DURHAM, N.C. -- Rewarding hospitals by paying them for performance may be ill-spent money, researchers here found, citing no significant difference in acute MI outcomes between hospitals in a Medicare incentive program and those that were not.
DURHAM, N.C., June 6 -- Rewarding hospitals by paying them for improved performance may be ill-spent money, researchers here found.
Compared with 446 hospitals given no financial incentive, 54 in a Medicare pay-for-performance pilot program did not significantly improve treatment or outcomes for patients with acute myocardial infarction, according to a report in the June 6 issue of the Journal of the American Medical Association.
Using all available data, this study found that the absolute improvement in quality attributable to pay-per-performance was only 1.6% over almost three years, found Eric D. Peterson, M.D., of Duke University here, and colleagues.
The results of this study raise concerns about the size of effect pay-for-performance should have to justify the cost, administrative burden, and potential unintended consequences of financial incentives, Dr. Peterson said.
In 2003, the Centers for Medicare and Medicaid Services (CMS) launched a large pay-for-performance pilot project in the U.S., including indicators for acute myocardial infarction. Participating hospitals with the highest performance measures would receive a financial bonus, whereas those with the poorest performance risked a future financial penalty.
To determine whether hospitals in the program improved in certain measures and outcome for treating MI patients beyond that in hospitals not participating in the program, the researchers undertook an observational, patient-level analysis of 105,383 patients with acute non-ST-segment elevation events.
The patients were enrolled in the Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines (CRUSADE), Duke's national quality-improvement initiative.
The patients were treated over three years from July 1, 2003 to June 30, 2006 at 54 hospitals in the CMS program and 446 control facilities.
The six targeted therapies included aspirin at arrival and at discharge, beta-blocker at arrival and at discharge, ACE inhibitor or angiotensin receptor blocker for left-ventricular systolic dysfunction at discharge, and smoking cessation counseling for active or recent smokers. A final composite score was included.
Among treatments subject to financial incentives, there was a slightly higher rate of improvement for two of the six targeted therapies--aspirin prescription at discharge and smoking cessation counseling--at pay-for-performance versus control hospitals, the researchers reported.
Comparing adherence scores at half-year intervals for aspirin at discharge, the odds ratio for pay-for-performance was 1.31(95% confidence interval 1.18-1.46) versus OR, 1.17 (CI, 1.12-1.21) at the control hospitals (P=.04).
For smoking cessation counseling, the OR was 1.50 (CI, 1.29-1.73) versus OR, 1.28 (CI, 1.22-1.35; P=.05).
However, the researchers reported, there was no significant difference in a composite measure of the six CMS-rewarded therapies between the two hospital groups (change in odds per half-year period of receiving CMS therapies: OR, 1.23 (CI, 1.15-1.30) versus OR, 1.17 (CI, 1.14-1.20; P=.16).
Furthermore, the researchers said, although rates of in-hospital mortality improved slightly over time, there was no evidence that improvements in hospital mortality were incrementally greater at pay-for-performance sites (change in odds of in-hospital death per half-year period, 0.91; CI, 0.84-0.99 versus 0.97; CI, 0.94-0.99; P=.21).
In addition, for composite measures of acute myocardial infarction treatments not subject to incentives, the rates of improvement were not significantly different (OR, 1.09; CI, 1.05- 1.14 versus OR, 1.08; CI, 1.06-1.09; P=.49), the researchers reported.
The pay-for-performance had a limited incremental impact on overall trends in processes of care for acute myocardial infarction, the researchers said. Using all available patient-level data, this study found that the absolute improvement in quality attributable to pay-per-performance of only 1.6% over almost three years was smaller than the 4.3% reported in another pay-incentives study.
Given this, and the challenge of conducting large-scale randomized controlled trials, the researchers said, different hospital groups in observational studies will be important for developing a robust evidence base.
Study limitations, the researchers wrote, included the fact that the study was limited to patients with non-ST-segment elevation myocardial infarction and therefore did not include two CMS measures: time to thrombolytic administration and percutaneous coronary intervention for ST-segment elevation MI.
Also, the researchers said, the financial-incentive performance was measured over only three years suggesting that a larger effect could be realized over a longer period.
In conclusion, the researchers wrote, this study is one of the first to evaluate the CMS financial-incentive pilot program. The results suggest only a small incremental benefit of pay-for-performance in some processes of acute myocardial infarction care without an associated improvement in patient outcomes, even without explicit consideration of the costs of such a program.
Additional studies of pay-for-performance are needed, they said, to determine its optimal role in quality-improvement initiatives.
Partnership, Schering Corp, and Berlex. John S. Rumsfeld. M.D., Ph.D., reported that he serves on the scientific advisory board for United Healthcare and is chief science officer for the American College of Cardiology National Cardiovascular Data Registries.
CRUSADE is funded by the Schering-Plough Corp. Bristol-Myers Squibb/Sanofi Pharmaceuticals Partnership provided additional funding support as did Millennium Pharmaceuticals. Dr. Peterson reported that he is the recipient of a grant from the National Institute on Aging.