The Effect of Group Exercise on Physical Functioning and Falls in Frail Older People Living in Retirement Villages: A Randomized, Controlled Trial


  • This study was funded by the National Health and Medical Research Council of Australia, New South Wales Health, and MBF (Australia).

Address correspondence to Professor Stephen Lord, Prince of Wales Medical Research Institute, Barker Street, Randwick, Sydney, NSW, 2031, Australia. E-mail:


Objectives: To determine whether a 12-month program of group exercise can improve physical functioning and reduce the rate of falling in frail older people.

Design: Cluster randomized, controlled trial of 12 months duration.

Setting: Retirement villages in Sydney and Wollongong, Australia.

Participants: Five hundred fifty-one people aged 62 to 95 (mean±standard deviation=79.5±6.4) who were living in self- and intermediate-care retirement villages.

Measurements: Accidental falls, choice stepping reaction time, 6-minute walk distance postural sway, leaning balance, simple reaction time, and lower-limb muscle strength.

Results: Two hundred eighty subjects were randomized to the weight-bearing group exercise (GE) intervention that was designed to improve the ability of subjects to undertake activities for daily living. Subjects randomized to the control arm (n=271) attended flexibility and relaxation (FR) classes (n=90) or did not participate in a group activity (n=181). In spite of the reduced precision of cluster randomization, there were few differences in the baseline characteristics of the GE and combined control (CC) subjects, although the mean age of the GE group was higher than that of the CC group, and there were fewer men in the GE group. The mean number of classes attended was 39.4±28.7 for the GE subjects and 31.5±25.2 for the FR subjects. After adjusting for age and sex, there were 22% fewer falls during the trial in the GE group than in the CC group (incident rate ratio=0.78, 95% confidence interval (CI)=0.62–0.99), and 31% fewer falls in the 173 subjects who had fallen in the past year (incident rate ratio=0.69, 95% CI=0.48–0.99). At 6-month retest, the GE group performed significantly better than the CC group in tests of choice stepping reaction time, 6-minute walking distance, and simple reaction time requiring a hand press. The groups did not differ at retest in tests of strength, sway, or leaning balance.

Conclusion: These findings show that group exercise can prevent falls and maintain physical functioning in frail older people.