Should readmissions penalties take patient socioeconomic status into account?

With the implementation of the Affordable Care Act (ACA), our health care system has been charged with reforming the delivery of care via implementation of programs including but not limited to the hospital readmission reduction program (HRRP). Essentially, HRRP requires that the Centers for Medicare and Medicaid Services (CMS) institute financial penalties for hospitals with excess readmissions via readmission payment adjustment factors. This program has received significant attention of late and is the source of a public debate stemming from concerns that the hospitals that provide care for a large vulnerable population with low socioeconomic status are disproportionately and unfairly penalized.

On the one hand, proponents state that there should be one standard of quality to which every hospital is accountable, regardless of patient or payer mix. On the other hand, opponents point to data that supports a link between social factors and readmission risk and advocate for the use of risk-adjustment for socioeconomic factors when assessing readmission penalties.

In a recently published Health Affairs article, Nagasako et al.¹ explored the impact of including social factors on the calculated hospital readmission rates. The investigators analyzed data from 71,793 index admissions and 59,554 unique patients at Missouri hospitals, specifically that of Missouri Medicare fee-for-service patients discharged with diagnoses of acute myocardial infarction, heart failure, or pneumonia.

The investigators reproduced the risk-standardized model currently used by CMS to gauge hospitals’ performance on 30-day readmissions and compared it to a model that incorporated census-tract poverty rate, educational attainment, median income, housing vacancy rate, and unemployment rate. The main results are displayed in Exhibit 1 of their manuscript.

Overall, the average risk-standardized readmission rates calculated via the baseline model did not differ significantly from those of the socioeconomic-factor-enriched model. However, there was a statistically significant decrease in the variance of the risk-standardized readmission rates for all principal diagnoses when including census-tract level socioeconomic factors.

Importantly, this analysis did not include data on social factors at the individual level, making investigators unable to distinguish effects resulting from neighborhood-level factors, individual-level factors, and other variables inherent to specific heath care provider-patient interactions. However, despite these limitations, the study raises a salient question of whether the current risk-standardized readmission rate calculus accurately reflects factors that are pertinent to the assessment hospital quality and performance.

The authors do not suggest that social factors be included in models that are used to assess hospital quality, performance, and penalties nor do they suggest that the inclusion of social factors in risk-adjustment be used to justify or explain disparities in health outcomes among hospitals that care for large disadvantaged populations and are disproportionately penalized. Instead, they assert that this data should be considered in the context of targeting interventions for improving care at the hospital level and evaluating limitations of current assessment methods.

Regardless of whether you are indifferent or subscribe to the school of thought that hospitals whose catchment area is disproportionately represented by patients of disadvantaged backgrounds and therefore should be not judged against the same outcomes and performance measures as other hospitals, or instead feel that all hospitals should be held accountable to one standard of care, there is no clear right answer. Yet, the stakes are high, particularly for safety net and academic hospitals that absorb the brunt of HRRP penalties.

Nagasako EM, et al. Health Affairs. 2014; 33: 786-91.

by

Darrell M. Gray, II, MD, MPH

Dr. Gray is a board-certified gastroenterologist and Mongan Commonwealth Fund Fellow in Minority Health Policy at Harvard Medical School and School of Public Health. He was recently appointed as an Assistant Professor of Medicine and Director of Community Engagement and Equity in Digestive Health in the Division of Gastroenterology, Hepatology, and Nutrition at The Ohio State University Wexner Medical Center, a position that he will start in August 2014.