OBJECTIVES: To study the cooperation of primary care physicians with a community-based prevention and health promotion program for older persons, to study physician factors related to cooperation, and to determine any relationship between physician cooperation and patient adherence to program recommendations and patient satisfaction with health care. DESIGN AND SETTING: A survey administered in subjects' homes and physicians' offices in Santa Monica, California. PARTICIPANTS: Patients (n = 81) were intervention group subjects in a 3-year, randomized, controlled trial of in-home comprehensive geriatric assessment paired with prevention and health promotion. Physicians (n = 50) were selected if they had been contacted at least once by a study nurse practitioner about one of these patients. MEASUREMENTS: Physician cooperation was rated by study nurse practitioners. Physicians were interviewed to identify factors associated with cooperation. Patients' satisfaction with health care and adherence were measured prospectively throughout the 3-year program. MAIN RESULTS: Physicians exhibiting better cooperation had fewer years in practice (P = .03) and were more likely to discuss the program with their patients (P = .005), see benefit for their patients from the program (P = .02), and rate program information as useful (P = .002). Higher physician cooperation did not predict higher patient satisfaction (P = .23) but did predict higher patient adherence to program recommendations (P = .02). CONCLUSIONS: Physicians rated as cooperative were more likely to have a positive appraisal of the program, and their patients had higher adherence to program recommendations. These findings suggest that strategies for increasing primary care physician cooperation might improve effectiveness of similar community-based prevention and health promotion programs.