Pay-for-Performance Not Linked to Better Patient Outcomes

Laurie Barclay, MD

March 28, 2012

March 28, 2012 — Hospitals participating in the Medicare Premier Hospital Quality Incentive Demonstration (HQID) had no improvement in 30-day mortality or other patient outcomes compared with control hospitals participating in public reporting alone, according to the results of a study using Medicare data published online March 28 in the New England Journal of Medicine.

"These results suggest that the way we have currently conceived of pay-for-performance is unlikely to have any meaningful impact on patient outcomes," lead author Ashish K. Jha, MD, MPH, associate professor of health policy and management at the Harvard School of Public Health (HSPH) in Boston, Massachusetts, said in a news release.

Thanks to the Affordable Care Act, pay-for-performance programs are dramatically increasing in the United States. In fact, this legislation mandates expansion of pay-for-performance by the Centers for Medicare and Medicaid Services to nearly all US hospitals in 2012.

To date, Medicare has spent tens of millions of dollars on pay-for-performance financial incentives to hospitals. The largest such program is Medicare's Premier HQID. However, the association of long- term patient outcomes with pay-for-performance has been poorly studied and understood.

Using Medicare data from 252 hospitals participating in Premier HQID, the investigators studied 30-day mortality rates for more than 6 million patients admitted between 2003 and 2009 for acute myocardial infarction, congestive heart failure, pneumonia, or coronary artery bypass graft surgery. The investigators compared these rates with those in 3363 control hospitals not participating in Premier HQID.

Overall, patient outcomes did not differ between hospitals in the Premier HQID program and control hospitals, with composite 30-day mortality being 12.33% for Premier HQID hospitals and 12.40% for control hospitals. Rates of decline in mortality per quarter were also similar in both types of hospitals, and mortality remained similar 6 years after implementation of the pay-for-performance program (11.82% for Premier HQID hospitals vs 11.74% for control hospitals).

Furthermore, outcomes in Premier HQID and control hospitals did not differ for patients with acute myocardial infarction and coronary bypass graft surgery, even though the pay-for-performance program attaches financial incentives directly to mortality rates for these conditions.

"Our findings suggest that both the size of the incentives and the targets matter," senior author Arnold Epstein, MD, HSPH professor and chair of the Department of Health Policy and Management, said in the news release. "In the Premier demonstration, the incentives were small and patient outcomes were not the major focus. It is not surprising, in retrospect, that this program failed to improve patient care."

Although poor-performing hospitals would theoretically benefit the most by improving quality of care, improvements in poor-performing Premier HQID were no better than those in poor-performing control hospitals. Poor-performing Premier HQID hospitals and poor-performing non-Premiere control hospitals had similar mortality rates at baseline (15.12% and 14.93%, respectively), similar rates of improvement in mortality rate throughout the course of the study (0.10% and 0.07%, respectively), and similar mortality rates at the end of the trial (13.37% and 13.21%, respectively).

Limitations of this study include that incentives were primarily focused on processes of care, the use of administrative data, and a lack of generalizability to other models of pay-for-performance programs other than the Premiere HQID model.

"We need to better align financial incentives with delivery of high quality care," Dr. Jha said in the release. "This study suggests that in order to improve patient care, we are going to have to work a lot harder to identify and implement an incentive program that works."

"Even though Congress has required that the [Centers for Medicaid and Medicare Services] adopt pay for performance for hospitals, expectations with regard to programs modeled after Premiere HQID should remain modest," Dr. Jha and the other authors conclude.

The Robert Wood Johnson Foundation funded this study. The authors have disclosed no relevant financial relationships.

N Eng J Med. Published online March 28, 2012.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.