Academic studies have found that hospitals treating a higher proportion of low-income patients are more likely than others to have higher readmission rates. The accompanying Kaiser Health News analysis is the first to examine the issue on an individual-hospital level for 3,119 facilities.
For the story, KHN looked at congestive heart failure, which is the most common cause of rehospitalization for Medicare patients. Readmission rates came from Medicare's Hospital Compare Web site, which publishes risk-adjusted rates for heart failure and two other conditions (pneumonia and heart attacks). The data cover patients who were readmitted within 30 days of discharge between July 2007 and June 2010. KHN excluded hospitals that Medicare said didn't have enough heart failure patients to fairly judge how well they did.
KHN relied on Medicare's declaration of whether a hospital's readmission rate was substantially better (lower), worse (higher) or average compared with the national readmission rate, which was 24.8 percent of patients. Medicare is cautious in its judgment: Of the hospitals in KHN's analysis, 92 percent were rated as having average readmission rates, so only true outliers are labeled as better or worse than average.
To gauge the socioeconomic status of each hospital's patient population, KHN relied on an index that the Centers for Medicare & Medicaid Services (CMS) uses to decide whether a hospital deserves "Disproportionate Share Hospital" (DSH) payments because it treats an excess of poor patients. That index, which CMS calculates and publishes for individual hospitals, is based on the prevalence of admitted patients who qualify for Medicaid, the joint federal-state health program for the poor, or Medicare's Supplemental Security Income benefit for the poor and disabled.
KHN divided the hospitals into five groups based on their number of low-income patients: Most Poor Patients (the 20 percent with the highest DSH index); Above Average Number of Poor Patients (second 20 percent); Average Number of Poor Patients (third 20 percent); Below Average Number of Poor Patients (fourth 20 percent); Fewest Poor Patients (bottom 20 percent). KHN excluded hospitals for which CMS reported no low-income patients out of concern about data irregularities or shortcomings in hospital reporting.
The analysis showed that 11.7 percent of the Most Poor Patients hospitals were ranked by Medicare as having worse readmission rates than the national average. Only 4.3 percent of the rest of the hospitals had worse-than-average readmission rates. In other words, those hospitals with the largest share of poor patients were 2.7 times as likely to have high readmission rates.
KHN devised its analysis after discussions with a number of experts: Paul Matsui, executive director at the Advisory Board; Eric Hammelman, a vice president in the analytics division of Avalere Health; Ashish Jha, a professor at the Harvard School of Public Health; and Karen Joynt, an instructor at Harvard School of Public Health. KHN discussed its approach with Patrick Conway, CMS's chief medical officer and director of the agency's Office of Clinical Standards and Quality. He said using the DSH formula was "a reasonable adjustment," one that officials have discussed using in the future in CMS's analyses.
Academic studies have noted the link between readmissions and poverty. A study that Joynt and Jha published in Circulation: Cardiovascular Quality and Outcomes found that patients discharged from hospitals in counties with low median incomes were more likely to be readmitted. They also found that hospitals with fewer financial resources -- those that were publicly owned, lacked cardiac services or had fewer nurses -- were more likely to have more readmissions.
A study commissioned by CMS last year found hospitals with a higher proportion of poor patients had "slightly higher readmission rates" for heart failure. That study, by Harlan Krumholz of the Yale-New Haven Hospital Center for Outcomes Research and Evaluation, also found that many hospitals with lots of poor patients had the same readmission rates as did hospitals with fewer such patients. Because of that, the study concluded "that hospitals with a higher share of lower income patients can perform at least as well on readmission measures."
The KHN analysis did not take race into account. But both the CMS study and a paper published by Jha and Joynt found that hospitals where many of the patients are black are more likely than others to have high readmission rates.