A randomized controlled trial of fall prevention strategies in old peoples' homes

Background: Falls are a major cause of morbidity in old age. A small number of fall prevention trials in cognitively intact community-dwelling older p eople have been effective. This study set out to examine the preventability of falls in older people living in institutional care. Objective: To evalu ate the effectiveness of falls risk factor assessment/modification and seat ed balance exercise training in reducing falls among elderly people living in residential care. Methods: 133 residents with a mean age of 84 +/- (SD) 6.8 years were allocated at random by home to receive either a 8-month fall s risk factor assessment/modification and seated balance exercise training programme (n = 77) or 6 months of reminiscence therapy (n = 55). The risk f actors targeted were postural hypotension, polypharmacy, visual acuity, and ambient lighting levels. Falls risk factor assessments and recommendation for modifications were performed at baseline in the intervention group and assessments repeated at 6 months. Functional reach, reaction time, timed up -and- go, grip strength, spinal flexibility, and Philadelphia Geriatric Cen tre Morale Scale and Mini-Mental State Examination scores were determined a t baseline and at 6 months by a 'blind' observer. Falls and fractures were then monitored in both groups during a 7- to 12-month falls-monitoring foll ow-up period. Results: Only 90 of 133 (67.7%) residents completed the 6-mon th intervention period, and 84 (63.2%) completed the 7- to 12-month falls-m onitoring follow-up period. Both prevalence of postural hypotension (p = 0. 0005) and poor visual acuity (p = 0.04) were reduced in the intervention gr oup. There was no difference between the groups in the number of falls sust ained, the risk of falling [odds ratio 0.45 (95% Cl 0.19-1.14)], or in the risk of recurrent falling [odds ratio 1.07 (95% CI 0.40-2.97)]. No signific ant differences were found between the groups with regard to change in othe r outcome measures. Conclusions: The high drop-out rate reduced the power o f th is study to detect a ny effect of the interventions used. It is possib le that either the exercises were not sufficiently vigorous or that to impr ove balance exercises must be performed standing. Further research is requi red to identify effective fall prevention strategies for elderly people in residential settings. Copyright (C) 2000 S.Karger AG, Basel.