Quality Outlook | Summer 2014 Edition by Premier, Inc.

QUALITY

STRATEGIC LEADERSHIP FOR FINANCIAL AND CLINICAL HEALTHCARE EXECUTIVES • A TWELVE MONTH OUTLOOK • SUMMER 2014

QUALITY OUTLOOK

FOR FURTHER INFORMATION To learn more about this publication, please visit premierinc.com/economicoutlook, or email economicoutlook@premierinc.com.

13034 Ballantyne Corporate Place Charlotte, NC 28277

444 N Capitol Street NW, Suite 625 Washington, DC 20001-1511

T 704 357 0022

T 202 393 0860 F 202 393 0864

PREMIERINC.COM

PUR4366 07/14

T W E LV E M O N T H O U T LO O K | SUMMER 2014

QUEST® Results

Drug Shortages

The Rise of Accountable Care

QUALITY

LETTER 01

EXECUTIVE LETTER

First, Do No Harm Mike Alkire, chief operating officer, Premier, Inc.

FEATURES | OUT OF HARM’S WAY

04 ACUTE RENAL FAILURE IN HOSPITALIZED PATIENTS

PERSPECTIVES 34 The rise of accountable care

08 16 OUT OF HARM’S ADDRESSING WAY: QUEST OPIOID RISKS RESULTS

20 IMPACT OF THE ONGOING DRUG SHORTAGE

30 CLOSED APIs: THERE’S A FEE FOR THAT

TRENDS 46 Partnering to improve community health analytics 48 Picture perfect: Reducing harm in radiology

OUTLOOK LEADERSHIP

MANAGING DIRECTOR Kayla Sutton

EDITORIAL STAFF

E XECUTIVE SPONSORS Mike Alkire, chief operating officer Durral Gilbert, president, supply chain services Amy Denny, vice president, strategy, supply chain services A special thanks to Madison Cyphers, Tina Harlan, Tim Lowe, Jill Mayrand, Dave Natale, Kellie Webb and Laura Yandell for their contributions to this edition of the Quality Outlook.

DESIGN AND PRODUCTION Chris Cardelli, director, creative services Sung Ginader, senior graphics designer, creative services Dave Dixon, associate graphics designer, creative services EDITORIAL SUPPORT Amanda Forster, vice president, public relations Alven Weil, director, public relations Bryan Alsop, senior manager, corporate communications Megan Jarrell, intern, Outlook

About the cover As the cover suggests, this edition of the Quality Outlook is intended to provide our readers with the insights and actionable knowledge needed to keep their patients “out of harm’s way.” Providers and patients alike understand that navigating the healthcare system can be complicated, and sometimes even risky. Although hospitals and other facilities often provide high-quality lifesaving care, we must recognize that the increasingly complex healthcare system that exists today can expose patients to unintended harm. However, identifying areas where safety and quality can be improved, and taking conscious action to eradicate and prevent harm can help healthcare stakeholders steer clear of the pitfalls and find the right path to exceptional quality of care and service for patients.

About the publication Stemming from Premier’s flagship publi-

A key aspect of the long-term strategy for

cation, the Economic Outlook, and grow-

the Outlook is to collaborate with internal

ing interest from our membership and the

and external subject matter experts to build

industry, we introduced our first Quality

consensus from diverse points of view.

Outlook in 2013. The Quality Outlook high-

The publication harnesses the expertise of

lights emerging industry trends, specific to

our network of healthcare leadership to

quality, safety and performance improve-

illuminate best practices and strategies

ment, impacting our membership and

needed to drive performance improvement.

shaping the healthcare landscape. Through

We strive to provide our members and

publication of the Quality Outlook, we seek

healthcare organizations with valuable,

to expand the breadth of information and

timely information and business intelli-

thought leadership we provide to alliance

gence derived from the industry’s most

members and other industry stakeholders.

progressive participants.

We welcome your comments and questions. For additional information, please email economicoutlook@premierinc.com. premierinc.com/economicoutlook outlookmarketplace.hostedbywebstore.com

EX E C U TI V E LE TTER

First, Do No Harm

an imprecise science. To do it correctly, hospitals often require real-time data feeds, predictive analytics and alert capabilities to notify clinical staff of potential problems. To give an example, Premier’s QUEST® collaborative has been evaluating harm for four years by examining 23 distinct measures of “potential harm” spanning across the hospital inpatient setting. Outliers are identified through QualityAdvisor™, and real-time alerts are set up through

MEMBERS OF THE PREMIER ALLIANCE,

SafetyAdvisor™. Hospitals track their performance over time using the QUEST harm reports, and providers exchange ideas on the best strategies for harm prevention through sprints and best practice forums.

hen our ancestors looked to the heavens,

In integrating a constellation of data sources and learning

they had a lot of unanswered questions.

modalities, QUEST hospitals have steadily improved in

Why did some stars appear in every season

reducing unjustified variation in harm rates, preventing an

and others only sometimes? Why did some

estimated 17,991 potential patient safety events.

stars disappear? Which stars could be used as guides? Although improvements overall have been steady, big

They didn’t have the tools to make sense of the pinpricks of light

reductions have been achieved in specific harm measure areas,

above. They had only a vague impression of their meaning.

including: • Central line (central venous catheter) associated blood

To make sense of it all, they organized the stars to create a clearer picture of the universe. Taking abstract inputs – each point of light – they developed constellations to connect the dots, paint a picture and create a map of the stars.

stream infection (CLABSI) rates by 82 percent; • Injury rates while in the hospital, including falls, by 71 percent; • Pressure ulcer rates by 65 percent; • Staph (septicemia) infection rates by 36 percent;

In today’s healthcare environment, we, too, are searching for

• Ventilator-associated pneumonia rates by 23 percent;

answers. With the market so in flux, we need a way to organize

• Catheter-associated urinary tract infection (CAUTI) rates

all our resources and all our data in a way that tells a story and guides action. We need our own constellations to make sense of

by 19 percent; and • Birth trauma (perineal laceration) rates by 18 percent.

the vast challenges we face, using them to create a map for total quality improvement.

Individual hospitals are also drawing constellations, pulling together disparate teams and methods to make a difference in

This is particularly the case when we look at harm. Although

patients’ lives.

preventing harm events continues to be a major focus for hospitals, proactively identifying and tracking inpatient harm is OUTLOOK

QTR 3.14

EX E C U TI V E LE TTER

First, Do No Harm

Consider the obstetrics ward of a hospital. Few areas are

Since implementing these processes, coupled with continuous

more complex, requiring care providers to treat two distinctly

monitoring of its successes through the QUEST harm

different people – simultaneously. Although harm events are

dashboards, Memorial has seen a three-fold decrease in post-

rare and often unavoidable, preventable events are usually the

operative respiratory failure.

result of communication gaps between care providers. When it comes to the challenge of harm prevention, our To prevent these gaps, Summa Akron City Hospital (Akron, OH)

alliance is turning the “impossible” into “it is” possible. We

is participating in the Premier Perinatal Safety Initiative, pulling

are integrating fragmented data sets and connecting the

together multi-disciplinary teams, data sets and tactics to ensure

dots to form solutions. Together, we have enabled a network

seamless communication and effective decision making.

of interconnections and insights that are serving as our constellations. And with these guideposts in place, each of our

It relies on a teamwork system originally developed by the

members will continue to be that shining example of excellence

Department of Defense designed to produce highly effective

in the healthcare universe.

medical teams that optimize the use of information, people and resources to achieve the best clinical outcomes for patients. Their teams perform drills and simulations using lifelike mannequins that are programmed to transition through the various stages of

– MIKE ALKIRE Chief operating officer

/ Premier, Inc.

labor and delivery that mothers and babies undergo during the birth process. They also practice the delivery of care bundles, or a set of processes that helps clinical staff remember to take all the necessary steps to provide optimal care to every patient, every time. And they harness data to track their performance over time, identify outlier cases and address root causes. In taking a holistic approach, Summa is drawing new connections, making sense out of the science and making a difference for mothers and babies alike. Memorial Health University Medical Center (Savannah, GA) is also drawing its own constellations to prevent harm. There, clinical teams conduct concurrent reviews of all medical records to identify patients who may be at risk for post-operative respiratory failure. They use a questionnaire that helps them identify those with obstructive sleep apnea, an often undiagnosed condition that significantly adds to the risk of death among those who must be sedated and intubated prior to surgery. Using this tool, Memorial is able to treat the sleep apnea prior to surgical procedures, and carefully monitor patients post-surgery with additional respiratory care bundles that ensure better outcomes.

We need our own constellations to make sense of the vast challenges we face, using them to create a map for total quality improvement.

FEATURES Acute renal failure in hospitalized patients, 4 Out of harm’s way: QUEST results, 8 Addressing opioid risks, 16 Impact of the ongoing drug shortage, 20 Closed APIs: There’s a fee for that, 30

Acute renal failure in hospitalized patients

A source of significant, potentially avoidable morbidity and costs

FEATURES

T F

Richard Bankowitz, MD

Chief medical officer / Premier, Inc.

eing admitted to a hospital in the United States can be risky. Though our nation’s hospitals provide

exceptional quality and in many cases life-saving care, we must recognize that today’s increasingly high-tech and complex care delivery system also exposes patients to unintended harm. Since the publication of the Institute of Medicine’s landmark report “To Err is Human,” which brought this phenomenon to national attention, many studies have shown that hospitalacquired conditions (those that result

In a sample 2,510 of 500,000 excess deaths random hospital admissions, hospitalAn additional acquired $472 million in the cost of care conditions resulted in: (five excess deaths per 1,000 admissions)

(almost $1,000 per admission)

from the care delivery process itself) occur with considerable frequency, impact clinical outcomes and add to the

a significantly increased risk of death

cost of care.

and higher cost of care. The Centers for Medicare & Medicaid

In a recent study published in the

any in a set of 13 conditions. The 2012 CMS Value-Based Purchasing program takes this a step

Services (CMS) has introduced three

further and places up to 2 percent of

Premier researchers concluded that

separate and overlapping mechanisms

hospital reimbursement at risk, based

a potential medical complication,

to impose financial consequences for

on performance on metrics that include

not present at the time of admission,

cases involving a subset of hospital-

a harm measure. CMS also recently

occurred in almost 16 percent of

acquired conditions. Beginning in 2010,

introduced financial penalties for

hospitalizations. Many of these

hospitals were ineligible for additional

hospitals in the lowest-performing

complications were associated with

reimbursement for the occurrence of

quartile of two composite harm

American Journal of Medical Quality,

OUTLOOK

QTR 3.14

FEATURES

T F

Fig.1

Patients with acute renal failure and other hospital-acquired conditions ACUTE RENAL FAILURE Number

Percent

Acute renal failure – only

80,932

52.2%

Respiratory failure

25,989

16.8%

Sepsis/(SIRS)

17,634

11.4%

Hypotension

16,748

10.8%

Pneumonia

12,669

8.2%

Encephalopathy

9,698

6.3%

Septic shock

8,904

5.7%

Aspiration pneumonia

6,758

4.4%

Cardiac arrest

5,309

3.4%

Embolism/thrombus

5,004

3.2%

Acute necrosis of the liver

4,316

2.8%

Acute myocardial infarction

4,276

2.8%

Cardiogenic shock

3,903

2.5%

Drug-induced neurologic disorder/delirium

3,885

2.5%

Gastrointestinal (GI) ulceration & hemorrhage

3,334

2.2%

Other shock

3,290

2.1%

C. difficile enteritis

3,115

2.0%

TOTAL

155,012

100.0%

Source: A Premier, Inc. maintained database

HOSPITAL-ACQUIRED CONDITIONS

measures. Some events, such as central

Patients (PFP) initiative have seen

line-associated blood stream infections

ongoing improvement in several areas

thromboembolism (48.8 percent);

(CLABSI), could potentially result in a

targeted for harm reduction. Premier

• Ventilator-associated pneumonia

triple penalty for a facility.

currently operates one of the largest

The good news is that the frequency

PFP hospital engagement networks

• Potentially preventable venous

(44.5 percent); and • Pressure ulcers stages three and four

of many of these hospital-acquired

(HEN), with 450 facilities participating.

(41.1 percent).

conditions has begun to decrease.

As of December 2013, the Premier HEN

Despite these successes, much

Increased national attention on such

has met its 40 percent improvement

work remains to be done. There are

things as CLABSI has resulted in greater

goals in reducing:

many high-impact hospital-acquired

use of infection prevention protocols

• Elective deliveries (70.8 percent);

conditions that occur frequently but

and enhanced awareness of proper

•C LABSI in small hospitals

are not yet receiving much national

technique. For some hospitals, the incidence of CLABSI is nearly zero. Similarly, Premier members participating in CMS’ Partnership for

(64.7 percent);

attention. For example, acute renal

•C LABSI in PICU/NICU (52.8 percent);

failure (ARF) has not been the subject of

•C LABSI in ICU and wards

any national payment focus or a national

(43.2 percent);

collaborative. However, that is set to

FEATURES

T F

targeted measure in its new advanced

Factors of acquired acute renal failure

possible to determine what might

Partnership for Patients initiative.

To get a better understanding of

account for the 52 percent of patients

Ongoing research at Premier has

what might be responsible for the

who experienced ARF as a sole acquired

determined that hospital-acquired ARF

relatively high incidence of ARF,

condition. Adverse drug events, contrast

is a significant complication of care,

Premier researchers examined data

media-associated renal failure, and use

contributing to increases in mortality,

from hospital discharges in 2013 to

of nephrotoxics are not conditions that

cost and length of stay.

determine which hospital-acquired

are easily coded into an ICD-9 discharge

conditions were also present in patients

summary. We believe, however, that

admissions to the hospital resulted in

who developed renal failure while

within that 52 percent majority, a deeper

ARF (the most frequently occurring

hospitalized.

review of the data will reveal many

change as CMS introduces ARF as a

In a recent study, 1.7 percent of all

hospital-acquired condition).

Out of 9.16 million inpatient hospital

Based on this data alone, it is not

potentially avoidable causes.

Study results showed patients who

admissions, 155,012 (1.7 percent)

experienced renal failure in the hospital,

had renal failure that occurred in the

understanding of common, yet seldom

all other things accounted for, had a 69

hospital. The majority of these patients

highlighted, conditions. If even 50

percent higher likelihood of death than

(52.2 percent) had no other acquired

percent of the cases of acquired ARF can

would be the case had this not occurred.

conditions from a list of 138 initially

be prevented, an enormous amount of

Similarly, hospital-acquired ARF

identified (Figure 1). With the remaining

pain, suffering and waste can be removed

patients stayed in the hospital nearly

patients, researchers were not able to

from the system.

nine days longer on average than those

determine which conditions preceded

who did not experience renal failure and

ARF, since multiple conditions can be

level of these infections that is

added an average of $19,725 each to the

coded for the same admission.

unavoidable. Had someone suggested

cost of hospitalization. The aggregate

From the study results, we can make

Our goal is to develop a better

Many believe that there is a baseline

10 years ago that we could eliminate all

additional cost of care was $26.75

certain inferences. Acquired ARF is

central line infections occurring within

million for the study sample.

frequently associated with respiratory

a hospital, it would have been met with

conditions. Respiratory failure can

skepticism. Yet a number of our health

more striking is that acquired ARF is

result from renal failure or both may be

systems today have gone a year or more

often preventable. Patients develop

caused from a third process, including

without any central line infections. We

renal failure in the hospital for a

shock. Thus, the respiratory failure

remain optimistic the same can be true

variety of reasons, many related to the

itself may not be causal but may have

of more complex conditions, including

treatment itself. For example, some

arisen from pneumonia or aspiration

renal failure. Improvement begins with

patients receiving contrast media for a

pneumonia that preceded renal failure.

an understanding of the causal drivers

diagnostic exam can experience acute

Similarly, sepsis and septic shock are also

of the condition, which will allow

renal failure, especially if they are older,

commonly associated with respiratory

us to devise strategies for enhanced

have pre-existing renal conditions, or

failure and may precipitate renal failure.

observation and early intervention.

What makes these numbers even

are not adequately hydrated prior to the procedure. Many drugs, such as some common antibiotics, are known to cause acute renal failure. Low blood pressure from any source – including bleeding, lack of hydration or drug effect – can lead to or worsen renal failure. Renal

REFERENCES

failure can also be a component of other

1. R. A. Bankowitz , et al., “Identifying Hospital-Wide Harm: A Set of ICD-9-CM-Coded Conditions Associated With Increased Cost, Length of Stay, and Risk of Mortality,” Am J Med Qual, 1062860613503896, first published on September 30, 2013 as doi:10.1177/1062860613503896 [Epub ahead of print].

conditions such as shock.

OUTLOOK

QTR 3.14

FEATURES

T C

ith declining

• Measures that more fully capture the

January 2014, a seventh measurement

reimbursements

domain was added to the model

essence of true engagement by patients

and other financial

(Figure 1). This domain, which focuses

and their families and caregivers;

pressures,

on community health, creates measures

• Safety (formerly known as harm)

healthcare providers are looking for

to address gaps as identified through the

measures such as AHRQ Patient

ways to go the extra mile in quality

Community Health Needs Assessment.

Safety Indicator (PSI) composites

performance to deliver safe patient

Measures added to QUEST 3.0 to

and CDC National Healthcare Safety

care, reduce costs and meet the

improve participants’ performance

Network (NHSN) definitions for

demands of value-based purchasing.

include:

healthcare acquired infections (HAIs);

In 2008, Premier launched the Quality, Efficiency, Safety and Transparency (QUEST®) initiative

• I npatient and outpatient

• Utilization measures (formerly

evidence-based care measures;

known as readmissions) that focus

•E fficiency measures that evaluate

to help members achieve top

additional opportunities for

performance in a changing healthcare

cost containment, such as

environment. More than five years

waste and resource use;

on readmissions and appropriate use of acute hospital care; • Outcomes-based mortality measurement that includes

later, the QUEST collaborative continues to prove itself as a driver of significant change for participants. The QUEST collaborative remains

Fig.1

QUEST’s seven domains of focus

a leader within the healthcare community by continuing to follow its four foundational goals of:

LEAD

• Driving rapid improvement

E R S H I P A N D C U LT U R E

in both cost and quality;

Evidencebased care

• Shaping policy and payment guidelines; • Developing and refining data collection, analytical tools, and

Cost and efficiency

Community health

collaborative methods in support of value-based purchasing and public reporting; and • Helping members reduce unjustified variation within their systems to

Mortality

Patient and family engagement

PATIENTS AND HEALTHY COMMUNITIES

achieve top performance in the seven QUEST focus areas as measured by “top performance thresholds” (TPTs). QUEST has experienced several

Appropriate hospital use

Safety

evolutions since its inception. The sixth domain of readmissions reduction was added in 2011, following the inclusion of readmission penalties as part of the ACA. When QUEST 3.0 launched in OUTLOOK

QTR 3.14

FEATURES

T F

30-day mortality rates for heart

improvement in each focus area,

of unjustified variation in mortality.

attack, pneumonia, COPD, stroke,

QUEST evaluates performance and

Implementation of best practices

sepsis, and heart failure; and

recalibrates its thresholds annually for

has created a 6 percent reduction in

mortality and readmissions and every

deaths related to strokes, heart failure

three years in the other domains.

and respiratory infections, including

â&#x20AC;˘A seventh measure domain focused on community health.

Data from the fifth year of the

pneumonia. Since focused interventions

grown from an initial 157 participants

collaborative suggest that the gap

began through the collaborative, QUEST

to 352 as of late 2013, has members in

between the top- and bottom-performing

hospitals have dramatically improved

nearly every state and is expanding

QUEST hospitals has narrowed over time

their mortality scores and are outpacing

every quarter.

as all hospitals have achieved significant

non-participants and national averages.

The QUEST collaborative, which has

In its first 5.75 years, QUEST has

improvements. This proves that QUEST

Overall, inpatient mortality rates,

experienced tremendous success in

can work in any hospital, regardless of

measured by a ratio of observed-to-

driving member hospital improvements.

size, location, and teaching or safety net

expected inpatient deaths, decreased

As of Q3 2013, the program has

status. Overall, QUEST members have

38 percent from 0.99 at the baseline to

cumulatively prevented 147,864

improved in each of the seven focus

0.61 in 2013. This reduction in mortality

inpatient deaths, reduced healthcare

areas and have demonstrated better

translates to almost 148,000 avoided

spending by $12.5 billion, and provided

performance than non-QUEST hospitals.

inpatient deaths over 5.75 years.5

116,689 additional patients with

The most dramatic advances have come

Although inpatient mortality has

quality, evidence-based care.2

in cost, patient mortality, and evidence-

decreased since 2006 for both QUEST

based care.

and facilities in a matched sample of

In addition to adding new measures, QUEST has redefined a few of its existing

non-QUEST hospitals, participating

measures as the program continues

Inpatient mortality rates

hospitals have seen accelerated

to evolve. For example, four measures

In reviewing inpatient mortality,

improvement, evidenced by the gap

within the evidence-based care (EBC)

QUEST participants found that sepsis

between QUEST and non-QUEST

domain topped out, signifying adherence

was the number one cause of death in

trends in mortality (Figure 2). In

rates nearing 100 percent. As a result,

excess of expected mortality. To tackle

fact, QUEST members experience

the latest cohort chose to use an all-or-

the problem, QUEST set up specific

mortality observed-to-expected rates

nothing scoring methodology to evaluate

sprints around sepsis, taking best

that are 14 percentage points lower

EBC performance. Through Q3 2013,

practices from the top performers.

than their matched counterparts.

QUEST participants have increased

These included techniques for early

the delivery of all recommended EBC

detection in the emergency department,

Evidence-based care

measures to 97 percent of the time,

early measurement of serum lactate

Numerous measures of EBC maxed out

a 6.6 percentage point improvement

levels, and prompt initiation of

their potential over the years, leading

since the project baseline.3

antibiotics after blood cultures and

participants to shift their focus to overall

aggressive fluid necessitation.

EBC adherence rates. QUEST members

QUEST has also raised some of its original thresholds for top performance

As a result of these interventions,

began with a baseline of providing

to match the improvements achieved by

participating hospitals have been able to

EBC 82 percent of the time. As of Q3

members. After demonstrating dramatic

reduce sepsis mortalities by 21 percent

2013, the QUEST EBC adherence rate

improvements in cost reduction,

over 5.75 years. In fact, as of 2012, sepsis

was 97 percent, equating to 116,689

the TPT was raised from the median

was no longer in the top 10 causes of

additional patients receiving every

to the top tercile of performance.

inpatient deaths for QUEST members.

recommended evidence-based care

Since the programâ&#x20AC;&#x2122;s ultimate goal is

QUEST members are now taking on

measure over 5.75 years (Figure 3).

for all members to make significant

other conditions that are leading causes

For instance, 37 percent more heart

FEATURES

T C

Fig.2

Mortality observed versus expected ratio, comparing QUEST and non-QUEST hospitals

1.2 1.1 1.0 Source: Data from Premier’s QUEST collaborative

0.9 0.8 0.7 0.6 2006 (Q3-Q4)

2007

2008

2009

Non-QUEST

2010

2011

2012 (Q1-Q3)

QUEST

Note: Q4 2012 - Q3 2013 data for this chart, which is provided by MedPar, has not yet been released. Updated figures will become available late 2014.

Fig.3

Sustained improvement in provision of EBC

116,689

120,000

93,995

100,000

80,000

66,817 Source: Data from Premier’s QUEST collaborative

60,000

42,860 40,000

24,091 20,000

9,427

0 2008

2009

2010

2011

2012

2013 (Q1-Q3)

OUTLOOK

QTR 3.14

FEATURES

T F

Fig.4

Adherence to evidence-based care

100% 97% 90% of facilities

90% 80%

82%

50% of facilities

70% 60% 50% 40% Baseline

2012 (Q4) - 2013 (Q3)

Source: Data from Premier’s QUEST collaborative

attack patients received angioplasty within 90 minutes of hospital arrival. Required reporting has increased adherence across the nation and has closed the gap to top performance. Figure 4 shows less variation among QUEST hospitals and higher adherence to strictly evidence-based care. Although performance is generally high, QUEST hospitals with the top

• Use of checklists to ensure all

Through improved discharge

evidence-based care has been

planning, medication management,

received prior to discharge;

targeting high-risk populations,

• Screening of all patients for flu

and additional case management

and pneumonia vaccinations; • Nurse-driven protocols for removing unnecessary urinary catheters; • Physician order sets that include all evidence-based care components; and • Involvement of the entire care

EBC adherence rates use strategies that

team (pharmacists, nurses,

can ensure consistent implementation

doctors, anesthesiologists) in

of the evidence. Key processes

delivering evidence-based care.

and care coordination, the median readmission observed-to-expected ratio fell from 0.97 to 0.89. Since 2010, 42 percent of QUEST members have achieved the TPT. More than 46,000 readmissions were prevented from 2011 to Q3 2013 (Figure 5).

Cost While hospitals nationwide have seen an

leveraged by top performers include:

Readmissions

increase in average cost per discharge,

to assure patients receive all

Similar to the success in harm

QUEST member increases have been

evidence-based care;

reduction, QUEST has also seen

lower than average. Although inpatient

accelerated reductions in readmissions

hospital costs per discharge have

compared to non-QUEST members.

increased by approximately 6 percent

•D aily review of medical records

•R egular feedback to clinicians on their compliance with scientific standards;

LEARNING FROM SUCCESS Harm prevention and reduction has been a major area of focus for QUEST cohort members. In 5.75 years, QUEST members’ average harm index score has decreased from 0.25 to 0.11.6 This reduction was accompanied by a significant decrease in range-of-harm scores. Cumulatively, nearly 20,000 harm events were prevented from 2010 – Q3 2013. In addition, QUEST members achieved: •8 2 percent reduction in central line-associated bloodstream infections (CLABSI); •7 1 percent reduction in hospital-acquired injuries; •6 5 percent reduction in pressure ulcers; •3 6 percent reduction in staph infections; and a •2 3 percent reduction in ventilator-associated pneumonia. These achievements were possible due to implementation of best practices in harm avoidance. Top performers in the area of harm have led the way for all QUEST members by sharing the following lessons learned: •E ffective standardization: The hospitals that made the biggest gains use strategies that “hardwire” best practices into the system. • I mplementation of care bundles: Using a collection of interventions that can be executed together results in better outcomes than interventions implemented individually. •E ncouragement of advocacy: Top performers have implemented training programs or promoted cultures that encourage communication and teamwork, while empowering staff to speak up and advocate for better, safer care. •L eadership involvement: Executive leaders of top performing systems were committed to improving safety and made this visible to frontline employees through structured leadership meetings in which staff could address their concerns. As the top performers have shown, creating a culture of safety requires a multifaceted approach to make it as easy as possible to take the right actions all of the time.

OUTLOOK

QTR 3.14

FEATURES

T F

Fig.5

Readmissions prevented over time

50,000

46,851

40,000

30,000

25,689 Source: Data from Premier’s QUEST collaborative

20,000

10,000

7,334

0 2011

per year since the collaborative began,

2012

To reduce labor expense inefficiency,

QUEST hospitals’ cost increases are 2.3

top performers cross trained employees

percent, just above the rate of inflation.

to either cut or eliminate the use of

Since the inception of QUEST,

2013 (Q1-Q3)

reliable and effective care, and improve patient safety and satisfaction. With quality improvements

contract or agency staff. Others created

necessary for survival in the new

national costs of inpatient care have

“float pools” that allow staff members

era of healthcare, QUEST enables

increased by about 37 percent, while

to be used where they are most needed.

its members to achieve maximum

costs for QUEST member hospitals have

Additionally, all top performers have

performance by considering all current

increased by approximately 14 percent.

case managers who work to reduce the

and potential measures that will be

Combined, QUEST hospitals saved $12.5

average length of stay and readmissions.

needed to ensure their future success.

billion in the last 5.75 years (Figure 6).

These efforts have eliminated excess

To put this into perspective, $12.5 billion

spending and brought efficiency and

Premier and its members are creating

would pay for every American to receive

value to QUEST hospitals.

a quality framework that can be

a free annual primary care checkup.

Through QUEST and other initiatives,

used to consistently deliver the most

Leading the charge

efficient, effective, and caring hospital

reduction analyses have identified labor

As healthcare executives, providers, and

experience to every patient, every time.

expense, excess readmissions, length

policymakers continue to seek solutions

of stay, and overuse of lab work and

that improve healthcare quality and

QUEST methodology

imaging as the largest sources of waste.

control spending, hospitals participating

Member hospitals that join QUEST work

The collaborative has worked specifically

in QUEST are already ahead of the game.

with program directors to identify areas

to reduce these measures as a result.

They have committed to collaborating

of weakness. After reviewing data from

to save lives, safely reduce costs, deliver

member institutions and collaborating

Furthermore, QUEST’s variation

FEATURES

T C

Fig.6

Cumulative cost savings

$14 billion $12.5B $12 billion $10.1B $10 billion $8 billion

$7.29B Source: Data from Premier’s QUEST collaborative

$6 billion $4.55B $4 billion $2.12B $2 billion

$683M

0 2008

with other healthcare executives,

2009

2010

2011

These activities are followed

2012

2013 (Q1-Q3)

based collaboratives. In addition,

hospital and health system leaders are

by “sprints,” a 90-day, rapid-cycle

program directors assist hospitals in

able to implement successful strategies

improvement series in which health

developing facility-specific action plans

that raise the quality and efficiency

systems work to improve specific

on a quarterly basis. These plans are

of their care. QUEST also works with

indicators. Members are given

designed to close performance gaps

member health systems through various

additional help through small issue-

in QUEST’s seven domains of focus.

engagement activities, including: • PremierConnect™, which offers a comprehensive, online best-practices forum for hospital executives, who can easily review successful techniques implemented at peer institutions; • Semiannual national meetings that provide health system executives with an opportunity to meet in person and discuss crucial issues; and • Individual health system assessments using Premier’s benchmarking and analytics tools that result in collaborative-specific, customized comparative reports.

REFERENCES 1. Premier, Inc., QUEST: High-Performing Hospitals, Year 5 Collaborative Findings, February 2014. 2. Ibid. 3. Ibid. 4. Ibid. 5. Ibid. 6. Ibid. 7. Ibid.

OUTLOOK

QTR 3.14

ÂŠ2014 AGENT ILLUSTRATEUR C/O THEISPOT.COM

How are your patients? Addressing opioid risks in the hospital setting

FEATURES

T F

Gina Pugliese, RN, MS Vice president / Premier Safety Institute

pioid analgesics provide

almost 14 percent of those who received

depression, one of the most serious

dose-dependent pain relief

opioids experienced an adverse event.

adverse effects, can lead to respiratory

with minimal toxic effects

Such incidents can include excessive

arrest and death.5,6

sedation and respiratory depression,

Opioid-associated adverse events

a variety of conditions, such as trauma-

nausea, vomiting, constipation, falls,

are also costly. A recent study, using

related injuries, post-surgical pain,

hypotension, hallucinations, delirium

a Premier database, found that post-

cancer and other diseases. They can be

and aspiration pneumonia. These

surgical patients who experienced

administered by a variety of delivery

symptoms can occur in patients receiving

an opioid-related adverse event were

systems, including oral, parenteral,

any opioid drug, and serious adverse

significantly more likely to incur greater

transdermal, epidural and spinal.

events can happen even when opioids are

expense, have a longer length of stay and

Examples of opioids include morphine,

prescribed correctly. Although the risk

experience more readmissions.7

fentanyl and others.

has been described as small, respiratory

for patients suffering from

There are strategies that can mitigate risks associated with inpatient opioid

Opioids associated with serious adverse events

use, including: •P rescreening to identify patients

A recent study showed that opioids

Opioid analgesics

were used in more than half of hospital

• Codeine

• Using non-opioid analgesia;

admissions of non-surgical patients at

• Dextropropoxyphene

•A dopting careful patient-controlled

286 U.S. hospitals.1 Two specific opioids

• Fentanyl

– fentanyl and morphine – are among

• Hydrocodone

the top 10 medications administered to pediatric inpatients.

Unfortunately, opioids are some of the most frequently associated drugs with adverse reactions. In one large study, opioids were responsible for 16 percent of drug-related adverse effects,3 while a study of post-surgical patients found that

• Hydromorphone • Meperidine • Methadone • Morphine • Oxycodone • Pentazocine

at risk;

analgesia (PCA) practices; and •C arefully monitoring patients during opioid therapy.

Prescreen to identify patients at risk Patient responses to opioids are highly variable due to biological differences, interactions with other drugs (e.g., sedatives, hypnotics), and OUTLOOK

QTR 3.14

FEATURES

T F

comorbid conditions.8 Clinical studies

side effects associated with each drug.11,12

regimens with NSAIDS, COX-2 inhibitors

and reports have identified several

There can also be a synergy when drugs

or acetaminophen, unless there is a patient

factors that put patients at higher risk

with different action mechanisms are

contraindication.15

for sedation and respiratory depression

combined.13 According to an American

due to opioid use. Clinicians should

Society of Anesthesiologists (ASA) task

carefully screen patients for these risk

force, pain management therapy should be

Carefully monitor patients on opioid therapy

factors and consider risk levels when

multimodal whenever possible to reduce

While assessing patient risk factors

creating pain management plans. They

the risks associated with opioids.

before initiating opioid therapy is

Clinicians should implement

necessary, that alone is not enough

should also question patients or their

representatives about past exposure to

multimodal pain plans for both short- and

to prevent serious adverse events.

opioids and any associated side effects.

long-term pain management, shifting

According to the Joint Commission,

from an opioid-based to an opioid-sparing

“Even at therapeutic doses, opiates can

Choose non-opioids to cut risks

approach. To reduce the use of opioids,

suppress respiration, heart rate and blood

Multimodal (balanced) therapy for pain is

using or adding non-narcotic medications

pressure, so the need for monitoring

the administration of two or more drugs

should be considered. These include

and observation is critical.”16 The Joint

(by the same or different routes) that use

NSAIDS, acetaminophen, regional

Commission, which recommends

diverse mechanisms to provide analgesia.10

infusion of local anesthetics, steroids,

performing serial assessments of

ketamine and gabapentinoids. The ASA

respiration and depth of sedation, notes

management allows lower doses of each

task force also recommends that pain

that continuous electronic monitoring

analgesic and may reduce the severity of

management plans include continuous

of oxygenation and ventilation may be

The use of multimodal pain

appropriate in some cases.17,18 A task force of the Anesthesia Patient Safety Foundation (APSF) advises healthcare providers to use continuous electronic monitoring for post-

Patient factors increasing the risk of sedation and respiratory depression with opioids9 • O bstructive sleep apnea (OSA) or high risk for OSA • M orbid obesity with high risk of sleep apnea • A dvanced age • L onger duration of general anesthesia • L ack of recent opioid exposure • P ost-surgical period, especially for procedures in the thoracic or upper abdominal areas

operative patients who receive opioids in an inpatient setting.19 This includes monitoring oxygenation (e.g., pulse oximetry) for those patients not receiving supplemental oxygen and ventilation (e.g., capnography) for patients who are on supplemental oxygen.

Adopt safe PCA practices Patient-controlled analgesia can be

• I ncreased opioid use requirement or habituation

effective in delivering pain medication

• C oncurrent use of other drugs with sedative effects, including

in an inpatient setting. However, the

antihistamines, benzodiazepines, sedatives, diphenhydramine and

safety of PCA is highly dependent on the

other CNS depressants

practices surrounding its use. A recent

• S moking • C omorbid pulmonary or cardiac disease or major organ dysfunction or failure (e.g., renal or hepatic)

national survey found significant gaps in PCA practices. For example, almost 20 percent of hospitals failed to assess patients’ previous exposure to opioids before initiating PCA.20

FEATURES

T F

Also disturbing is the fact that approximately one-third of survey respondents reported that “alarm

apnea or who recently underwent

administration by family members and

thoracic surgery;

others without proper training. They

• Incorporating non-opioid analgesics

fatigue” prevented them from

to create a balanced multimodal pain

implementing continuous electronic

management plan; and

monitoring. With the documented gaps

respiration and level of sedation and

failure of clinicians to respond to or

strongly considering continuous

silence PCA alarms represents a serious

electronic monitoring for all patients.

patient safety concern.

Meanwhile, healthcare leaders should

with PCA, the Joint Commission

of safety checklists.

• Carefully monitoring patients’

in safety practices related to PCA, the

To reduce the risks associated

should also consider ordering the use

mandate the use of safe PCA practices, including the avoidance of PCA

For more information on opioids, see Opioids and Patient Safety on the Premier Safety Institute website at www.premierinc.com/opioids.23 For additional information, see Premier Safety Institute’s review: Opioid analgesics: a double threat to patient safety.

recommends careful patient selection, ongoing monitoring of those receiving PCA, and possible use of infusion pumps with dosage error reduction software. The Joint Commission dedicated a Sentinel Event Alert to the risks associated with PCA given by family members and others who are not authorized to administer the drugs (“PCA by proxy”).21 The Alert recommended education of staff and family members about the danger of PCA by proxy and reiterated the importance of careful patient selection and monitoring. To ensure consistent safety practices, organizations should consider using a PCA safety checklist, such as the one created by the Physician Alliance for Health and Safety.22

Conclusion Opioid analgesics are widely used in the inpatient setting and can be associated with respiratory depression and other adverse events, even when prescribed correctly. Clinicians can take steps to reduce the risk of opioid-associated adverse events, including: • Prescreening all patients to identify those at higher risk for these events,

REFERENCES 1. S. J. Herzig et al., “Opioid Utilization and Opioid-Related Adverse Events in Nonsurgical Patients in US Hospitals,” J Hosp Med 9 (2014): 73–81. 2. T. Flasket al., “Estimating Pediatric Inpatient Medication Use in the United States,” Pharmacoepidemiol Drug Saf 20, no.1 (2011): 76-82. 3. E. C. Davies et al.,” Adverse Drug Reactions in Hospital In-patients: A Prospective Analysis of 3695 Patient Episodes,” PLoS One 4, no.2 (2009): e4439. 4. E. R. Kessler et al., “Cost and Quality Implications of Opioid-Based Postsurgical Pain Control Using Administrative Data From a Large Health System: Opioid-Related Adverse Events and Their Impact on Clinical and Economic Outcomes,” Pharmacotherapy 33, no. 4 (2013): 383-91. 5. C. Koo and M. Eikermann, “Respiratory Effects of Opioids in Perioperative Medicine,” Open Anesthesia Journal 5, suppl. no. 1-M6 (2011): 23–34. 6. F. J. Overdyk et al., “Continuous Oximetry/Capnometry Monitoring Reveals Frequent Desaturation and Bradypnea During Patient-Controlled Analgesia,” Anesth Analg 105, no.2 (2007): 412-8. 7. G. M. Oderda et al., “Effect of Opioid-Related Adverse Events on Outcomes in Selected Surgical Patients,” J Pain Palliat Care Pharmacother 60, no. 1 (2013): 62-70. 8. C. Koo and M. Eikermann, “Respiratory Effects of Opioids in Perioperative Medicine,” Open Anesthesia Journal 5, suppl. no. 1-M6 (2011): 23–34. 9. Joint Commission, “Safe Use of Opioids in Hospitals,” Sentinel Event Alert 49 (2012):1-5. 10. American Society of Anesthesiologists Task Force on Acute Pain Management, “Practice Guidelines for Acute Pain Management in the Perioperative Setting: An Updated Report by the American Society of Anesthesiologists Task Force on Acute Pain Management,” Anesthesiology 116, no. 2 (2012): 248-73. 11. Elia, et al., “Does Multimodal Analgesia with Acetaminophen, Nonsteroidal Antiinflammatory Drugs, or Selective Cyclooxygenase-2 Inhibitors and Patient-Controlled Analgesia Morphine Offer Advantages Over Morphine Alone? MetaAnalyses of Randomized Trials,” Anesthesiology 103, no. 6 (2005): 1296-304. 12. E. Maund et al., “Paracetamol and Selective and Non-selective Non-steroidal Anti-inflammatory Drugs for the Reduction in Morphine-related Side-effects After Major Surgery: A Systematic Review,” Br J Anaesth 106, no. 3 (2011): 292-7. 13. Fernández-Dueñas et al., Synergistic Interaction between Fentanyl and a Tramadol: Paracetamol Combination on the Inhibition of Nociception in Mice,” J Pharmacol Sci. 118, no. 2 (2012):299-302. 14. American Society of Anesthesiologists Task Force on Acute Pain Management, “Practice Guidelines for Acute Pain Management in the Perioperative Setting: An Updated Report by the American Society of Anesthesiologists Task Force on Acute Pain Management,” Anesthesiology 116, no. 2 (2012): 248-73. 15. Ibid. 16. Joint Commission, “Safe Use of Opioids in Hospitals,” Sentinel Event Alert 49 (2012):1-5. 17. Ibid. 18. Joint Commission, “Patient Controlled Analgesia by Proxy,” Sentinel Event Alert 33 (2010):1-2. 19. Anesthesia Patient Safety Foundation, No Patient Shall Be Harmed by Opioid-Induced Respiratory Depression, 2011. Available at: http://www.apsf.org/newsletters/html/2011/fall/01_opioid.htm (accessed March 25, 2014). 20. Wong M, Mabuyi A, Gonzalez B. First national survey of patient-controlled analgesia practices. Physician-Patient Alliance for Health & Safety. 2013. Available at: http://ppahs.files.wordpress.com/2013/10/first-national-survey-of-patient-controlledanalgesia-practices.pdf (accessed March 26, 2014). 21. Joint Commission, “Patient Controlled Analgesia by Proxy,” Sentinel Event Alert 33 (2010):1-2. 22. Physician-Patient Alliance for Health & Safety (PPAHS), Patient Controlled Analgesia (PCA) Safety Checklist, 2013. Available at http://ppahs.org/category/patient-monitoring (accessed March 26, 2014). 23. “Opioids and Patient Safety,” Premier Safety Institute Website. www.premierinc.com/opioids.

such as those with obstructive sleep

OUTLOOK

QTR 3.14

IMPACT OF THE ONGOING

DRUG

SHORTAGE

FEATURES

T F

he U.S. is facing an ongoing drug shortage that challenges care providers to properly

treat patients in a cost-effective

REASONS FOR THE ONGOING DRUG SHORTAGE INCLUDE:

manner. In February 2014, the Food

Drug recalls

and Drug Administration (FDA) reported 38 new drug shortages from

Quality of active pharmaceutical ingredients

January through September 30, 2013 â&#x20AC;&#x201C; a significant decrease from the 117

Industry consolidation

identified during calendar year 2012.1 However, long-standing shortages still affect more than 300 drugs.

Quality control issues at manufacturing plants

Drug shortages have been associated with delays or cancellations of necessary medical procedures; added risk of

Lack of manufacturing line capacity

healthcare errors and adverse events; and increased costs for providers through the substitution of more

Just-in-time inventories

Regulatory and financial pressures

expensive therapeutic alternatives. Premier recently surveyed its member pharmacy experts to learn more about the magnitude of current drug shortages

least one shortage in the last six months

a two-fold improvement over 2010 (15.9

and how they impact hospitals. This

that may have caused a medication safety

percent). The percentage of respondents

survey updates research conducted at

issue or error in care. That compares to

with six or more occurrences decreased

the height of the drug crisis in 2010.

91.4 percent in 2010 (Figure 1). However,

45.7 percent compared to 2010 (15.0

Published in 2011,2 it was one of the

the prevalence of shortages affecting

percent versus 27.6 percent).

earliest instances of documentation

patient care seems to be decreasing.

regarding the shortage.

Compared to 2010, the percentage

Financial impact

of respondents with fewer shortages

Survey results suggest shortages

Quality and safety impact

(between one and five) over the last six

continue to increase unnecessary costs.

Drug shortages may result in the delay

months increased by 30 percent (50.4

Over the last six months, more than 99

or cancellation of necessary medical

percent versus 35.5 percent in 2010).

percent of those surveyed said a shortage

procedures. They may also result in

Meanwhile, the percentage experiencing

led to the purchase of a more expensive

substitutions that can increase the time

six or more occurrences decreased 26

generic alternative. This was similar to

and expense required to protect against

percent (39.7 percent versus 53.5 percent

responses in 2010 (Figure 3).

medical errors and adverse events. This

in 2010).

is particularly true if prescribers are not

In addition, fewer respondents

Also, 96.4 percent reported having to purchase a more expensive therapeutic

as familiar with alternative dosing and

experienced drug shortages in the last six

alternative during the past six months

potential interactions with other drugs.

months that delayed or cancelled patient

because of a shortage â&#x20AC;&#x201C; again, equivalent

Survey results suggest drug shortages

care (Figure 2). Of those reporting, 35

to results from 2010 (Figure 4).

continue to put patients at risk, with 90

percent did not experience a shortage

percent of respondents experiencing at

that could have delayed or cancelled care,

At the same time, 93.7 percent of respondents said a shortage necessitated

FEATURES

T C

Fig.1

Number of times a drug shortage may have caused a potential medication safety issue or error in patient safety over the last six months 60%

2010 2014

50%

40%

30%

20%

10%

0% 0

1-5

6-10

11 or more

Source: Premier, Inc. drug shortage survey

Fig.2

Number of times a drug shortage resulted in a delay or cancellation of a patient care intervention in the last six months

60%

2010 2014

50%

40%

30%

20%

10%

0% 0

1-5

6-10

11 or more

Source: Premier, Inc. drug shortage survey

OUTLOOK

QTR 3.14

FEATURES

T F

Fig.3

Number of times a drug shortage required the purchase of a more expensive generic alternative in the past six months 60%

2010 2014

50%

40%

30%

20%

10%

0% 0

1-4

5-20

21-50

51-100

Source: Premier, Inc. drug shortage survey

the purchase of a more expensive

Premier’s Purchasing Partners services,

Drug shortage strategies

product directly from a manufacturer

the annualized financial impact of more

Care providers have become increasingly

over the last six months, compared to

expensive generic alternatives averaged

capable of handling supply chain

91.5 percent in 2010 (Figure 5).

approximately $56 million in 2011, $77

disruptions, implementing more

million in 2012, and $61 million in 2013.

effective programs to cope with long-

Overall, the drugs most often cited for affecting patient safety and cost were: • Electrolytes, IV fluids and parenteral

From a national standpoint, the

term shortages. According to survey

analysis suggests that all U.S. hospitals

results (Figure 7):

nutrition solutions, which provide

incurred additional costs of close to

• 89.9 percent added back-up inventory

hydration and nutrition intravenously

$199.3 million in 2011, $280.7 million

or adjusted par levels for critically

to patients unable to take oral fluids

in 2012, and $209 million in 2013

important drug categories;

or food;

(Figure 6). The total economic impact

• Cardiovascular agents used to treat heart

is likely much higher, since the financial

disease and other cardiac conditions

analysis excludes drugs purchased from

(e.g., nitroglycerin IV solution); and

off-contract distributors or those with

• Surgical agents needed for surgery

• 86.9 percent increased communications about shortages to internal stakeholders; • 82.8 percent implemented

therapeutic alternatives. The research

restrictions and/or rationing for

preparation/anesthesia and sedation

also does not include indirect costs, such

short-supply drugs;

(e.g., propofol).

as labor needed to manage drug shortages and secure alternative supplies.

Analysis of added costs

• 56.6 percent added regular meetings with internal pharmacy staff to formulate and prioritize

Based on current member hospitals

actions needed to ameliorate drug

and other non-acute care sites that use

supply problems;

FEATURES

T C

Fig.4

Number of times a drug shortage required the purchase of a more expensive therapeutic alternative in the past six months

45%

2010

40%

2014

35% 30% 25% 20% 15% 10% 5% 0% 0

1-4

5-20

21-50

51-100

Source: Premier, Inc. drug shortage survey

Fig.5

Number of times a drug shortage required the purchase of a more expensive product from a direct manufacturer source in the past six months

45%

2010

40%

2014

35% 30% 25% 20% 15% 10% 5% 0% 0

1-4

5-20

21-50

51-100

Source: Premier, Inc. drug shortage survey

OUTLOOK

QTR 3.14

FEATURES

T F

Fig.6

Additional hospital spending on generic substitutes $300,000,000

$250,000,000

$200,000,000

$150,000,000

$100,000,000

$50,000,000

$0 2011

2012

2013

Note: Purchasing data supplied by Premier, Inc. to estimate additional replacement cost of on-contract back-ordered drugs versus available generic equivalent drugs. Price differential was determined by subtracting the actual cost of the generic equivalent from the actual cost of the back-ordered drug and multiplying this by the volume purchased for the hospital sample. Projected additional replacement cost was determined by dividing the total volume cost by the number of staffed beds for the hospitals in the sample and multiplying this rate by the number of staffed beds reported for U.S. hospitals by the American Hospital Association (due to unavailability of data, staffed beds from 2012 was used for 2013). Additional replacement cost only applies to the cost differential for the hospital (lost revenue) and does not take into consideration the cost of hospital and pharmacy operations or the labor required for administration of the medication, or the cost of the drugs purchased from off-contract distributors or directly from the manufacturer. The projections should thus be considered conservative.

Unlike many shortage events, which

capture monetary compensation when

measures for drugs not normally used

can be caused by a manufacturer leaving

shortages occur and a more expensive

to avoid errors in medication dosing;

the market or closing a production

and

line – and can be managed by advanced

• 52.5 percent introduced special safety

alternate drug must be purchased; • 42.1 percent noted advocating for

notice to the FDA and providers –

legislative changes that encourage the

this particular shortage was largely

FDA to respond to drug shortages and

in managing severe supply chain “spikes”

unpredictable. That created a “shock”

provide early notice/communication

in demand that can create shortage

to the healthcare system that hospitals

about manufacturing issues;

situations of an undetermined time

were not prepared to handle. As a result,

frame. For example, after this survey was

these shortages remain an area of

additional manufacturers to contract

conducted, Premier received numerous

concern across the supply chain.

portfolios, and providing information

• 19.2 percent hired additional staff. Providers continue to face challenges

reports of shortages for intravenous

• 34.2 percent recommended adding

about therapeutic alternatives to drugs

solutions, particularly saline used to

The role of GPOs

hydrate patients suffering from a variety

Premier members indicated their group

of conditions. According to the FDA, the

purchasing organization (GPO) can help

alternative supply sources in shortage

shortage appears to have been triggered

in managing drug shortages (Figure 8).

situations; and

by quality issues related to IV pumps in

Of those surveyed:

addition to increased and unexpected

•5 2.6 percent cited failure to supply

demand from hospitals using the solutions to treat patients with the flu.3

clauses in contracts to help hospitals

in short supply; • 32.9 percent suggested locating

• 30.3 percent wanted increased sharing of information and more networking on the topic among providers.

FEATURES

T C

Fig.7

Methods providers are using to reduce the financial and quality impact of drug shortages

Added back-up inventory for critically important drug categories or changed par levels for drugs

Increased communications about shortages to internal stakeholders

Implemented restrictions and/or rationing for short supply drugs

Added regular meetings with internal pharmacy staff to formulate and prioritize actions needed to ameliorate existing and new drug supply problems

Implemented special safety measures for drugs that are not normally used to avoid errors in medication dosing

Added staffing/labor to help manage the drug shortage problem

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Source: Premier, Inc. drug shortage survey

OUTLOOK

QTR 3.14

FEATURES

T F

Fig.8

Role of GPOs in assisting providers in managing drug shortages

Offered failure to supply clauses in contracts to help hospitals recoup credits

Provided early notice and communication about manufacturing issues that may lead to a shortage

Advocated for legislative changes that encourage the FDA to respond to drug shortages

Provided information about therapeutic alternatives to drug shortages

Added additional manufacturers to the contract portfolio

Located alternative sources of supply in shortage situations

Provided knowledge sharing and networking on the topic between members

Provided information on safe sourcing and how to avoid gray market distributors

Used purchasing scale to entice new manufacturers into the market to produce shortage drugs

0% Source: Premier, Inc. drug shortage survey

10%

20%

30%

40%

50%

60%

FEATURES

T C

Conclusion

(200 to 500 beds) and large hospitals

Drug shortages can present significant

(500+ beds) and 39 percent from small

risk to public health. Based on our survey

hospitals (less than 200 beds), with 69

findings, hospitals are using a variety

percent from non-rural areas.

of strategies to address hundreds of persistent drug shortages. This requires ongoing and costly management by hospital personnel, as well as significant efforts by GPOs. In addition to the drug shortage problem, the supply chain is also vulnerable to spikes that significantly disrupt patient care and hospital operations. Today, one such spikerelated issue involves intravenous solutions. Steps have been taken by lawmakers, the FDA, manufacturers, hospitals and GPOs to reduce the incidence and manage drug shortages. While these efforts have had a positive impact, drug shortages continue to plague hospitals and require an urgent commitment from all stakeholders.

Methodology In 2010, Premier conducted an electronic survey over a six-month period (July to December 2010) of 330 pharmacy experts representing Premier alliance member hospitals (urban/rural, small/ midsized/large) to obtain information about their experiences with drug shortages. Findings were published in a March 2011 white paper.4 A similar survey was conducted over a five-week period from December 2013 to January 2014. Responses were collected from 124 pharmacy experts (margin of error +/- 5.3 percent to 8.8 percent). The majority of respondents were pharmacists. Approximately 61 percent were from midsized hospitals

REFERENCES 1. Margaret A. Hamburg, M.D., Commissioner of Food and Drugs, Department of Health and Human Services Food and Drug Administration, First Annual Report on Drug Shortages for Calendar Year 2013, report to Congress required by Section 1002 of the Food and Drug Administration Safety and Innovation Act Public Law 112-144, February 5, 2014, http://www.fda.gov/downloads/Drugs/DrugSafety/DrugShortages/UCM384892.pdf. 2. C. Cherici, et al., “Navigating Drug Shortages in American Healthcare: A Premier Healthcare Alliance Analysis,” March 2011, www.premierinc.com/about/news/11-mar/drug-shortage-white-paper-3-28-11.pdf. 3. U.S. Food and Drug Administration, Drug Safety and Availability, http://www.fda.gov/drugs/drugsafety/ ucm382255.htm. 4. C. Cherici, et al., “Navigating Drug Shortages in American Healthcare: A Premier Healthcare Alliance Analysis,” March 2011, www.premierinc.com/about/news/11-mar/drug-shortage-white-paper-3-28-11.pdf.

OUTLOOK

QTR 3.14

FEATURES

T F

FACE TIME WITH

Keith J. Figlioli I Senior vice president, healthcare informatics, Premier, Inc.

Closed APIs: There’s a fee for that Keith J. Figlioli is senior vice president of healthcare informatics, Premier, Inc., and a member of the Office of the National Coordinator Health Information Technology Standards Committee, which recommends EHR technical capabilities needed for meaningful use.

What is an application programming interface (API)? Can you provide an example of how it is successfully used in industries beyond healthcare? An API is a set of functions and procedures used by computer programs to communicate with one another. You could call it a user guide for developers to design applications that will work on a given technology platform. There are two types of APIs, open and closed. With open APIs, it’s as simple as having one system talk to another without the need for manual intervention. On the other hand, closed APIs are not openly available; you must be part of a small user group to access them. Take Twitter, for example. Millions around the world use the app as a means of communication. But how useful would Twitter be if you had to contact the company and get a custom code each time you wanted to tweet?

FEATURES

T F

Application Programming Interface (API) facts The number of APIs has more

The U.S. loses $300-$450 billion in

than tripled since 2011.

annual economic potential because

of closed data systems.2 Examples of open APIs include

The U.S. electronic health record

Twitter, YouTube and Facebook.

(EHR) market is projected to reach $9.3B by 2015.3

Despite increased spending, hospital C-suite executives are increasingly dissatisfied with their EHR systems.4

Closed APIs prevent billions of new apps from coming to

In other words, not only are providers spending millions on

the market and harm new entrants that may have good ideas

one-off apps, they’re also missing opportunities for innovation,

for innovation.

since EHRs won’t allow them to access their own data.

Is healthcare where it needs to be with API interoperability?

money. It should be the exact opposite – helping providers

It’s well known that healthcare is an industry that lags with

one system to another should be free. This is the cornerstone of

regard to API interoperability. Today, data is frequently locked

an open API.

Interoperability in healthcare should not be about making reduce costs and improve population health. Moving data from

away in proprietary systems, holding back efforts to create an electronically connected, data-driven, healthcare IT (HIT) environment. EHR systems are a major culprit. Suppose, for instance, that

How can creating a more open environment drive population health improvements? Connected data is essential for better predicting, diagnosing

providers want to link EHR and pharmacy data so they can be

and managing population health. Without free-flowing data,

alerted when patients haven’t refilled their medications. The

we can’t predict or effectively deal with a patient safety event

providers must first contact their EHR vendors. To add insult

across acute or non-acute settings.

to injury, providers have to pay a fee of up to $15,000 to have an app developed to meet their needs and specific technical

Closed APIs only set us back as we work to connect care to improve population health management.

requirements. Often unbeknownst to them, these fees are built into the cost of EHR implementation.

OUTLOOK

QTR 3.14

FEATURES

T F

The business of healthcare should not be predicated on keeping data trapped behind proprietary walls.

How do we move the industry forward?

problem. This discontent is a main reason we formed the

We’re seeing a growing number of organizations and experts

MTIC, which serves as an advisory committee to the Premier

advocating for interoperability. For instance, we were very

Board of Directors, much as our existing value and quality

pleased with the recent report to the Agency for Healthcare

improvement committees operate.

Research and Quality and ONC5 for recommending open API

The purpose of MTIC is to provide a forum for Premier

use in healthcare data systems. The Boston-based SMART

member CIOs to offer strategic insight to the alliance in

Platform has been working with the HIT community to create

identifying member technology needs and solutions. It will

an ecosystem of substitutable apps that can run on any EHR

also guide alignment and development of federal and state

system. However, more action must come from both the public

policy positions related to the adoption of open data and

and private sectors.

interoperability initiatives and policies.

In the private sector, providers and patients need to advocate

The business of healthcare should not be predicated

for open APIs and usable data if we’re to get the most out of

on keeping data trapped behind proprietary walls. I truly

healthcare applications. This increased demand for open

feel that open APIs are the missing link in HIT. They’re

access can drive market forces to prevent closed systems

essential to ensure that patient, administrative and resource

from being introduced for a single vendor’s financial gain. In

information systems can be accessed.

addition, EHR vendors can differentiate themselves with the

We’re getting close to a tipping point. This has to happen in

diversity and utility of the apps that are built to work with their

our industry. We aren’t there yet, but I think it’s only a matter

systems, creating an added value to end users.

of time.

With the public sector, we need the government to play a role by enabling an environment that promotes innovation. One way this could be achieved would be for the Office of the National Coordinator to require open APIs for health data. In an optimal environment, vendors should have to demonstrate that information can be extracted via open APIs and leveraged by third-party software developers.

Please tell us about the Premier Member Technology Improvement Committee (MTIC). I’ve been hearing from Premier members on a daily basis about their interest in working toward a solution to this

REFERENCES 1. SmartBear, “The API Revolution [Infographic],” April 10, 2014, http://blog.smartbear.com/ api-testing/the-api-revolution/. 2. James Manyika, et al., “Open Data: Unlocking Innovation and Performance with Liquid Information,” McKinsey Global Institute, October 2013, http://www.mckinsey.com/insights/ business_technology/open_data_unlocking_innovation_and_performance_with_liquid_ information. 3. Accenture, “Getting EMR Back in the Fast Lane,” February 10, 2014, www.accenture.com/ us-en/Pages/insight-getting-emr-back-fast-lane.aspx. 4. Premier, Inc.’s spring 2014 Economic Outlook survey. 5. HealthIT.gov, “A Robust Health Data Infrastructure,” April 2014, http://www.healthit.gov/ sites/default/files/ptp13-700hhs_white.pdf

PERSPECTIVES The rise of accountable care, 34

PERSPECTIVES

ccording to results from

Large hospitals (those with more

Southeast had an existing ACO, this

Premierâ&#x20AC;&#x2122;s semiannual

than 500 beds) were most likely to

region holds the greatest potential

health system executive

participate in an ACO within the

for growth, with two-thirds of its

survey, many health

foreseeable future (Figure 2). Only

respondents planning to join or create

systems are either already participating

a small portion (5 percent) of large

an ACO in the near future. Only a small

or plan to participate in an accountable

hospitals represented had no plans to

portion (14 percent) of respondents in

care organization (ACO). Approximately

move to an ACO model.

the region had no plans to move to an

one-third of C-suite respondents

According to our survey, the Midwest

ACO model (Figure 3).

reported that their organizations have

currently has the highest concentration

already moved to an ACO model

(43 percent) of facilities participating

rise overall among survey respondents,

(Figure 1), compared to only 18 percent

in an ACO. While only 19 percent of

there are a variety of reasons why

in the fall 2013 survey. In spring 2012,

respondents from facilities in the

a health system may choose not to

Although ACO participation is on the

only 5 percent of respondents were already part of an ACO. Another 10 percent represented in the survey

Fig.1

ACO participation (C-suite only)

projected their organizations will create or join an ACO by the end of this year. Additionally, in spring 2012, 30 percent of respondents reported that

We already have an ACO in place

their organizations had no plans to join an ACO. That figure fell to nearly a quarter (23 percent) of fall 2013

By the end of 2014

respondents and has since declined to approximately 20 percent, revealing that an increasing number of health systems are planning long-term to make

By the end of 2015

the switch to accountable care.

planning to be part of an ACO, the pace

Source: Premier online survey for spring 2014 Economic Outlook publication

Despite the increase in established participation in ACOs and long-term

By the end of 2016

of adoption was slower than what was originally anticipated. Two years ago, 52 percent of respondents anticipated adoption of an ACO model by the end of

After 2016

2013, but by fall 2013, only 24 percent expected to meet that deadline. The current survey reveals that nearly one-third of respondents had an ACO in

My health system will not be creating or joining an ACO in the foreseeable future

place at the beginning of 2014. 0%

10%

15%

20%

25%

OUTLOOK

30%

35%

QTR 3.14

PERSPECTIVES

T F

Fig.2

ACO participation by facility size (C-suite only) Large hospital Midsized hospital

We already have an ACO in place

Small hospital

My health system plans to join or create an ACO

My health system will not be creating or joining an ACO in the foreseeable future

10%

30%

20%

40%

50%

60%

Source: Premier online survey for spring 2014 Economic Outlook publication

Fig.3

ACO participation by region (C-suite only) Northeast/Mid-Atlantic Southeast We already have an ACO in place

Midwest West

My health system plans to join or create an ACO

My health system will not be creating or joining an ACO in the foreseeable future

10%

Source: Premier online survey for spring 2014 Economic Outlook publication

20%

30%

40%

50%

60%

70%

PERSPECTIVES

T F

participate. When asked what primary

and other areas that help keep patients

factors deterred them from moving to

healthy and reduce healthcare costs.

an ACO model, respondents from the

The top areas of resource use have

Population health management is designed to ensure that patients receive the right care in the right place at the

Southeast were twice as likely as others

remained stable from the fall 2013

right time. It also aims to reduce the

to cite patient population or location.

survey. The category of lifestyle and

use of emergency or urgent care by

Among all respondents, hospital size,

wellness coaching remained the

enhancing care that helps people stay

lack of perceived value, and patient

primary focus area for population health

healthy. To deliver the proper care at the

population or location were most often

management. Community engagement

appropriate time, providers must have a

cited for ACO abstention.

programs, a new response option in the

holistic understanding of their patients’

spring 2014 survey, was selected by 63

needs and the factors influencing their

Population health management

percent of respondents as a top area of

health. This allows providers to craft

Accountable care models – along with

resource dedication (Figure 4). Payer

care plans that truly reflect patients’

various ACA incentives such as value-

partnerships rose from 40 percent of

needs and address the most critical

based purchasing – require healthcare

respondents in the fall to 49 percent this

drivers of their health.1

providers to be involved in their

spring, while transitional or end-of-

patients’ care beyond the acute care

life care and patient-centered medical

knowledge gap between the information

setting. Unlike the traditional care that

homes saw slight decreases since the

providers have historically had about

health systems have provided, “well

fall, though less than 10 percentage

their patients with the more substantial

care” extends beyond the four walls of

points, respectively.

understanding that is aligned with population health management.

the hospital and includes home health

Fig.4

Data and analytics can bridge the

Top areas of resource dedication for population health management Fall 2013

Lifestyle and wellness coaching

Spring 2014

Community engagement programs Home health

Source: Premier online survey for spring 2014 Economic Outlook publication

Transitional and/or end-of-life care Partnering with payers Patient-centered medical homes Virtual care/telemedicine Patient risk stratification Integrated clinical, supply chain, and financial data Patient registry 0%

10%

20%

30%

40%

50%

60%

70%

Note: Community engagement programs was not a response option in the fall 2013 survey.

OUTLOOK

QTR 3.14

PERSPECTIVES

T F

Fig.5

Popular approaches to population health management using data 80% Source: Premier online survey for spring 2014 Economic Outlook publication

70% 60% 50% 40% 30% 20% 10% 0% Fall 2013

Integrating clinical and claims data

Spring 2014

Using an integrated data solution to reduce silos

Predictive analysis

Integrating supply chain and clinical data

According to our survey, 68 percent of

likely than non-rural facilities to approach

respondents are integrating clinical

population health by integrating their

and provider leaders were the top

and claims data, 51 percent are using an

supply chain and clinical data (48 percent

collaboration for all respondents; 92

integrated data solution to reduce

versus 36 percent), but non-rural

percent of those from large hospitals

silos across their various databases,

facilities are more likely to use the other

mentioned it compared to 75 percent at

43 percent are using predictive

three approaches.

small or midsized hospitals (Figure 6).

analytics, and 42 percent are integrating

As health systems continue to develop

Partnerships with physician

Size also appeared to impact whether

supply chain and clinical data to better

strategies for population health,

hospitals partnered with public payers,

manage population health (Figure 5).

partnerships with other organizations

large local employers, and external

Because no single approach to

and individuals will be necessary

providers at a local and national level.

implementing data for population health

to successfully deliver care across

However, large hospitals were less

management is below 40 percent, it is

communities. Top areas of partnership

likely (42 percent) to be engaged in

safe to say that advanced data analytics

cited by our survey respondents include:

partnerships with health and wellness-

capabilities remain a key component in

• Physician and provider leadership

focused community groups, compared to

providing better healthcare. Overall, facilities within an IDN are more likely to currently use data

within the organization (74 percent); • Health and wellness-focused community groups (54 percent);

midsized (55 percent) and small hospitals (56 percent). Respondents from small hospitals cited partnerships with local

for population health management.

• Private payers (51 percent); and

public health departments (38 percent)

Additionally, rural facilities are more

• P ublic payers (45 percent).

and health-focused community groups

PERSPECTIVES

T F

Fig.6

Top partnership opportunities for population health by facility size

Large hospital Midsized hospital

Physician and provider leadership

Small hospital

Private payers (commercial)

Public payers (Medicare, Medicaid)

Large local employers

External providers at a local level Source: Premier online survey for spring 2014 Economic Outlook publication

Health and wellness-focused community groups

Local public health or health departments

External providers at a national level

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

OUTLOOK

QTR 3.14

PERSPECTIVES

T F

Fig.7

Number of payer partnerships per health system

35%

Fall 2013 Spring 2014

30% 25% 20% 15% 10% 5% 0% 0

5-7

8 or more

Source: Premier online survey for spring 2014 Economic Outlook publication

(56 percent) as two of the top three

involving local public health agencies.

specific patient populations that help

partnership opportunities.

Over half of IDN respondents partnered

incentivize patients to stay healthy

with public payers (52 percent), private

and help health systems meet medical

apparent. Respondents from the West

payers (60 percent), and community

management goals, such as formulary

appeared significantly more engaged

groups (56 percent), while an

development.2

with public payers than those from

overwhelming majority partnered with

other regions (Northeast/Mid-Atlantic,

physician leadership (85 percent).

Geographic differences were also

Since fall 2013, the percentage of respondents who do not have at least

Southeast, and Midwest). Even so, the

For many providers hoping to

one payer partner dropped from 29 to 24

West reported the lowest engagement

better manage population health,

percent, while the percentage with only

with local, external providers.

payer partnerships are a means to

one payer partner increased from 14 to

advance health systemsâ&#x20AC;&#x2122; efforts by

20 percent (Figure 7).

A similar difference occurred between

One-third of non-acute respondents

rural and non-rural facilities, with

limiting investment risk and rewarding

rural facilities preferring community

improvement in health outcomes and

do not currently have a payer

partnerships and non-rural facilities

cost reduction. Payers can also provide

partnership in place, compared to 22

preferring partnerships with payers.

health systems access to claims data

percent of acute respondents. Acute

Facilities within an IDN were

that can help clarify what services are

respondents were more likely

more likely to organize any form of

used by whom and whether they are in-

(58 percent) to have between one and

partnership than facilities not in an

or out-of-network. In addition, payers

four payer partnerships, compared to

IDN, with the exception of partnerships

can develop value-based benefits for

non-acute respondents (43 percent),

PERSPECTIVES

T F

Fig.8

Types of payment arrangements

Fall 2013 Spring 2014 Shared savings, upside

Bundled payment

Capitation

Care management fees Source: Premier online survey for spring 2014 Economic Outlook publication

Shared savings, downside

Other

10%

20%

30%

40%

50%

60%

OUTLOOK

QTR 3.14

PERSPECTIVES

T F

Fig.9

Types of payment arrangements, IDN and non-IDN

IDN Shared savings, upside

Non-IDN

Bundled payment Source: Premier online survey for spring 2014 Economic Outlook publication

Capitation

Care management fees

Shared savings, downside

Other

10%

20%

30%

40%

50%

60%

70%

but were slightly less likely (19 percent)

shared savings upside and capitation

Northeast/Mid-Atlantic and West

to have five or more payer partnerships

programs. They were also nearly three

reported the highest participation in

compared to non-acute facilities

times as likely to be in bundled payment

upside-only shared savings programs

(20 percent).

programs. IDN and non-IDN respondents

(Figure 10). Respondents from the West

reported similar rates of participation in

also cited the highest participation in

care management fee programs.

shared savings downside programs,

The most common type of payment arrangement was upside-only shared savings, which increased from 49

Rural respondents were more likely

capitation programs and care

percent of respondents in fall 2013 to

than non-rural to participate in care

management fees, compared to those in

55 percent in spring 2014 (Figure 8).

management fees (42 percent compared

the Northeast/Mid-Atlantic, Southeast

Those engaged in capitation programs

to 27 percent) and had slightly higher

and Midwest.

increased 39 percent from fall 2013

participation in capitation programs.

to spring 2014. During the same

The rate of participation in shared

making strides to move care beyond

period, respondents engaged in care

savings programs, both upside and

the hospital walls and better connect

management fee programs with payers

downside risk, were nearly twice as high

patients with resources that can keep

decreased 27 percent.

among non-rural respondents.

them healthy. The rise of accountable,

There was also variation in

value-based care coincides with health

Respondents from an IDN reported

Itâ&#x20AC;&#x2122;s clear that health systems are

greater rates of participation in payer

participation in different types of

systems working to build the foundation

partnerships than those outside of an

payment arrangements by geographic

for population health through a variety of

IDN (Figure 9). Respondents from IDNs

region. Survey respondents from the

partnerships, initiatives and investments.

were significantly more likely to be in

PERSPECTIVES

T F

Fig.10

Types of payment arrangements by geographic region

Northeast/Mid-Atlantic Southeast Shared savings, upside

Midwest West

Capitation

Shared savings, downside

Care management fees Source: Premier online survey for spring 2014 Economic Outlook publication

Bundled payment

Other

10%

20%

30%

40%

50%

60%

70%

OUTLOOK

QTR 3.14

PERSPECTIVES

T F

Methodology From February to March 2014, Premier,

Fig.11

Inc., in collaboration with Customer

Role of survey respondents C-suite

Care Measurement and Consulting, of approximately 10,000 healthcare

Source: Premier online survey for spring 2014 Economic Outlook publication

Supply chain or materials management

LLC, commissioned an online survey

Service line or practice area manager/director

leaders across our membership, representing both the acute and non-

Office administrator/manager

acute healthcare markets (n=522; response rate=5 percent). The survey

Finance and/or accounting

collected data on members’ perspectives about the healthcare supply chain,

Physician/clinician

population health management, quality Quality improvement

incentives, and financial and economic trends impacting the industry. The

Other

survey results shown in this article include only respondents familiar with

10%

15%

20%

25%

30%

35%

accountable care and population health initiatives within their organizations or C-suite only, where noted.

Fig.12

Types of respondent organizations

The majority (80 percent) of respondents were C-suite, supply chain,

Large hospital (>500 beds) Midsized hospital (200-500 beds)

or practice area executives. Slightly

Small hospital (<200 beds)

more than half (51 percent) came from a multi-hospital system or integrated delivery network. Urban and rural areas were almost equally represented, and there were approximately equal numbers of respondents from the following geographic areas: Northeast and Mid-Atlantic, Southeast, Midwest, and West (includes Southwest, Northwest and West Coast). An overview of the respondent profile is shown in Figures 11 and 12.

Source: Premier online survey for spring 2014 Economic Outlook publication

materials management, service line

Critical access hospital (<25 beds) Ambulatory or outpatient center Multi-specialty group practice Surgery center Senior-living facility Single-specialty group practice Physician-owned specialty hospital Other 0%

REFERENCES 1. Premier’s Accountable Care Collaboratives Guidebook: Creating a People-Centered Foundation,Version 2.0 (Charlotte: Premier, Inc., February 2011), https://www.premierinc.com. 2. Amanda J. Forster, Blair G. Childs, Joseph F. Damore, Susan D. DeVore, Eugene A. Kroch, and Danielle A. Lloyd, Accountable Care Strategies: Lessons from the Premier Health Care Alliance’s Accountable Care Collaborative (New York: The Commonwealth Fund, August 2012), http://www.commonwealthfund.org/publications/fund-reports/2012/aug/accountable-care-strategies.

10%

15%

20%

25%

30%

35%

TRENDS Partnering to improve community health analytics, 46 Picture perfect: Reducing harm in radiology, 48

James Studnicki, ScD, MPH, MBA Irwin Belk endowed chair in health services research and professor of public health science at the University of North Carolina at Charlotte

Dr. James Studnicki is the Irwin Belk endowed chair in health services research and professor of public health sciences at the

The University of North Carolina at Charlotte (UNCC) and Premier recently announced a partnership designed to help care providers improve population health in communities nationwide. Can you describe it? The agreement has two major components. The first is to transfer technology that my

University of North Carolina at Charlotte. He

research team at UNCC has developed to Premier’s platform. Since Premier’s analytical

was the first director of the Master of Health

capabilities are so robust, they can handle more volume and users than we can in an

Science program at The Johns Hopkins School

academic research enterprise.

of Hygiene and Public Health. Subsequently,

The second component – and this is the most exciting to us – is for UNCC, along with

he chaired the Department of Health Policy

Premier and its members, to develop new applications from this technology, starting

and Management and was director for the

with community health status. The quality implications for this are vast and expansive,

Center for Health Outcomes Research at the

so we’re really looking forward to these new applications.

University of South Florida Health Sciences Center. He has also been a senior hospital executive and president of a technology company started in a university incubator. Dr. Studnicki’s research has focused on using large-scale databases and associated information technology to analyze outcomes at the patient, hospital and community levels.

How did this partnership come together? We recognized 20 years ago that there was this tremendous amount of event-level (e.g., mortality) data that would be extraordinarily useful if we could: • Bring it all together and organize it in a data warehouse; • Integrate it in such a way that the databases could interact meaningfully; and • Build in strong analytical capabilities. For more than two decades, our research team has been addressing the problem of taking various publicly available, event-level datasets and creating something more useful in terms of analytics.

TRENDS

T F

In the past several years, in line with the emergence of electronic health record (EHR) systems and meaningful use criteria, we noticed that various organizations began to recognize the importance of what we were doing. They also wondered how they could use the tsunami of information available from their own EHRs. We were already able to take event-level data, integrate it and analyze it when we were approached by Sean Cassidy, vice president and general manager of Premier’s Data Alliance Collaborative. He wanted to discuss how UNCC and Premier might work together. Our sense is that Premier is making real strides to improve the healthcare system and that is very important to us in a partner. We considered other potential partners across the country, but we really felt Premier was on a parallel course with us. We each had different pieces of the puzzle. Together, we’re uniquely suited to do something special. Plus, Premier has a wide reach and relationships with providers across the country that we thought could serve as test beds for some of these analytics. Our first test environment is Catholic Health Partners (Cincinnati, OH).

What are the benefits for hospitals in turning to UNCC and Premier compared to conducting this analysis and research internally? This technology will make it possible for hospitals to develop priorities from existing community health data. They’ll have access to various reports, customized to their hospital or health system, and will not have to do any of the work themselves, which will save time and internal resources. I believe that larger health systems and IDNs, which want to use the total capabilities of the data warehouse, will be able to access all of the population data, surgical procedures, health behaviors and many other datasets themselves. The data warehouse and associated analytic tools will enable those types of hospital users to be as creative as they want in terms of what information they can see and what they can do with it. There’s no end to what you can do analytically once you have

tax-exempt status. They need to demonstrate community benefit. Part of that demonstration is doing a CHNA, determining priorities, explaining how they arrived at those priorities, and making that whole process available to the public. Language in the Affordable Care Act regarding ACOs suggests that health systems should be thinking about their geographically defined population data. The data generated from a clinically defined population isn’t the same as that from a geographically defined population. This means ACOs need to think about population trends and needs. The advantage of all of this community or event-level data is that it allows us to define a population in geographic terms. CHNAs really come alive when there is individual, event-level data as well as multi-dimensional data (e.g., all procedures). This gives providers actionable information. You can identify disparities or variations while going after the issue and creating a solution. CHNAs haven’t really changed in 5060 years, so it’s an area where there has been little advanced technology applied. We think the event-level data we have will really change the face of CHNAs and their ability to improve community health. This is particularly true since many community health status approaches have been limited to a set of county level aggregate indicators.

If CHNAs haven’t changed in 50-60 years, what are we adding now? This whole idea of creating community priorities is really subtle. The way it’s been done in the past varied widely. What has been lacking is an objective, quantifiable, consistent, verifiable way to compare data. That comes from having a system like this. We used to meet with communities across the U.S., and we found that often their priorities were determined by a piece of legislation or funding availability. We think that community priorities need to be established using some kind of objective methodology. The technology we’ve developed, combined with

this technology.

Premier’s scale and membership data, will allow health systems

How does prioritization of health issues within the community impact hospitals’ strategic plans for population health and/or ACOs?

consistent, evidence-based manner.

to identify, analyze and impact community priorities in a

All nonprofit hospitals have to conduct community health needs assessments (CHNAs) every three years to justify their OUTLOOK

QTR 3.14

TRENDS

T F

PICTURE PERFECT:

REDUCING HARM IN RADIOLOGY

TRENDS

T F

hen Wilhelm

While radiology procedures have de-

than seven times as much ionizing

Roentgen produced

livered huge benefits and improvements

radiation from medical procedures as in

the first image using

in the diagnosis and treatment of medical

the early 1980s. This is according to a

X-rays in 1895, even

conditions, these services require

report on population exposure from the

he couldn’t have foreseen the impact

complicated and costly equipment –

National Council on Radiation Protection

his discovery would have on medicine.

along with highly skilled, well-trained

and Measurements (NCRP), an organi-

Throughout the next century, radiological

clinicians and ancillary staff. Improper

zation created by Congress more than 40

research revealed a tremendous amount

operation of this equipment can result

years ago.2 The increased exposure was

about how X-rays interact with human

in medical errors and danger to

due mostly to the higher use of CT imag-

tissue. That, in turn, has resulted in

patients and healthcare providers.

ing and nuclear medicine that together

spectacular advances in anatomical

Despite the many technological

contributed to 75 percent of the medical

imaging and therapeutic treatment of

advances that ensure the proper amount

disease. Radiation technologies have

of radiation is precisely targeted to a

given clinicians new tools for diagnosis,

specific area, operator negligence can pro-

radiography-related harm include:

treatment, and information sharing.

duce unwanted outcomes. For example,

• Medication errors;

Over the past decade, attention has in-

several years ago, more than 200 stroke

• Mislabeled images leading to

creasingly focused on radiography-related

patients in California received sustained

medical services, thanks to:

radiation overdoses during computed

• Rapid introduction and adoption of new

tomography (CT) brain perfusion scans.1

technologies;

An additional concern is the recogni-

• Increasing regulatory demands;

tion of the long-term cumulative effect of

• Changes in reimbursements; and

radiation exposure, which increases the

• Recent patient safety incidents involv-

probability of radiation-induced cancers

ing harm and deaths.

for patients and healthcare workers alike. In 2006, Americans were exposed to more

radiation exposure of the U.S. population.3 Other examples of reported

misdiagnoses; • Communication errors resulting in omission of serious findings; • Infection risks from breaks in aseptic injection techniques; • Contrast material-induced nephropathy; and • Administration of inappropriate radiation doses for children. OUTLOOK

QTR 3.14

TRENDS

T F

reflect the following recent pediatric

PEDIATRIC IMPACT

trends, including:

While medical imaging is a beneficial

• Greater availability of CT;

technology for patients of all ages, over-

• Improvements in CT diagnostic

exposure to radiation is most harmful

capabilities; and

for children. Pediatric patients are

• Strong desire on the part of physicians,

at a greater risk of radiation dose

patients, and their families for

effects than adults due to

diagnostic certainty.5

their bodies’ increased radiosensitivity and the

As a result of these trends and the wide

longer length of time to

variability in radiation doses, many children

manifest the radiation impact.

receive high-dose, potentially harmful

Although adolescents are at a high risk, the use of CT among this population

Increased media attention surround-

examinations. Recent research recommends

has increased at a rate far exceeding the

the implementation of reduction strategies

growth in patient volume, despite discus-

that target the highest quartile of doses.

sion among medical professionals about

This could significantly reduce the number

risks involved.4 Reasons for this may

of radiation-induced cancers.6

benefits while minimizing risks to

This criteria, once used in response to

ing reports and studies about overex-

patients, clinicians and public health in

those accusing radiology of being the

posure have raised fears among patient

general. To face this challenge, providers

primary cause for out-of-control health-

populations concerning routine radiology

must be committed to preventing the

care costs, is now just as important as a

procedures. The American College of

inappropriate use of imaging and opti-

means for controlling patient exposure

Radiology urges patients to understand

mizing radiation technologies to obtain

to radiation.

that the risk of radiation exposure from

the best image quality with the lowest

diagnostic imaging is much less than

radiation dose.

Dose management programs also allow facilities to track, report, and

the risk of declining a diagnostically

monitor dosing to eliminate unnecessary

warranted examination. In order to

Finding the perfect dose

exposure. A dedicated program should

take advantage of radiation technolo-

The American College of Radiology has

ensure that the right test is performed

gies, clinicians must reassure patients

developed evidence-based guidelines

with the appropriate dose for the specific

that the benefits of ionizing radiation,

to assist referring physicians and other

patient. To develop a dose management

properly and responsibly applied, far

providers in radiation dose manage-

program, a multidisciplinary team

outweigh the risks.

ment. These “appropriateness criteria”

should consider:

supplement a clinician’s judgment as

• Reviewing workflows and key

The challenge in doing this is ensuring a balanced approach to the safe use of

to whether a patient is a candidate for

radiology services that addresses the

the given treatment, test, or procedure.

participants; 8

TRENDS

T F

• Establishing radiation monitoring and dose data collection;

The wide use of radiation technologies

of contracts that offer a broad range

in medicine calls for a multidimensional

of radiation safety-related products

• Defining roles for risk management;

approach to minimize the health risks

and services to members. These

• Educating and training staff; and

associated with radiation exposure while

include radiation dose tracking,

• Improving patient communications.

maximizing the benefits of diagnostic

dosimetry monitoring, dose-reducing

Unfortunately, there is no exact

and therapeutic radiology. As additional

image reconstruction, and physics

science to radiation dose manage-

research is completed, further guidance

consulting services.

ment. In fact, a variety of strategies

will be available regarding appropriate

have emerged. When it implemented a

imaging, dose optimization, imaging

thoroughly assess their capability to

comprehensive radiation dose reduc-

parameters, and dose limitation. Along

ensure the highest levels of radiation

tion program, Women and Children’s

with these guidelines, a conscious effort

safety for patients and staff, and to

Hospital of Buffalo took a multifaceted

to appropriately administer and monitor

consider the new radiation safety

approach that included:

radiation exposure will ensure that both

portfolio as a viable option to enhance

• Creating a radiation dose reduction

patients and technicians feel safe using

their own safety program initiatives.

committee;

Premier encourages members to

medical imaging technologies.

• Enlisting the services of a prominent medical physicist group;

Moving forward together

• Using a national dose index registry;

Premier’s Imaging & Radiation Oncology

• Implementing a methodology to re-

Committee has addressed the issue by

view results and adjust protocols; and

creating and awarding a new group

The primary contents of this article appeared on Premier’s blog, actionforbetterhealthcare.com, by Dave Natale, director, contract management, and on the Premier Safety Institute website, www.premierinc.com/safety.

• Initiating a marketing campaign to promote radiation dose awareness. Legislation has also been enacted in three states, with more to come, which requires the tracking and reporting of patients’ radiation dose levels. Proposals related to quality in the CMS 2014 Medicare Physician Fee Schedule include new measure groups for CT dose tracking.9 Other recommendations, sample policies, and safety initiatives have been developed by numerous local, state, federal, and international organizations. Professional societies have also launched campaigns to enhance safe use of radiation services, including Image Wisely® and Image Gently®, both designed to improve understanding of radiation overexposure implications.

REFERENCES 1. Paula Gould, “Radiation Overdose in 200 Patients Leads to FDA Safety Notice,” BMJ 339 (2009), doi: http://dx.doi.org/10.1136/ bmj.b4217. 2. National Council on Radiation Protection and Measurements, Ionizing Radiation Exposure of the Population of the United States (Bethesda, MD: National Council on Radiation Protection and Measurements, 2009). 3. Ibid. 4. Joshua Broder, Lynn A. Fordham, and David M. Warshauer, “Increasing Utilization of Computed Tomography in the Pediatric Emergency Department, 2000–2006,” Emergency Radiology 14, no. 4 (2007): 227-232, doi:10.1007/s10140-007-0618-9 (accessed May 23, 2014). 5. Ibid. 6. D. L. Miglioretti, E. Johnson, A. Williams, R. T. Greenlee, S. Weinmann, L. I. Solberg, H. S. Feigelson, et al., “The Use of Computed Tomography in Pediatrics and the Associated Radiation Exposure and Estimated Cancer Risk,” JAMA Pediatrics 167, no. 8 (2013): 700-707 (accessed May 23, 2014). 7. Joel N. Shurkin, “Physicists: Fear of Diagnostic Radiation Is Overblown,” US News & World Report, (modified January 12, 2012), http://www.usnews.com/science/articles/2012/01/17/phsycists-fear-of-diagnostic-radiation-is-overblown? 8. American College of Radiology, “About the ACR Appropriateness Criteria,” (modified November, 2013), http://www.acr.org/ Quality-Safety/Appropriateness-Criteria/About-AC. 9. Centers for Medicare & Medicaid Services, 2014 Medicare Physician Fee Schedule, U. S. Department of Health and Human Services, 2013, http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeeSched/PFS-Federal-Regulation-Notices-Items/CMS-1600-FC.html (accessed May 23, 2014).

OUTLOOK

QTR 3.14

QUALITY

STRATEGIC LEADERSHIP FOR FINANCIAL AND CLINICAL HEALTHCARE EXECUTIVES • A TWELVE MONTH OUTLOOK • SUMMER 2014

QUALITY OUTLOOK

FOR FURTHER INFORMATION To learn more about this publication, please visit premierinc.com/economicoutlook, or email economicoutlook@premierinc.com.

13034 Ballantyne Corporate Place Charlotte, NC 28277

444 N Capitol Street NW, Suite 625 Washington, DC 20001-1511

T 704 357 0022

T 202 393 0860 F 202 393 0864

PREMIERINC.COM

PUR4366 07/14

T W E LV E M O N T H O U T LO O K | SUMMER 2014

Renal Failure

Out of Harm’s Way

Drug Shortages