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

This review4 includes 41 randomised trials involving 25 422 participants. The average age of the participants was 83 years, and around three quarters were women. Many of the participants were cognitively impaired. The review reported the findings of the 30 trials conducted in nursing care facilities separately from the 11 trials in hospital settings. This reflects differences in participant characteristics between the two settings and that the differences in the environment may warrant different types of interventions, possibly implemented by people with different skill mixes. Only two trials, both in hospital settings, recruited participants with a particular health condition: respectively, stroke and hip fracture.

The review authors grouped and described interventions using the fall prevention classification system developed by the Prevention of Falls Network Europe: ProFANE.5 Where possible, the authors reported two outcomes: rate of falls (RaR) (based on the number of falls per person year) or risk of falling (RR) (based on number of trial participants who fell during follow-up). Also reported, where available, were fall related fractures. While sought, none of the included trials provided data on adverse effects of the interventions, although the possibility that some interventions increase the rate of falls was discussed in the review. None of the trials included cost effectiveness or cost utility evaluations. A summary of some of the main results of this review follows below.

Most of the trials conducted in both settings tested single interventions. Various types of supervised exercise were investigated in 11 trials in nursing care facilities, and pooled data from seven of these showed no significant difference in rate of falls or risk of falling. However, these results were inconsistent and may relate to differences among the trials in the types and intensity of exercise, the study populations, and variation in methodological quality. Probably influenced by the type of exercises, there appeared some possibility that exercise programmes for frail nursing care facility residents may increase their risk of falls. In contrast, results from three small trials conducted in subacute hospital settings showed a significant reduction in risk of falling from supervised exercise.

Pooled data from five large studies evaluating vitamin D supplementation in nursing care facilities showed a significant reduction in the rate of falls, but not risk of falling.

Pooled results from the majority of nine trials testing multifactorial interventions, which target known risk factors, in nursing care facilities did not show significantly reduced rate of falls or risk of falling. These, however, were a heterogeneous set of trials evaluating a complex intervention. A post hoc subgroup analysis pointed to a potential effect for multifactorial interventions where provided by a multidisciplinary team. There was evidence from four trials that multifactorial falls prevention programmes in hospitals for patients who have longer lengths of stay (at least 3 weeks) are effective.

Differences in study characteristics and findings for various interventions support the separate grouping of the trials by nursing care facilities and hospitals in this review. It is also pertinent to view the findings in the context of those of a related Cochrane review of interventions for preventing falls in community dwelling older people.6 Comparison of the findings of these two reviews is aided by their use of similar methodology including the using the ProFANE classification system. In contrast to the inconsistent findings for supervised exercise in nursing care facilities, a key finding of the second review is that group exercise is effective in reducing falls in older people in the community. Conversely, vitamin D was not found to be effective in reducing falls in community dwellers, although some exception was mooted for those with vitamin D insufficiency. Underlying vitamin D insufficiency may well be the basis for the protective effect found for vitamin D in nursing care facilities. It is noteworthy too that people with dementia were specifically excluded from the majority of the studies in the community dwellers review.

Together with those of over 4000 other Cochrane reviews, the full versions of these reviews are available in The Cochrane Library (http://www.thecochranelibrary.com). Both reviews highlighted in this article are registered with the Cochrane, Bone, Joint and Muscle Trauma Group (http://www.bjmtg.cochrane.org). The work of this group involves preparing, maintaining, and promoting the accessibility of systematic reviews on different aspects of the prevention, treatment, and rehabilitation of musculoskeletal injuries. People interested in contributing to this work can contact Lindsey Elstub, the Managing Editor, at lindsey.elstub{at}manchester.ac.uk.