Benzodiazepines appear to increase the incidence of hip fractures. Their role as a time-dependent risk factor remains unclear. We therefore conducted a case-crossover study to determine whether the new use of benzodiazepines is associated with a rise in hip fractures. We analysed 49 months of the statutory health insurance Gmünder Ersatzkasse (GEK) and enrolled all first hip fractures. The index date was the day of hospital admission. Exposure to new use of benzodiazepines was compared within the five periods preceding the index date. Out of 1630 subjects, 223 (13.7%) had at least one prescription of benzodiazepines in the preceding 150 days before the index date, 74 (4.5%) of them as a first prescription. The average age of the population was 79.8 years (SD: 7.7). Odds ratio (OR) of hip fracture was highest during the initial 5 days of new use (OR: 3.43; 95% CI 1.15-10.20) and then declined to a non-significant OR of 1.59 (95% CI 0.96-2.63) after 30 days. In conclusion, the start of a new benzodiazepine is associated with an increased risk of hip fractures. However, the population attributable risk (PAR) and, therefore, the percentage of preventable events is small (PAR: 0.55%; 95% CI 0.05-1.06%).