Hip protectors in aged-care facilities: a randomized trial of use by individual higher-risk residents

OBJECTIVES: To investigate the effect of use of external hip protectors on occurrence of hip fracture. DESIGN: Randomized controlled trial, with randomization at the individual level. SETTING: residential aged-care facilities in urban areas of New South Wales, Australia. PARTICIPANTS: 174 women, aged 75 years and older, who had had two or more falls or one fall requiring hospital admission in the previous 3 months, and who lived in hostels or nursing homes. Eighty-six subjects were in the intervention group and 88 in the control group. INTERVENTION: Use of external hip protectors and encouragement by nurses to use the protectors. MEASUREMENTS: Follow-up visits at approximately 2 weeks and 2, 10 and 18 months to determine falls and fall injury (including hip fracture); we also measured adherence to hip protector use. RESULTS: The mean age of participants was 85; they lived in 32 different aged-care facilities, two-thirds of which were nursing homes. Intervention and control groups had similar baseline characteristics, with a mean Barthel index of 58 at enrollment and a mean Short Portable Mental Status Questionnaire score of six errors, indicating severe disability and major cognitive impairment. During follow-up, a mean of 4.6 falls per person occurred. There was no difference in mortality, with 28 deaths in each group. Eight hip fractures occurred in the intervention group and seven in the control group (hazard ratio 1.46; 95% confidence interval 0.53-4.51). No hip fractures occurred when hip protectors were being worn as directed. Adherence was about 57% over the duration of the study and hip protectors were worn at the time of 54% of falls in the intervention group. Adherence varied markedly between institutions, but the greatest was about 80%. CONCLUSION: Hip protectors were not effective in reducing the incidence of hip fractures in this study, but because of low statistical power, a reduction in risk of hip fracture of up to 50% may not have been detected. There was limited adherence with their use, resulting in a large number of falls occurring without hip protectors in place. All hip fractures in the intervention group occurred when hip protectors were not being used.