[Skip to Content]
[Skip to Content Landing]
Invited Commentary
April 2016

Reducing Readmissions—Destination or Journey?

Author Affiliations
  • 1Office of Research and Development, Department of Veterans Affairs, Health Services Research and Development, Washington, DC
  • 2Department of Medicine, Oregon Health & Science University, Portland
  • 3Veterans Affairs Portland Healthcare System, Portland, Oregon
JAMA Intern Med. 2016;176(4):493-495. doi:10.1001/jamainternmed.2015.8603

Hospital readmissions often look like a golden opportunity to achieve the elusive triple aim of improved patient experience, better population health, and lower costs. Programs to better coordinate transitions from hospital to home could produce positive returns if they could reduce a small fraction of those repeat hospitalizations. Under the Affordable Care Act, Medicare is now reducing reimbursements to hospitals that exceed national averages for all-cause readmission rates for 6 conditions: acute myocardial infarction, heart failure, pneumonia, joint replacement, chronic lung disease, and cardiac bypass surgery. Not surprisingly, every hospital in the United States has been focusing intently on how best to reduce preventable readmissions, and 30-day rehospitalization rates fell from 58.2 to 50.1 per 1000 Medicare beneficiaries between 2009 and 2013.1

Often lost in this discussion is a more careful examination of which readmissions are truly preventable. Too often a readmission within 30 days is taken as presumptive evidence that some aspect of care was inadequate, that the patient was sent home too early, they didn’t understand the care plan or have the help they needed, or the follow-up was inadequate. In reality, readmission is often just a sign of the precarious health of most patients who end up in the hospital. Even with optimal care, an elderly patient with multiple chronic diseases may need more frequent inpatient care as their underlying conditions wax and wane. Readmissions can also be an adverse effect of our treatment success. Within the Veterans Affairs health care network, mortality from chronic heart failure has fallen significantly with improved treatment, but readmissions have remained steady in part because we are now able to keep sicker patients with very low ejection fractions alive.2

The HomeRUN study3 represents an important contribution to our understanding of preventability. Auerbach and colleagues reviewed over 1000 readmissions of patients discharged from general medicine services at 12 academic medical centers. In 27% of cases, trained reviewers judged that there was a greater than 50% chance that the readmission was preventable. A unique strength of the study was that in addition to reviewing hospital records, Auerbach and colleagues conducted interviews with patients and surveyed the various physicians involved in their care (primary care physician, discharging physician, and readmitting physician). An additional strength is that all readmissions were examined, rather than targeting specific high-risk populations such as patients who were frail and elderly. The methodology nonetheless has limitations. Because all the data were obtained and reviewed after readmission, there is possible attribution bias, wherein patients or clinicians blame a readmission on an event that might actually not have been causal. As a result, 27% likely represents an upper bound for the proportion of readmissions that are truly preventable.

How many readmissions we can really prevent is a source of much debate. To date, intensive transitional care interventions, focused mostly on high-risk populations, have had modest effects—about a 3% to 4.5% absolute reduction in 30-day readmission or 18% to 25% relative reduction.4 This likely indicates the large variety of factors contributing to preventable readmissions. There are many pathways to readmission and many targets for improvement. In fact, multimodal interventions that address several gaps in care tend to be more successful in reducing readmissions than interventions focused on single issues such as early follow-up or medication safety.4

Some of the elements of preventing readmissions (ie, patient education, self-management support, care coordination, post-discharge follow-up care, and medication safety) have been well recognized for many years. Yet, despite their familiarity, the HomeRUN study3 suggests that related transitional care gaps persist even at the leading academic hospitals in their study.

The HomeRUN study3 also identified 3 factors that contributed to a substantial proportion of potentially preventable readmissions but that have not received as much attention. These, thus, represent opportunities for future research. First, we need to test how to improve advanced care planning around the time of care transition, given the burden of hospitalizations toward the end of life.5 There are many opportunities to test innovative approaches that would facilitate cross-site palliative care consultation, greater input of a patient’s outpatient health care provider during hospitalization, and better communication of advanced care plans to the in-patient team. Second, the HomeRUN study suggests that discharging patients too early contributed to readmissions. We need to better understand what factors contribute to discharge decision making and identify opportunities for improvement. For example, many providers may not be well equipped to assess readiness for discharge, given that postgraduate medical education rarely includes transitional care curricula and training on assessing discharge readiness. The finding that inpatient assessment of physical needs was incomplete also suggests the value of incorporating assessments from a multidisciplinary care team (ie, nurses, physical therapists, and pharmacists).

Finally, the single biggest contributor to preventable admissions in HomeRUN3 was what study reviewers considered an inappropriate decision by the emergency department physician to admit the patient. This is an admittedly problematic judgment for reviewers. Drawing on data collected long after the readmission, including the clinical course of the patient during the hospital stay, makes it easy to second-guess the admitting physician—a problem known as hindsight bias. Nevertheless, this finding underscores that the benefits of better care transition support do not end at an arbitrary point after discharge. Indeed, the single factor that accounts for more variability in readmission rates than any other is the community admission rate.6 The health systems gaps that contribute to preventable readmissions probably contribute to preventable admissions as well. These likely include lack of available alternatives to hospitalization, poor local health care access, inadequate communication between emergency departments, primary care, and hospitals, and variation in admission threshold. An exclusive focus on readmission obscures important opportunities to address these deeper sources of health care fragmentation and associated gaps in care as patients transition across different parts of the health care system. Some policy experts have argued that attention to community-based hospitalization rates, along with current hospital-based readmission rates, would help spur improvements in community supports and systems integration.7

Over the last decade, reducing readmissions has been viewed as a goal in its own right, and our concerted efforts have yielded modest progress toward that goal. On the other hand, debate has grown over the value of readmissions as a measure of hospital quality because of the low rates of preventable readmissions, mixed success of interventions, problems with measurement, possible adverse effects on safety net hospitals, and questionable relationships between readmission rates and other measures of quality.8 But the HomeRUN study3 and favorable trends in readmission rates remind us that the real value in paying attention to readmissions is that it forces us to explore the interstitial spaces of our health care system—those areas where supporting structures need to be strengthened. Real improvement will not come from adding 1 new check box to the discharge form. Improvement requires creative ways to enable the complex array of professionals and caregivers involved in the care of very sick patients to work as a team across boundaries created by professional roles, geography, and time. The goal of reducing unnecessary hospitalization continues to beckon on the horizon, but we should not lose sight of the journey itself, which is just beginning and represents the real prize.

Back to top
Article Information

Corresponding Author: David Atkins, MD, MPH, Director, Health Services Research and Development, Office of Research and Development, Department of Veterans Affairs (10P9H), 810 Vermont Ave NW, Washington, DC, 20420 (David.atkins@va.gov).

Published Online: March 7, 2016. doi:10.1001/jamainternmed.2015.8603.

Conflict of Interest Disclosures: None reported.

Disclaimer: The views and opinions expressed are solely those of the authors and do not represent official policy of the Department of Veterans Affairs or any federal agency.

Daughtridge  GW, Archibald  T, Conway  PH.  Quality improvement of care transitions and the trend of composite hospital care.  JAMA. 2014;311(10):1013-1014.PubMedGoogle ScholarCrossref
Heidenreich  PA, Sahay  A, Kapoor  JR, Pham  MX, Massie  B.  Divergent trends in survival and readmission following a hospitalization for heart failure in the Veterans Affairs health care system 2002 to 2006.  J Am Coll Cardiol. 2010;56(5):362-368.PubMedGoogle ScholarCrossref
Auerbach  AD, Kripalani  S, Vasilevskis  EE,  et al.  Preventability and causes of readmissions in a national cohort of general medicine patients  [published online March 7, 2016].  JAMA Intern Med. doi:10.1001/jamainternmed.2015.7863.Google Scholar
Kansagara  D, Chiovaro  JC, Kagen  D,  et al.  So many options, where do we start? An overview of the care transitions literature.  [published online November 9, 2015].  J Hosp Med. 2015. doi:10.1002/jhm.2502.PubMedGoogle Scholar
Teno  JM, Gozalo  PL, Bynum  JPW,  et al.  Change in end-of-life care for Medicare beneficiaries: site of death, place of care, and health care transitions in 2000, 2005, and 2009.  JAMA. 2013;309(5):470-477.PubMedGoogle ScholarCrossref
Epstein  AM, Jha  AK, Orav  EJ.  The relationship between hospital admission rates and rehospitalizations.  N Engl J Med. 2011;365(24):2287-2295.PubMedGoogle ScholarCrossref
Jencks  SF, Brock  JE.  Hospital accountability and population health: lessons from measuring readmission rates.  Ann Intern Med. 2013;159(9):629-630.PubMedGoogle ScholarCrossref
Jha  AK.  Seeking rational approaches to fixing hospital readmissions.  JAMA. 2015;314(16):1681-1682.PubMedGoogle ScholarCrossref