To determine if the risk of hip fracture difference between persons receiving benzodiazepines with long (greater than or equal to 24 hours) or short (less than 24 hours) elimination half-lives, we conducted a nested case-control study among residents of the Canadian province of Saskatchewan who were 65 years of age and older. We identified 4501 cases occurring between 1977 and 1985 from computerized hospital records and 24,041 population controls. Current benzodiazepine use, defined as having filled a prescription in the past 30 days, was ascertained from computerized pharmacy records. The relative risk of hip fracture was 1.7 (95% confidence interval, 1.5 to 2.0) for current users of long half-life benzodiazepines, in contrast to that of 1.1 (95% confidence interval, 0.9 to 1.3) for current users of short half-life drugs. This finding was not altered by sex, age, calendar year, nursing home residence, or history of hospitalization. Medical record review for a sample of 189 cases suggested that this finding was not due to confounding by dementia, ambulatory status, functional status, or body mass.