Socioeconomic Research: New Evidence on the Soundness of CMS Readmission Penalties

blog & resources


For all its sophistication, the US’s national readmission rate penalty system misses a crucial factor. For years this has been the call of thoughtful commentators such as Joynt and Jha and Sahni, Cutler, and Kocher. Now four Missouri researchers have shown in stark terms how the Centers for Medicare and Medicaid Services (CMS)’s judgments of hospital over- or under-performance would largely be abandoned after consideration of the socioeconomic status of each patient population.

In their Health Affairs article Adding socioeconomic data to hospital readmissions calculations may produce more useful results Nagasako, Reidhead, Waterman, and Dunagan have shown that even fairly crude measures of socioeconomic status (SES) make an immense difference in each hospital’s risk-standardized readmission rate (below, “RSRR”). The authors have taken into account census-tract-based indicators of poverty, educational level, and housing vacancy. With these SES adjustments in place —

  • Extremely “over-performing” hospitals now look just slightly better than average;
  • Extremely “under-performing” hospitals now look just slightly worse than average; and
  • The average hospitals still look, basically, average.


The chart above may have been relegated to Appendix Exhibit 10, but it shouldn’t be missed. It shows how, for heart attack patients, the range of hospital readmission rates, from the highest to the lowest, shrinks from 6.5 percentage points to 1.8 once SES is factored in. This is a 72% drop.

The effect is similar, if a little less pronounced, for heart failure (14.0 to 7.4; a 47% drop) and pneumonia (7.4 to 3.7; a 50% drop).

Why is this decreased range significant? It’s one thing to penalize a hospital when its rate, as currently counted by CMS (“Baseline” in the chart), is 4 or 5 points above the state average. It’s quite another when the difference is a single point (“SES-Enriched”). Controlling for the SES of the patient population, all Missouri hospitals are performing at a very, very similar level. It’s hard to see justification for any penalties at all—certainly not penalties that amount to hundreds of thousands or even millions of dollars per year.


A Parallel

Compare the authors’ findings to those from the literature on the No Child Left Behind policy enacted for the nation’s schools, and more generally on efforts to reward or penalize schools and teachers based on their students’ test scores. In their classic 2002 article Volatility in school test scores: Implications for test-based accountability systems, Tom Kane and Doug Staiger show how the kind of problem that affects RAR penalties

“can wreak havoc in school accountability systems. To the extent that test scores bring rewards or sanctions, school personnel are subjected to substantial risk of being punished or rewarded for results beyond their control.”

To this they add an astute point, one that again seems to have a direct parallel for hospitals and post-discharge care:

“Moreover, to the extent such rankings are used to identify best practice in education, virtually every educational philosophy is likely to be endorsed eventually, simply adding to the confusion over the merits of different strategies of school reform.”


Back to Blog & Resources