Advocates for Evidence-Based Health Policy Speakers
How Patient Case-Mix Adjustment affects Pay-for-Performance Payments
The Centers for Medicare & Medicaid Services (CMS) has a pay-for-performance pilot program that links hospital process performance to financial incentives. This paper sought to determine whether patient case mix alters overall performance ratings relative to other hospitals and whether adjustments made based on patient characteristics would alter hospitals’ eligibility status for financial incentives. Based on the adherence to such performance measures, CMS rewards hospitals in the top 20% of the pay-for-performance program and reduces payments for hospitals in the bottom 20%. Data was collected on 8 performance measures concerning acute myocardial infarction: aspirin at admission and on discharge, beta-blockers at admission and on discharge, ACE inhibitors for left ventricular systolic dysfunction, smoking cessation counseling, thrombolytics within 30 minutes of arrival and primary percutaneous coronary interventions within 90 minutes of arrival. Performance was then adjusted for baseline patient characteristics, including age, race/ethnicity, sex, BMI, insurance status, past medical history, and systolic blood pressure.
Overall, 8.2% of institutions would have benefited in their ranking when patient characteristics were taken into account. The same percentage, 8.2%, of institutions would have been negatively impacted by such an adjustment. Nearly 1 out of 6 institutions changed their initial pay-for-performance ranking after accounting for patient case mix. Hospitals with the worst performance tended to care for a higher proportion of racial or ethnic minority groups who had higher incidence of comorbidities. These hospital therefore would be more likely penalized by pay-for-performance programs.
The authors pose arguments for and against the use of case-mix adjustments. The lack of incorporation of adjustments may deprive some institutions who care for the underserved. However, those institutions with a complicated case-mix population may have partial justification for poor performance and would otherwise not be incentivized to overcome disparities in the quality of care. The paper proposes stratifying hospital rankings based on various patient subgroups; it also emphasizes pay-for-improvement rather than pay-for-performance.
Commentary:
Both academic literature and media coverage has brought attention to medical errors, and as a result, quality improvement has become one focus of evidenced-based medicine. This study highlights areas for improvement within pay-for-performance programs. Rewarding institutions based on their performance without taking into account uncontrollable factors that might influence that performance is certainly not the answer. The paper makes a relevant comparison to the ill-reputed No Child Left Behind and public education programs in an effort to explain how hospital performance might be stratified. A critical comparison lies in the unrealistic expectations of both these public education programs and of poor performing hospitals; those that serve the underserved are inherently not on a level playing field from which they can be equally judged. Penalizing institutions that are underresourced is counterintuitive and threatens the progress that our health care system requires. This study provides insight on potential unexpected and undesirable effects of pay-for-performance programs.
by
Kameron Leigh Matthews, MD, Esq.
You must be logged in to post a comment.