A new study analyzes how much the difference in readmission rates between safety net and non-safety net hospitals can be explained by measurable factors, such as patient characteristics.
The implementation of Medicare’s 2012 Hospital Readmissions Reduction Program (HRRP), which standardizes payment penalties for excess hospital readmissions, has always been controversial since patient socioeconomic status may play a vital part in some of the variation in readmission rates. As a result, hospitals who treat a disproportionate number of poor, uneducated, or unemployed patients (called safety net institutions) tend to have higher rates of readmission.
The HRRP already has several policies in place to avoid potential inequalities in penalties, such as risk adjustment factors to account for the discrepancies with safety net institutions. Proposals to benefit these safety nets include adding socioeconomic factors to the current risk-adjustment method and calculating penalties by comparing hospitals within peer groups based on their share of low-income patients. But would this change mask potential disparities in quality of care? A recent study published in Health Affairs aimed to analyze this debate.
Data was compiled from readmission rates for acute myocardial infarction, heart failure, and pneumonia based on Medicare hospital claims for the years 2009 and 2012. Based on HRRP parameters, each index admission was then classified as having an eligible readmission or not. Patient characteristics were compiled from Medicare and hospital characteristics from submitted hospital cost reports. The study then compared readmissions between safety net and non-safety net hospitals.
Analysis of the data determined how much of the difference in readmission rates between the 2 types of hospitals can be explained by measurable factors like patient characteristics and how much can be explained by unmeasured factors such as hospital performance. As expected, combined readmission rates for safety net hospitals were higher than other hospitals.
However, the predicting factors for readmission between the 2 hospital types with the greatest impact were the hospitals’ teaching and profit status, as well as discharge destination. That is, patients from safety net hospitals who were discharged to other facilities were more likely to get readmitted than those who were discharged home. This was opposite in patients from other hospitals. Interestingly, socioeconomic factors were similar in both.
Much of the remaining difference in readmission rates (approximately 40%) was attributed to unmeasured or unobserved factors, such as quality of care. Finally, the study found that safety-net hospitals have experienced only slightly higher readmission penalties under the HRRP than other hospitals have, anyway.
Together, these findings suggest that greater evaluation is needed before implementing alternatives that will factor socioeconomic status into penalty calculations for excess readmissions. Adequate indicators of socioeconomic status and the impact of these on payment must be determined. The study is an important first step in determining what needs to be done to resolve this policy debate.
“There are very good reasons to vigorously pursue the research mandate to inform continuing policy deliberations,” the study authors wrote.