Although osteoporosis is common in older adults, it is often under-diagnosed and under-treated. We developed community-based patient- and physician-directed interventions for fracture prevention and compared them in a 2 x 2 factorial randomized controlled trial. The study population included older adults who were enrolled in a state-run pharmacy benefits program (The Pharmaceutical Assistance Contract for the Elderly in Pennsylvania) for Medicare beneficiaries. We randomly assigned 826 primary care physicians and their 31,715 patients to one of four trial arms--no patient and no physician intervention, patient but no physician intervention, physician but no patient intervention, both patient and physician interventions. The patient intervention consisted of targeted communication about fall and fracture prevention and osteoporosis diagnosis and treatment. It was delivered through several mailings. The physician intervention entailed one-on-one academic detailing encounters covering the same topics. The composite primary endpoint consisted of use of osteoporosis medication or a bone mineral density test. Other endpoints included patient's knowledge and attitudes towards fractures and osteoporosis, use of lower extremity strengthening to prevent falls, and the occurrence of fractures. All outcomes will be analyzed using random effects models accounting for clustering of subjects within physicians' practices.