Background: Clinical trials have demonstrated that drug therapy can reduce osteoporosis-related fracture risk in women over 50 years of age. Noncompliance could considerably limit the effectiveness observed in actual practice, however. The objective of this study was therefore to estimate fracture risk in relation to compliance with osteoporosis medication in actual practice. Methods: Demographic, prescription drug use, physician services, and hospitalization information for women with osteoporosis who were dispensed an osteoporosis medication between 1996 and 2001 was obtained from the Saskatchewan health data files. Compliance to treatment was defined as drug available to cover 80% of the time. Subsequent fractures were identified via hospitalizations or physician contacts with a relevant diagnostic or procedure code. The risk of fractures in relation to compliance was examined using a Cox proportional hazards model with time-dependent covariates. The impact of other patient characteristics, including age, having suffered a prior fracture, and prior use of osteoporosis medication and steroids, was also examined. Results: 11,249 women suffering from osteoporosis were identified with a mean age at the time of the index prescription of 68.4 years and average follow-up of 2 years. The overall fracture rate was 4.5% per year. Patients who complied experienced a 16% lower fracture rate. This association was maintained within subgroups and after controlling for other patient characteristics that independently predict the fracture rate. Conclusion: These results indicate that improving compliance in actual practice may significantly decrease osteoporosis-related fracture risk.