Comparison of a fall risk assessment tool with nurses' judgement alone: a cluster-randomised controlled trial

BACKGROUND: the impact of fall risk assessment tools on clinical endpoints is unknown. OBJECTIVE: we compared a standardised fall risk assessment tool alongside nurses' clinical judgement with nurses' judgement alone. DESIGN: a 12-month cluster-randomised controlled trial. SETTING: nursing homes in Hamburg (29 per study group). SUBJECTS: 1,125 residents (n = 574 intervention group, IG; n = 551 control group, CG). INTERVENTIONS: all homes received structured information on fall prevention before randomisation. The IG monthly administered the Downton Index, and the CG did not use a tool. Measurements were number of participants with at least one fall, falls, fall-related injuries and medical attention, fall preventive measures, physical restraints. RESULTS: the mean follow-up was 10.8 +/- 2.9 months in both groups: 105 (IG) and 114 (CG) residents died or moved away. There was no difference between the groups concerning the number of residents with at least one fall (IG: 52%, CG: 53%, mean difference -0.7, 95% confidence interval -10.3 to 8.9, P = 0.88) and the number of falls (n = 1,016 and n = 1,014). All other outcomes were also comparable between the IG and CG. CONCLUSIONS: application of a fall risk assessment tool in nursing homes does not result in the better clinical outcome than reliance on nurses' clinical judgement alone.