Objectives: to evaluate the effectiveness of a policy of making hip protectors available to residents of nursing homes.

Design: a cluster randomised controlled trial of the policy in nursing and residential homes, with the home as the unit of randomisation.

Setting: 127 nursing and residential homes in the greater Belfast area of Northern Ireland.

Participants: 40 homes in the intervention group (representing 1,366 occupied beds) and 87 homes in the control group (representing 2,751 occupied beds).

Interventions: a policy of making hip protectors available free of charge to residents of nursing homes and supporting the implementation process by employing a nurse facilitator to encourage staff in the homes to promote their use, over a 72-week period.

Main outcome measures: the rate of hip fractures in intervention and control homes, and the level of adherence to use of hip protectors.

Results: there were 85 hip fractures in the intervention homes and 163 in the control homes. The mean fracture rate per 100 residents was 6.22 in the intervention homes and 5.92 in the control homes, giving an adjusted rate ratio for the intervention group compared to the control group of 1.05 (95% CI 0.77, 1.43, P = 0.76). Initial acceptance of the hip protectors was 37.2% (508/1,366) with adherence falling to 19.9% (272/1,366) at 72 weeks.

Conclusions: making hip protectors available to residents of nursing and residential homes did not reduce the rate of hip fracture. This research does not support the introduction of a policy of providing hip protectors to residents of nursing homes.