Few studies have addressed the association between polypharmacy and hip fracture using population data. We conducted a population-based case-control study to investigate whether polypharmacy increases the risk for hip fracture in the elderly. We used insurance claims data from the Taiwan Bureau of National Health Insurance, a universal insurance program with a coverage rate of more than 98% of the population in Taiwan. We identified 2328 elderly patients with newly diagnosed hip fracture during the period 2005-2007. We randomly selected 9312 individuals without hip fracture to serve as the control group. Patient characteristics, drugs prescribed by physicians, and all types of hip fracture were ascertained. The odds ratio (OR) of hip fracture in association with the number of medications used per day in previous years was assessed.We found that patients were older than controls, predominantly female, and more likely to use 5 or more drugs (22.2% vs. 9.3%, p < 0.0001). The OR of hip fracture increased with the number of medications used per day and with age. Multivariate logistic regression analysis revealed that the overall OR for patients using 10 or more drugs was 8.42 (95% confidence interval [CI], 4.73-15.0) compared with patients who used 0-1 drug per day. However, age-specific analysis revealed that the risk for hip fracture was 23 times greater for patients aged > or = 85 years who used 10 or more drugs than for those aged 65-74 years who used 0-1 drug after controlling for covariates (OR, 23.0; 95% CI, 3.77-140).We conclude that the risk of hip fracture in older people increases with the number of medications used, especially in women. Age interacts with the daily medications for the risk of hip fracture.