Objective: To examine the characteristics of benzodiazepines usage and their associations with hip fractures. Method: All subjects were aged 65 and older and enrolled in the National Health Insurance program in Taiwan, 2001-2004. Cases (N = 217) were elderly patients who were identified with hip fractures for the first time in their outpatient claims. They were individually matched to 1,214 comparison patients based on age, gender, and index year. Benzodiazepine usage (doses, duration, half-life) and the other covariates including comorbidities, health care utilization, and psychotropic medications used in the 180 days before index events were constructed. Results: Using nonusers as reference group, use of benzodiazepines was significantly associated with hip fractures (adjusted odds ratio [AOR] = 1.7, 95% confidence interval [CI] = 1.2-2.5). Such risks appear to be particularly high during the first month (AOR = 5.6, 95% CI = 2.7-11.8) of exposure, doses higher than 3.0 mg/day in diazepam equivalents (AOR = 1.8, 95% CI = 1.1-3.1), and using short-acting benzodiazepines (AOR = 1.8, 95% CI = 1.3-2.7). Conclusions: Benzodiazepine exposure in the elderly increases the risk of hip fractures. This is true even with modest dosage, short-acting agents and short-term exposures. Clinicians should prescribe benzodiazepines judiciously with the elderly to minimize drug-related hip fractures.