Studies examine effectiveness of financial incentives in primary care

Two randomized trials suggest that certain financial incentives may be associated with some clinical improvements in primary care, although both the study authors and editorialists said that additional research is needed.

The first study, published in the Sept. 11 Journal of the American Medical Association, was a cluster-randomized trial of 12 Veterans Affairs outpatient clinics, involved 83 primary care physicians and 42 nonphysicians and looked at the effect of individual-level financial incentives, practice-level financial incentives, both kinds of incentives, or no incentives on hypertension care. Five performance periods and a 12-month washout period were included.

Total payments over the course of the study were $4,270 for combined incentives, $2,672 for the individual incentives, and $1,648 for the practice-level incentives. Only the individual-incentive group saw a significantly greater increase in blood pressure control or in appropriate response to uncontrolled blood pressure compared with the control group. No incentive, however, affected use of guideline-recommended medications or increased hypotension incidence, and the effect of incentives was not sustained after the washout period.

The second study, published in the same issue of JAMA, was a cluster-randomized trial of small primary care clinics, all of which had electronic health records. Clinics in the incentive group were paid for each patient whose care met performance criteria for aspirin or thrombotic prescriptions, blood pressure control, cholesterol control, and smoking cessation. Payments were higher for patients who had comorbid conditions, those who had Medicaid, and those who were uninsured. Maximum payments during the study, which ran from April 2009 through March 2010, were $200 per patient and $100,000 per clinic.

Greater adjusted absolute improvement in appropriate antithrombotic prescription, blood pressure control and smoking cessation was seen among the 42 incentive clinics compared with the 42 control clinics. Medicaid and uninsured patients did better on all variables at incentive clinics, with the exception of cholesterol control. The authors concluded that the incentive program modestly improved certain measures of cardiovascular care but noted that studies longer than a year are needed.

An accompanying editorial said that both studies contribute to the field of knowledge regarding the role of financial incentives in quality improvement but noted that it is not clear how much such incentives can actually affect health care costs. The findings, said the editorialists, seem to suggest that differences in performance may be related to “systematic shortcomings” rather than differences in individual clinician performance.

The two trials “provide a great opportunity to ask the question of how to best use the limited but expensive primary care physician workforce in the most effective manner in the evolving health care delivery system,” the editorialists wrote. “But this question also suggests that a clear understanding of system-level goals and responsibilities is needed to enable the transformation of clinical care.”