OBJECTIVES: To compare the risk of fracture associated with initiating opioids with that of nonsteroidal anti‐inflammatory drugs (NSAIDs) and the variation in risk according to opioid dose, duration of action, and duration of use.

DESIGN: Retrospective cohort study.

SETTING: Two statewide pharmaceutical benefit programs for persons aged 65 and older.

PARTICIPANTS: Twelve thousand four hundred thirty‐six initiators of opioids and 4,874 initiators of NSAIDs began treatment between January 1, 1999, and December 31, 2006. Mean age at initiation of analgesia was 81; 85% of participants were female, and all had arthritis.

MEASUREMENTS: Cox proportional hazards models, adjusted for several potential confounders, quantified fracture risk. Study outcomes were fractures of the hip, humerus or ulna, or wrist, identified using a combination of diagnosis (International Classification of Diseases, Ninth Revision, Clinical Modification) and procedure (Common Procedural Terminology) codes.

RESULTS: There were 587 fracture events among the participants initiating opioids (120 fractures per 1,000 person‐years) and 38 fracture events among participants initiating NSAIDs (25 fractures per 1,000 person‐years) (hazard ratio (HR)=4.9, 95% confidence interval (CI)=3.5–6.9). Fracture risk was greater with higher opioid dose. Risk was greater for short‐acting opioids (HR=5.1, 95% CI=3.7–7.1) than for long‐acting opioids (HR=2.6, 95% CI=1.5–4.4), even in participants taking equianalgesic doses, with differential fracture risk apparent for the first 2 weeks after starting opioids but not thereafter.

CONCLUSION: Older people with arthritis who initiate therapy with opioids are more likely to experience a fracture than those who initiate NSAIDs. For the first 2 weeks after initiating opioid therapy, but not thereafter, short‐acting opioids are associated with a greater risk of fracture than are long‐acting opioids.