OBJECTIVE:

To determine whether benzodiazepines are associated with an increased risk of hip fracture.

DESIGN:

Case-control study.

PARTICIPANTS:

All incident cases of hip fracture not related to traffic accidents or cancer in patients over 65 years of age. 245 cases were matched to 817 controls.

SETTING:

Emergency department of a university hospital.

MAIN OUTCOME MEASURES:

Exposure to benzodiazepines and other potential risk or protective factors or lifestyle items.

RESULTS:

The use of benzodiazepines as determined from questionnaires, medical records, or plasma samples at admission to hospital was not associated with an increased risk of hip fracture (odds ratio 0.9, 95% confidence interval 0.5 to 1.5). Hip fracture was, however, associated with the use of two or more benzodiazepines, as determined from questionnaires or medical records but not from plasma samples. Of the individual drugs, only lorazepam was significantly associated with an increased risk of hip fracture (1.8, 1.1 to 3.1).

CONCLUSION:

Except for lorazepam, the presence of benzodiazepines in plasma was not associated with an increased risk of hip fracture. The method used to ascertain exposure could influence the results of case-control studies.