Fund support to Robert Miller, Ph.D., has enabled a study of the costs and benefits of implementing electronic health records in solo or small group physician practices. His survey, published in Health Affairs,⁽⁹⁾ found that practices that had adopted the technology reaped financial benefits from improved billing and reduced personnel costs. While the practices had improved access to data, few used the electronic health records to systematically improve chronic and preventive care. Initial costs averaged $42,000 per provider and ongoing costs averaged $8,100. The average practice paid for its system in less than three years.

In the coming year, a grant to the Johns Hopkins School of Medicine will support a survey of physicians in 156 Texas hospitals to assess the structural and functional capabilities of their IT systems and determine whether these capabilities translate into improved quality and lower costs.

Programs that align payments to health care providers with the quality of care they deliver have been blossoming. More than 100 insurers nationwide have implemented so-called pay-for-performance initiatives, and Medicare is considering adopting such policies as well.

Despite this activity, evidence about the impact of performance incentives is still scarce. With Fund support, Meredith Rosenthal, Ph.D., at the Harvard School of Public Health conducted one of the first formal evaluations of a large pay-for-performance program, implemented by the PacifiCare health plan in 2003, among more than 200 California group practices. In May 2005, Rosenthal presented preliminary findings to members of the House Subcommittee on Employer-Employee Relations, as well as to the IOM Committee on Redesigning Health Insurance Benefits, Payment and Performance Improvement Programs. Her study, published in the Journal of the American Medical Association,⁽¹⁰⁾ shows that, compared with PacifiCare's physician groups in Oregon and Washington that did not participate in the incentive program, PacifiCare's California network demonstrated greater quality improvement on one of three clinical measures. Although practices that were historically high performers earned the most rewards, lower-performing practices improved significantly.