A new study shows a statistically significant correlation between lower readmission rates and higher mortality rates for patients with heart failure and pneumonia, renewing questions about the efficacy of the Center for Medicare and Medicaid Services' Hospital Readmissions Reduction Program (HRRP).
The new study, published in JAMA, looked specifically at changes in mortality rates in the 30- and 45-day periods after discharge for patients suffering from heart failure, heart attack or pneumonia. The 30-day data showed accelerated mortality rates for those with heart failure or pneumonia, though the 45-day data did not.
The HRRP has come under previous scrutiny for potentially raising mortality rates by offering hospitals a perverse incentive to reduce admissions in ways that don’t necessarily meet the patient’s best interest, such as discharging them after emergency room treatment or placing them in a temporary observation unit rather than readmitting them. In other studies, however, the program’s success at reducing overall readmissions has been hailed as broadly effective.
While the study’s limitations don’t necessarily indicate a causal relationship between the HRRP and the higher mortality rates, it does raise some potential red flags, according to Rishi Wadhera, M.D., a cardiology fellow at Brigham and Women’s Hospital and the study’s lead author.
Wadhera notes that the increase in mortality rates comes amid troubling context. While post-discharge deaths for heart patients with heart failure had already begun to increase in the years prior to the HRRP, the trend actually accelerated after the program was established. At the same time, the increase in mortality rates for pneumonia patients followed a period of stability in the years prior to the HRRP. Those conditions also saw significant decreases in the number of readmissions after discharge.
Whether hospitals are acting in ways that don’t necessarily benefit patients remains an open question, but Wadhera sees important lessons in the lack of clarity around how the program actually functions.
“I think one of the issues is we really haven’t done a deep dive into how physicians and hospitals have responded to the HRRP,” he said. “All we’ve really observed are the decline in admission rates and we’ve assumed that that was due to improvements in quality of care, but we don’t actually know that.”
Given the potential for unintended harm, Wadhera believes policymakers should slow down and take a more measured approach. “Our takeaway from the findings is that the HRRP should not be expanded to all hospitalized conditions, which is what some policymakers have advocated for recently, and that we should be more cautious before widespread expansion of the program.”
The findings also offer useful lessons for developing nationwide incentive-based quality programs in the future, Wadhera said.
At the very least, he recommends more limited pilot programs or randomized studies before rolling programs out nationwide. In cases where that’s not possible, he recommends specifying analyses to identify potential unintended consequences before implementation, and monitoring after implementation with well-defined stop-loss thresholds if that analysis uncovers a signal of harm.
“The HRRP teaches us about the importance of evaluating policies rigorously before we implement them nationwide, and I think CMS and specifically the Centers for Medicare and Medicaid Innovation, which have some leeway to test policies, could really take the lead on evaluating policies before they’re implemented nationwide,” Wadhera said.