Population-based interventions for preventing falls and fall-related injuries in older people

Abstract

Background: Around one-third of older adults aged 65 years or older who live in the community fall each year. Interventions to prevent falls can be designed to target the whole community, rather than selected individuals. These population-level interventions may be facilitated by different healthcare, social care, and community-level agencies. They aim to tackle the determinants that lead to risk of falling in older people, and include components such as community-wide polices for vitamin D supplementation for older adults, reducing fall hazards in the community or people's homes, or providing public health information or implementation of public health programmes that reduce fall risk (e.g. low-cost or free gym membership for older adults to encourage increased physical activity).

Objectives: To review and synthesise the current evidence on the effects of population-based interventions for preventing falls and fall-related injuries in older people. We defined population-based interventions as community-wide initiatives to change the underlying societal, cultural, or environmental conditions increasing the risk of falling.

Search methods: We searched CENTRAL, MEDLINE, Embase, three other databases, and two trials registers in December 2020, and conducted a top-up search of CENTRAL, MEDLINE, and Embase in January 2023.

Selection criteria: We included randomised controlled trials (RCTs), cluster RCTs, trials with stepped-wedge designs, and controlled non-randomised studies evaluating population-level interventions for preventing falls and fall-related injuries in adults ≥ 60 years of age. Population-based interventions target entire communities. We excluded studies only targeting people at high risk of falling or with specific comorbidities, or residents living in institutionalised settings.

Data collection and analysis: We used standard methodological procedures expected by Cochrane, and used GRADE to assess the certainty of the evidence. We prioritised seven outcomes: rate of falls, number of fallers, number of people experiencing one or more fall-related injuries, number of people experiencing one or more fall-related fracture, number of people requiring hospital admission for one or more falls, adverse events, and economic analysis of interventions. Other outcomes of interest were: number of people experiencing one or more falls requiring medical attention, health-related quality of life, fall-related mortality, and concerns about falling.

Main results: We included nine studies: two cluster RCTs and seven non-randomised trials (of which five were controlled before-and-after studies (CBAs), and two were controlled interrupted time series (CITS)). The numbers of older adults in intervention and control regions ranged from 1200 to 137,000 older residents in seven studies. The other two studies reported only total population size rather than numbers of older adults (67,300 and 172,500 residents). Most studies used hospital record systems to collect outcome data, but three only used questionnaire data in a random sample of residents; one study used both methods of data collection. The studies lasted between 14 months and eight years. We used Prevention of Falls Network Europe (ProFaNE) taxonomy to classify the types of interventions. All studies evaluated multicomponent falls prevention interventions. One study (n = 4542) also included a medication and nutrition intervention. We did not pool data owing to lack of consistency in study designs. Medication or nutrition Older people in the intervention area were offered free-of-charge daily supplements of calcium carbonate and vitamin D₃. Although female residents exposed to this falls prevention programme had fewer fall-related hospital admissions (with no evidence of a difference for male residents) compared to a control area, we were unsure of this finding because the certainty of evidence was very low. This cluster RCT included high and unclear risks of bias in several domains, and we could not determine levels of imprecision in the effect estimate reported by study authors. Because this evidence is of very low certainty, we have not included quantitative results here. This study reported none of our other review outcomes. Multicomponent interventions Types of interventions included components of exercise, environment modification (home; community; public spaces), staff training, and knowledge and education. Studies included some or all of these components in their programme design. The effectiveness of multicomponent falls prevention interventions for all reported outcomes is uncertain. The two cluster RCTs included high or unclear risk of bias, and we had no reasons to upgrade the certainty of evidence from the non-randomised trial designs (which started as low-certainty evidence). We also noted possible imprecision in some effect estimates and inconsistent findings between studies. Given the very low-certainty evidence for all outcomes, we have not reported quantitative findings here. One cluster RCT reported lower rates of falls in the intervention area than the control area, with fewer people in the intervention area having one or more falls and fall-related injuries, but with little or no difference in the number of people having one or more fall-related fractures. In another cluster RCT (a multi-arm study), study authors reported no evidence of a difference in the number of female or male residents with falls leading to hospital admission after either a multicomponent intervention ("environmental and health programme") or a combination of this programme and the calcium and vitamin D₃ programme (above). One CBA reported no difference in rate of falls between intervention and control group areas, and another CBA reported no difference in rate of falls inside or outside the home. Two CBAs found no evidence of a difference in the number of fallers, and another CBA found no evidence of a difference in fall-related injuries. One CITS found no evidence of a difference in the number of people having one or more fall-related fractures. No studies reported adverse events.

Authors' conclusions: Given the very low-certainty evidence, we are unsure whether population-based multicomponent or nutrition and medication interventions are effective at reducing falls and fall-related injuries in older adults. Methodologically robust cluster RCTs with sufficiently large communities and numbers of clusters are needed. Establishing a rate of sampling for population-based studies would help in determining the size of communities to include. Interventions should be described in detail to allow investigation of effectiveness of individual components of multicomponent interventions; using the ProFaNE taxonomy for this would improve consistency between studies.