCM, CRRN, CMCN Gina F Browne, RN, CMCN Cara M Burgos, RN, BSN, ACM-RN, CMCN Brittany Cancienne, RN, CMCN Fiona H Cayer, BSN, RN, CCM, CMCN Alexandria B Comrie, RN, CMCN Antoinette Credit, LPN, CMCN Tina Darwish, LVN, NCP, CMCN Laurie J Ellis, RN, CMCN Katrina M Etter, BSN, RN, CMCN Fonda Sue Ferrari, RN, CMCN Mirlande Frederic-St. Hilaire, BSN, RN, CMCN Tammy D Gardner, RN, CMCN Denine Hamp-Mondragon, RN, CMCN Alexanrda Jones, RN, CMCN Kali Kivi, RN, CMCN Mary Ann Kovalchin, BSN, RN, CMCN Scott Latham, RN, CMCN Angel Lemoine, RN, CMCN Giovanni Maya, MSN, RN, CMCN

Susan M McCoy, RN, BSN, CCM, CCDS, CMCN Beth A McGovern, RN, MSN, CCM,CMCN Kemisha M McMullen, RN, CMCN Penny A Mobberley, RN, BSN, CMCN Olusegun Olumuyide, RN, CMCN Claudin Pierre-Louis, RN, CMCN Stana Popovac, LVN, CMCN Catherine Arcel Roma, MSN-Ed, BSN, RN, CMCN Lisa Rudd, RN, CMCN Monica L Scott, BSN, RN, CMCN Joia M Shafer, RN, BSN, CMCN Deborah L Smith, LPN, CMCN Charity Smith-Alexander, RN, CMCN Marie Souffrant, RN, CMCN Janet Stevenson, LVN, CMCN Amy J Vasconcellos, RN, CMCN Cara S Voorhorst, RN, CMCN Kelley A Wood, RN, CMCN Karen G Zeller, BSN, RN, CMCN

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

45

Welcome New AAMCN Members!

Susan Albright Rosalene Alexander Denise Alves Ashleigh Andre Marianne Arcania Bethany Archer Sheila Asare Stephanie Babcock Julie Baker Zinetta Baker, RN Nakisha Barr Camille Bascom Collette Bates Dwia Beal Deborah Boe Maria Boyd Kimberly Brandt Linda Briscoe, RN, CMCN Jennifer Brock David Brown Virginia Bucholtz Laurie Bushek Nancy Claflin, RN, PhD, FNAHQ Davina Coker David Cole, RN Rosemond Cole, RN Kimberly Collins Regina Cook Maria Cosmiano Jayne Cox Denise Crandall Theresa Dallas Amanda Davis Kaynetha Davis Allison Dawson Heatherlee Smith Dawson Randy Dellinger Kathleen Diaz Jaime Dixon Renee Doan Kimberly Driessen Casey Dunham Linda Duvall Kayla Fazi Nicole Fenimore Regina Fenner 46

Kimberly Fergerson Erin Ferraro Karen Fitzpatrick Matt Flemr Brittany Friend Elizabeth Galloway Michele Garrett Lisa Garrett-Stielau Christine Geen Matthew Green Giselle Grob Lisa Hamilton Melissa Heath Cheryl Hebert Barbara Henry Cortina Herbert Michael Herbert Dana Herman Rochelle Hickman John Hill Stacie Howe Sharon Huerta Nicholas Hummerich Caroline Ikpeze Angie Johnson Barbara Johnson Glemel Jones Anne Kamau Brenda Keaton Leslie Kendrick Bakary Kinteh Susan Lawrence Ann Learned Viktoriya Lee, RN Cara Levy Leslie Lockwood Kathryn Loewke Lorraine Lubba Sara Luzunaris Cassandra Machler Sarah Mahaffey, PN Kimberly Maxwell, RN Christine McCary Justin McClure Rebecca Meisenhelder Susan Mellott, RN, PhD, FNAHQ

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

Angela Miller Karen Mitchell Sandra Munson Katherine Murphy Kelly Mytych Poonam Nandal Ann Nemirovsky Rhonda Nohava Charles Oliver Amber Ondriezek Christine Ordonez, VN Amen Otunba Chloe Ovinsky Hariett Parker, LVN Francesca Patarino Misha Patel, LPN Emily Payne Elba Perez, RN Kimberly Peters Tara Phillips Leah Pollard Victoria Pompey Encalade Melissa Posey Aaron Primm Melissa Rackard Chermaine Ramos Debra Reiland Tricia Reuter July Rivera Karen Robinson Opal Robinson, BRN Rachel Rush Melissa Saddler Kathy Sankey Christina Santelli Paula Schwarte Elizabeth Sharp Lisa Sheffield-Howell Sandra Sherman, RN Tracy Shultz-Halpern Alyssa Shumway Wendy Sidorenko Cathleen Smith Angela Stacy Katherine Stefaniw Katherine Steinberg Christina Subia Tomasz Suchodolski Masako Suzuki

Belinda Sweet Rabiatou Sylla Justine Tallon-Satink Erin Taylor Pamela Taylor Elizabeth Thompson Chepchir Tirop, RN Alana Tomlinson Kathleen Tornow, RN, FNAHQ, PhD Kyla Uâ&#x20AC;&#x2122;Ren Jill VanAalst Kelly VanParys Carmen Vasquez Laura Vaughn Christine Vicens Lisa Waker-Carter Richard Ward Shanda Watson Amy Watts Claudeus White Amy Wilson Elizabeth Wilson, RN Robin Winskas, RN Jamie Wolfe Heather Woodruff Susan Wyffels Twyla Yezsin Danielle Zander, RN Roxanne Zupparo

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

47

48

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

49

50

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

www.aamcn.org | Vol. 7, No. 1 | Journal of Managed Care Nursing

51