Interventions for preventing falls in older people in nursing care facilities and hospitals

For each of the last few years, the interest in ways to prevent falls among older people made “Interventions for preventing falls in elderly people” the most accessed of all Cochrane reviews. It was accessed once every 59 minutes on average during the 2008 leap year in the Wiley InterScience version of The Cochrane Library alone. This review has now been split in two, providing separate assessments of the evidence on interventions for use in the community and on those for use in nursing care facilities and hospitals. The community review was published in 2009 and the residential facilities review, which is the subject of this Cochrane Journal Club, appeared in The Cochrane Library in Issue 1 of 2010.

The review includes randomised trials or quasi-randomised trials comparing any intervention designed to reduce falls in older people versus any other intervention, usual care or placebo; where the majority of the participants were in nursing care facilities or hospitals. “Nursing care facilities participants” were defined as residents of establishments primarily engaged in providing residential nursing and rehabilitation services, generally for an extended period of time. Such establishments would have a permanent core staff of registered or licensed practical nurses who, along with other staff, provide nursing and continuous personal care services. “Hospital participants” were defined as patients who had been admitted to in-patient wards. The hospitals were divided into those providing acute and those providing subacute care. The latter was defined as medical and skilled nursing services provided to patients who were not in an acute phase of an illness but who required a level of care higher than that provided in a long-term care setting. Emergency departments, outpatient departments and “hospital at home” settings were excluded. Trials were considered for inclusion if most of the participants were over 65 years of age, or the mean age in the trial was over 65 years.

The primary outcomes for the review were falls (meaning the number of falls, which might be expressed as the fall rate per person year, and yields a rate ratio) and fallers (meaning the number of people who fall). Secondary outcomes were severity of falls (for example, number of falls resulting in injury), fractures, deaths and complications of the interventions; but there were insufficient data on these for most of the trials. The meta-analyses were done separately for the results from trials in nursing care facilities and those in hospitals, with the main statistics being either a rate ratio from comparing the rate of falls in the intervention group versus the control group, or a risk ratio from the number of participants in each group with one or more falls during each trial.

The review includes data from 41 trials, with more than 25,000 participants. There are also a further 34 studies that were excluded, three studies awaiting classification and ten ongoing studies for which the results are not yet available. The 41 included trials were from a total of 13 different countries. Thirty studies (18,388 participants) were from nursing care facilities and 11 studies (7054 participants) had been done in hospitals. The mean age of residents in the nursing care facilities was 84 years, compared to 80 years for the hospital patients.

In nursing care facilities, seven trials of supervised exercise interventions had inconsistent results, with considerable heterogeneity in the meta-analyses. When the trials of multifactorial interventions were combined the overall rate of falls and risk of falling were not significantly different between the intervention and control groups. However, a post hoc subgroup analysis found that multifactorial interventions provided by a multidisciplinary team reduced the rate of falls (rate ratio [RaR] 0.60, 95% CI 0.51 to 0.72; 4 trials, 1651 participants) and risk of falling (risk ratio [RR] 0.85, 95% CI 0.77 to 0.95; 5 trials, 1925 participants). Vitamin D supplementation reduced the rate of falls in nursing care facilities (RaR 0.72, 95% CI 0.55 to 0.95; 4 trials, 4512 participants), but not risk of falling (RR 0.98, 95% CI 0.89 to 1.09; 5 trials, 5095 participants). In hospitals, multifactorial interventions reduced the rate of falls (RaR 0.69, 95% CI 0.49 to 0.96; 4 trials, 6478 participants) and risk of falling (RR 0.73, 95% CI 0.56 to 0.96; 3 trials, 4824 participants). The three trials (131 participants) of supervised exercise interventions in subacute hospitals showed a significant reduction in risk of falling (RR 0.44, 95% CI 0.20 to 0.97).

In conclusion, the current evidence reveals that vitamin D supplementation is effective in reducing the rate of falls in nursing care facilities, multifactorial interventions reduce falls and risk of falling in hospitals, and exercise appears effective in subacute hospital settings. However, the effectiveness of multifactorial interventions and exercise in nursing care facilities remains uncertain. The Cochrane review also highlights topics for future randomised trials, including the examination of current approaches to falls prevention for which there is a limited research base, such as increased supervision of patients deemed to be at high risk, and the use of technologies, such as monitoring and alarm systems.