A randomized, controlled trial was conducted to assess the effects of a financial and office systems intervention to increase preventive care in physicians' offices for patients aged 65 years or older. A total of 1,914 patients from 10 primary-care medical practices in central North Carolina were randomized within practices to an intervention and a usual-care control group. The intervention consisted of full Medicare reimbursement to physicians for preventive care and health promotion packages (thus making these services free for patients), regular prompting of physicians to routinely schedule preventive care visits, a new office system in which nurses carried out many preventive procedures, and a form for charting preventive care. The performance of screening tests dramatically increased in the intervention group relative to control (P < 0.001), but there was evidence of lack of follow-up of abnormal findings by physicians. At the 2-year follow-up, there were minimal differences between intervention and control groups in health-related quality-of-life indicators. Relative to the $294 per patient 3-year cost to Medicare for waivered services, the intervention was reimbursed-cost neutral or slightly cost reducing ($190 over 3 years) for Medicare. It is concluded that adding reimbursement for preventive services to Medicare--even with the office systems changes made in this study--will not by itself lead to effective implementation of preventive services in community medical practices. To enhance patient benefit from preventive services, greater attention needs to be focused on an organized approach to patient follow-up.