Medicare's Reliance on 3 Conditions to Calculate Quality, Penalties Called "Incomplete" by Study

Condition-specific readmissions measures for heart failure, pneumonia, and heart attack may not accurately or fairly reflect hospital quality, according to a study published this week in the Annals of Internal Medicine.

The study found significant differences in hospitals' performance when readmissions were assessed for non-Medicare patients and for conditions other than those currently reported, showing that when these additional factors are taken into account, half of the hospitals would be subject to a change in their financial penalty status.

The Hospital Readmissions Reduction Program (HRRP), a national mandatory penalty-for-performance program, was unveiled by CMS with the passage of the Affordable Care Act. Although the HRRP has worked to drive down readmission rates, CMS recently hit 2600 hospitals with penalties for failing to hit targets. Hospital leaders have long been concerned about the size of the penalties and the lack of adjustment for socioeconomic and clinical factors.

Calculations known as excess readmission ratios (ERRs) examine hospitals' readmissions for heart failure, heart attack, and pneumonia among Medicare beneficiaries. These ERRs are currently used to assess care quality in the form of consumer report cards and to determine the financial implications for hospitals.

In this study, researchers reviewed data from the Healthcare Cost and Utilization Project's Nationwide Readmission Database. They zeroed in on more than 2100 hospital observations in 2013 and 2014 to investigate whether 30-day readmission measures for publicly reported conditions among Medicare patients reflect hospital performance on readmissions more broadly in an all-payer national sample. They found that 29% of hospitals currently being penalized for readmissions would no longer incur a penalty if unreported conditions were used as the basis of the calculations.

The difference widened when examining non-Medicare readmission rates: 40% of penalized hospitals would not have penalties if performance was based on readmission rates for non-Medicare patients hospitalized for the same 3 conditions.

The senior author of the study said that the reliance on Medicare data led to “an incomplete picture."

"Significant attention has been given to hospitals' overall performance as determined by the public reporting of a small number of specific conditions and patient populations. It's a little bit like issuing a final grade based on a few homework assignments and not a full semester's worth of work," said Robert W. Yeh, MD, MSc, director of the Smith Center for Outcomes Research in Cardiology at Beth Israel Deaconess Medical Center (BIDMC), in a statement.

Another researcher, Neel Butala, MD, MBA, an investigator at the Smith Center at BIDMC and cardiology fellow at Massachusetts General Hospital, said another aim was to find out if hospitals that reported high readmission rates for fee-for-service Medicare patients would also have high readmission rates for patients with other payers. They also wanted to see if high readmission rates for the 3 specified conditions held for other unreported conditions.

"This tells us that similar hospital or patient characteristics may influence readmissions more than similar disease conditions and suggests that efforts to prevent readmissions may be more successful by targeting hospital-wide processes rather than condition-specific processes," said Butala.

Medicare metrics alone may not be sufficient when judging hospital quality for readmissions, said Yeh.

Futher study is needed to identifiy what is driving the difference between the Medicare and non-Medicare patient groups, as well as to come up with risk mitigation interventions, the study said.

Reference

Butala NM, Kramer DB, Shen C, et al. Applicability of publicly reported hospital readmission measures to unreported conditions and other patient populations: a cross-sectional all-payer study [published online March 27, 2018]. Ann Intern Med. doi: 10.7326/M17-1492.