Research Activities, January 2014
Readmission rates vary moderately from year to year, suggesting need for other measures of hospital care quality
Patient Safety and Quality of Care
Hospitals' annual readmission rates are not very stable, according to a new study. The importance for hospitals of their annual readmission rates (for specific conditions) began when the Centers for Medicare & Medicaid Services (CMS) began reporting these rates to consumers through the Hospital Compare Web site. Subsequently, a provision in the Affordable Care Act makes $500 million available to hospitals and community-based organizations to implement collaborative programs for reducing readmission rates. Finally, CMS is reducing payments to hospitals with higher-than-expected readmission rates.
Because of the growing importance of readmission rates to hospitals and patients, Andrew P. Ryan, Ph.D., of Weill Cornell Medical College in New York City, and his colleagues compared the change in readmission rate rankings (by quartile) for heart attack, heart failure, and pneumonia between 2009 and 2011.
The researchers found that there was little difference between the lowest rate for heart attack readmissions among the fourth quartile ("worst") hospitals and the highest rate in the first quartile ("best") hospitals. In addition, only 49 percent of the 545 hospitals in the 2009 fourth quartile remained in that quartile in 2011, while 51 percent had moved into the first through third quartiles. Similarly, among 564 hospitals in the first quartile for heart attack readmissions in 2009, 58 percent remained in that quartile in 2011, while 42 percent had moved to lower quartiles. The 2009–2011 changes in readmission rates for heart failure and pneumonia were similar.
For all three conditions, many hospitals with higher-than-average readmission rates in 2009 showed somewhat improved rates in 2011 and many hospitals with lower-than-average readmission rates in 2009 showed somewhat worsened rates in 2011. Teaching hospitals tended to have higher readmission rates than nonteaching hospitals. Also, the condition-specific readmission rates in 2011 correlated only weakly, or even negatively, with other recognized quality indicators for each condition, such as mortality rates.
Based on their findings, the researchers suggest that policymakers consider augmenting the use of readmission rates with other measures of hospital performance during care transitions. The study was funded in part by AHRQ (HS13903).
More details are in "Limits of readmission rates in measuring hospital quality suggest the need for added metrics," by Matthew J. Press, M.D., M.Sc., Dennis P. Scanlon, Ph.D., M.A., Dr. Ryan, and others in the June 2013 Health Affairs 32(6), pp. 1083-1091.