|| Checking for direct PDF access through Ovid
Adverse drug reactions constitute an important health problem among older people. Many older suffer from multi-morbidity, including cognitive, visual and balance impairment, which further increases the risk for adverse drug reactions. Fall-induced hip fractures are especially related to these impairments. In this study we examine the risk of fall-induced hip fracture shortly after initiation of drug treatment with benzodiazepines, Z-hypnotics, antipsychotics and antidepressants.We applied the case-crossover design to a population-based sample of 99 357 people (28.5% men and 71.5% women) with fall-induced hip fractures identified in the inpatient register in Sweden during February 2006 to December 2013. Register information on prescribed medications dispensed in the 28 day period prior to the date of hospitalisation due to hip fracture was compared to medication during a period of equal length 4 months earlier. Odds ratios (OR) and 95% confidence intervals (CI) were estimated using conditional logistic regression.Dispensed drugs were common in close relation to time of hip fracture, 5.9%–21.6% had a prescription filled for drugs in each of the groups benzodiazepines, antipsychotics, Z-hypnotics, and antidepressants drugs within the 28 day period prior to hospitalisation. The analyses showed increased risk for hip fracture 0–28 days after dispensed benzodiazepines (OR=1.21 (95% CI 1.13–1.30)), antipsychotics (OR=1.32 (95% CI 1.16–1.49)), Z-hypnotics (OR=1.36 (95% CI 1.28–1.45)), and antidepressants prescriptions (OR=1.25 (95% CI 1.17–1.34)). Higher risks were observed for shorter time intervals (i.e. the week) before hospitalisation).Our results confirm that use of benzodiazepines, antipsychotics, Z-hypnotics and antidepressants are associated with an increased risk of hip fracture. Hence, caution should be taken when prescribing these drugs to a frail elderly patient at high risk of hip fracture. Our results are unlikely to be explained by confounding, owing the case-crossover design of the study.