While readmission rates are a measure of quality, looking at all-cause readmissions alone may not reflect a hospital's overall quality performance, according to a study in Health Affairs.
Researchers compared hospitals' readmission rates in 2009 and 2011 with commonly used quality measures such as mortality rates, volume, teaching status and compliance with process standards. They found that hospitals with low readmission rates in 2009 tended to have higher readmission rates in 2011, while hospitals with high readmission rates in 2009 tended to have lower rates in 2011.
Regression to the mean — a form of statistical noise — was responsible for some of the changes in readmission rates, indicating that readmission rate variation may not reflect real changes in quality of care, according to the study. In addition, readmission rates in 2011 had a weak or negative correlation with the other quality measures.
The authors suggest that while readmission rates may indicate quality, they should not be used alone to measure hospital performance. They offer the following recommendations for improving the use of all-cause readmission rates as a quality measure:
• Account for regression to the mean when comparing readmission rates within and between hospitals over time. People need to understand the limits of all-cause readmission rates as a quality measure.
• Base hospital payments on more than all-cause readmission rates. For example, measures of preventable readmissions and quality during care transitions may be used to measure hospital performance, according to the study.
• Apply readmission policies to a broader range of providers. Since care transitions involve providers across the continuum of care, readmission policies could include care by outpatient physicians, home health agencies and nursing homes.
RARE Campaign Helps Minnesota Hospitals Avoid 4,570 Readmissions
CMS: Medicare Readmissions Declined in 2012
Researchers compared hospitals' readmission rates in 2009 and 2011 with commonly used quality measures such as mortality rates, volume, teaching status and compliance with process standards. They found that hospitals with low readmission rates in 2009 tended to have higher readmission rates in 2011, while hospitals with high readmission rates in 2009 tended to have lower rates in 2011.
Regression to the mean — a form of statistical noise — was responsible for some of the changes in readmission rates, indicating that readmission rate variation may not reflect real changes in quality of care, according to the study. In addition, readmission rates in 2011 had a weak or negative correlation with the other quality measures.
The authors suggest that while readmission rates may indicate quality, they should not be used alone to measure hospital performance. They offer the following recommendations for improving the use of all-cause readmission rates as a quality measure:
• Account for regression to the mean when comparing readmission rates within and between hospitals over time. People need to understand the limits of all-cause readmission rates as a quality measure.
• Base hospital payments on more than all-cause readmission rates. For example, measures of preventable readmissions and quality during care transitions may be used to measure hospital performance, according to the study.
• Apply readmission policies to a broader range of providers. Since care transitions involve providers across the continuum of care, readmission policies could include care by outpatient physicians, home health agencies and nursing homes.
More Articles on Hospital Readmissions:
Why Hospitals Need to Look Beyond 30-Day ReadmissionsRARE Campaign Helps Minnesota Hospitals Avoid 4,570 Readmissions
CMS: Medicare Readmissions Declined in 2012