We conducted a randomized controlled trial within the setting of a large drug benefit plan for Medicare beneficiaries. Primary care physicians and their patients were randomized to usual care, patient intervention only, physician intervention only, or both interventions. There was no difference in the probability of the primary composite endpoint (BMD test or osteoporosis medication) or in either of its components comparing the combined intervention group with usual care (risk ratio = 1.04; 95% CI, 0.85-1.26).
Introduction: Fractures from osteoporosis are associated with substantial morbidity, mortality, and cost. However, only a minority of at-risk older adults receives screening and/or treatment for this condition. We evaluated the effect of educational interventions for osteoporosis targeting at-risk patients, primary care physicians, or both.
Materials and methods: We conducted a randomized controlled trial within the setting of a large drug benefit plan for Medicare beneficiaries. Primary care physicians and their patients were randomized to usual care, patient intervention only, physician intervention only, or both interventions. The at-risk patients were women >or=65 yr of age, men and women >or=65 yr of age with a prior fracture, and men and women >or=65 yr of age who used oral glucocorticoids. The primary outcome studied was a composite of either undergoing a BMD test or initiating a medication used for osteoporosis. The secondary outcome was a hip, humerus, spine, or wrist fracture.
Results: We randomized 828 primary care physicians and their 13,455 eligible at-risk patients into four study arms. Physician and patient characteristics were very similar across all four groups. Across all four groups, the rate of the composite outcome was 10.3 per 100 person-years and did not differ between the usual care and the combined intervention groups (p = 0.5). In adjusted Cox proportional hazards models, there was no difference in the probability of the primary composite endpoint comparing the combined intervention group with usual care (risk ratio = 1.04; 95% CI, 0.85-1.26). There was also no difference in either of the components of the composite endpoint. The probability of fracture during follow-up was 4.2 per 100 person-years and did not differ by treatment assignment (p = 0.9).
Conclusions: In this trial, a relatively brief program of patient and/or physician education did not work to improve the management of osteoporosis. More intensive efforts should be considered for future quality improvement programs for osteoporosis.