New Health Affairs Study: Medicare’s Readmissions Reduction Program Less Effective Than Widely Believed

For all of the work and effort from all sides that has been involved in the Hospital Readmissions Reduction Program (HRRP) since its establishment as one element in the Affordable Care Act (ACA), which was passed into law in 2010, a new research study just published in Health Affairs may prove to be discouraging for all involved.

Researchers Christopher Ody, Lusy Msall, Leemore S. Dafny, David C. Grabowski, and David M. Cutler, writing in the just-published January issue of Health Affairs, in an article entitled “Decreases In Readmissions Credited to Medicare’s Program To Reduce Hospital Readmissions Have been Overstated,” have written that “Medicare’s Hospital Readmissions Reduction Program (HRRP) has been credited with lowering risk-adjusted readmission rates for targeted conditions at general acute care hospitals. However, these reductions appear to be illusory or overstated,” they wrote. “This is because a concurrent change in electronic transaction standards allowed hospitals to document a larger number of diagnoses per claim, which had the effect of reducing risk-adjusted patient readmission rates. Prior studies of the HRRP relied upon control groups’ having lower baseline readmission rates, which could falsely create the appearance that readmission rates are changing more in the treatment than in the control group. Accounting for the revised standards reduced the decline in risk-adjusted readmission rates for targeted conditions by 48 percent. After further adjusting for differences in pre-HRRP readmission rates across samples, we found that declines for targeted conditions at general acute care hospitals were statistically indistinguishable from declines in two control samples. Either the HRRP had no effect on readmissions, or it led to a systemwide reduction in readmissions that was roughly half as large as prior estimates have suggested.”

Here's what the researchers found. On the one hand, they wrote, “A number of studies have documented that thirty-day risk-adjusted readmission rates declined after the HRRP was established.” On the other hand, they noted, speaking of their own research, “This study presents new evidence on why risk-adjusted readmission rates have decreased since the HRRP was established and why reductions were larger for patients with targeted conditions treated at general acute care hospitals than for other patients.”

Very importantly, they noted, “The majority of the decrease was generated by increased patient risk scores, rather than by actual lower readmission rates. However, the study did not determine why patient risk scores increased, and the authors noted that the changes could have resulted from either increased patient risk or increased coding of diagnoses. The HRRP bases patient risk scores on age, sex, and comorbidities calculated using patient diagnoses from inpatient and outpatient claims for the twelve months before hospitalization for the targeted condition. Crucially, the HRRP risk scores exclude many diagnoses coded only during the targeted admission. As a result, it is unclear how much hospitals could have manipulated patient risk scores. To ‘game’ the program’s risk adjustment, hospitals would need to code patient diagnoses more aggressively for care received before the program’s targeted admission.”

As a result, the researchers wrote, “We argue that the increased coding of patient risk scores has a more mundane explanation: Between the March 2010 establishment of the HRRP and the October 2012 introduction of penalties, the Centers for Medicare and Medicaid Services (CMS) changed the electronic transaction standards that hospitals use to submit Medicare claims, allowing for an increased number of diagnosis codes. This change coincided with the time window in which risk-adjusted readmission rates declined the fastest.” Walking readers through their nuanced methodology, they wrote that their own study and other studies have shown evidence that the readmissions reduction program has actually achieved far weaker results than have popularly been perceived.

As the authors put it, “The Hospital Readmissions Reduction Program has been cited as one of the successes of value-based payment, which fosters the view that targeted financial incentives can lead to large changes in behavior. However, altering two seemingly small details related to data and methodology meaningfully weakens the evidence that the HRRP lowered risk-adjusted readmission rates for targeted conditions and targeted hospitals. By coincidence, the HRRP was implemented just before a change in electronic transaction standards that increased diagnostic coding and therefore created the illusion that risk-adjusted readmission rates had decreased. Furthermore, given the higher rate of readmissions for targeted conditions at targeted hospitals than at nontargeted hospitals and nontargeted conditions, the decreases in readmission rates for targeted conditions and targeted hospitals were not atypically large.” And they wrote, “We note in closing that if the HRRP has not lowered readmission rates, then the rationale for the program’s existence becomes substantially weaker.”