Patient compliance with osteoporosis drug therapy is often poor in clinical practice and may be associated with higher risk of fracture. A nested case-control study was undertaken using a US health insurance claims database. The source population included all women aged >or=45 years who began drug therapy for osteoporosis. Cases consisted of those who experienced an osteoporosis-related fracture; they were matched to controls without osteoporosis-related fracture. Compliance with osteoporosis drug treatment was assessed in terms of the number of therapy-days received and medication possession ratio (MPR). Conditional logistic regression was employed to examine the relationship between compliance and fracture risk. A total of 453 women with osteoporosis-related fracture were identified and matched to 2,160 controls. Fracture risk was significantly lower for patients with >180 days of therapy [181-360 days: odds ratio (OR) = 0.70, 95% CI = 0.49-0.99; >360 days: OR = 0.65, 95% CI = 0.43-0.99) versus those with <or=30 days. Risk was also lower for patients with MPR >or=90% (OR = 0.70, 95% CI = 0.52-0.93) versus those with MPR <30%. Fracture risk decreased as compliance increased (p(trend) < 0.05). Among women initiating drug therapy for osteoporosis, better compliance is associated with reduced risk of fracture.