Hip protectors appear to be effective in reducing the incidence of hip fractures. However, compliance is often poor. Therefore, the objective of this study was to examine the compliance and determinants of compliance with external hip protectors. A prospective study was performed in residents from apartment houses for the elderly, homes for the elderly and nursing homes with a high risk for hip fracture (n=276). The study was performed within the framework of the Amsterdam Hip Protector Study, a randomized controlled trial examining the effect of external hip protectors on the incidence of hip fractures. Compliance was assessed by unannounced visits at 1, 6 and 12 months after inclusion in the study. During the visits, a member of the research team checked whether the participant was wearing the hip protector and, if so, whether it was worn correctly. Furthermore, data on potential determinants of compliance were collected by interviewing the participants or their nurses. Compliance was 60.8% after 1 month (n=217), 44.7% after 6 months (n=246), and 37.0% after 12 months (n=230). Of those wearing the hip protector, 86.7%, 91.7% and 96.5% of the participants were wearing the hip protector correctly after 1, 6 and 12 months respectively; and 14.8%, 16.1% and 8.8% respectively reported wearing the hip protector at night. Compliance after 12 months was predicted by the compliance after 1 month (RR=2.04; 90% CI: 1.05-3.96). Furthermore, people who experienced one or more falls in the half year before baseline had a lower probability of being compliant at 6 months (RR=0.72; 90% CI: 0.52-0.99). In conclusion, compliance is a very important issue in hip protector research and implementation. Although, the compliance percentages were moderately high during the unannounced visits in this study, not everyone was wearing the protector correctly and most participants did not wear the hip protector during the night.