Interventions to Prevent Falls in Community-Dwelling Older Adults: A Systematic Review for the U.S. Preventive Services Task Force [Internet]

Rockville (MD): Agency for Healthcare Research and Quality (US); 2018 Apr. Report No.: 17-05232-EF-1.


Objective: We conducted this systematic review to support the U.S. Preventive Services Task Force in updating its recommendation on the prevention of falls in older adults. Our review addressed the following questions: 1) Is there direct evidence that primary care interventions to prevent falls in community-dwelling older adults at average or high risk for falls, used alone or in combination, reduce falls or fall-related injury, improve quality of life, reduce disability, or reduce mortality? 1a) How is high risk assessed in the included trials? 2) What are the adverse effects associated with primary care interventions to prevent falls in community-dwelling older adults?

Data Sources: We searched MEDLINE, PubMed publisher-supplied records, Cumulative Index for Nursing and Allied Health Literature, and Cochrane Central Register of Controlled Trials for relevant English-language literature published between January 1, 2010, and August 30, 2016. Additionally, we re-evaluated all studies included in the 2010 review. We supplemented our searches with reference lists from relevant existing systematic reviews, suggestions from experts, and to identify ongoing trials. We conducted ongoing surveillance through November 22, 2017 to identify any major studies published in the interim.

Study Selection: We included the following study designs: randomized placebo-controlled trials on the effectiveness of interventions to prevent falls in older adults; randomized controlled trials on the harms of fall interventions; and systematic reviews and randomized control trials on the harms of vitamin D to prevent falls in older adults. Two investigators independently reviewed identified abstracts and full-text articles against a set of a priori inclusion and quality criteria.

Data Analysis: One investigator abstracted data into an evidence table and a second investigator confirmed these data. Two investigators independently assessed study quality using methods developed by the USPSTF. We qualitatively synthesized the data for each key question and meta-analyzed trial results when appropriate.

Results: We identified 62 trials (n=35,058) examining seven intervention types aimed at reducing the risk of falls and fall-related outcomes. The largest bodies of literature evaluated multifactorial and exercise interventions with 26 and 21 trials, respectively. Our findings suggest that there is a fall-related benefit associated with both multifactorial and exercise interventions, but evidence is most consistent across multiple fall-related outcomes for the exercise trials. Meta-analysis of multifactorial intervention trials showed a 21 percent reduction in falls with substantial heterogeneity (17 RCTs; n=9,737; incidence rate ratio [IRR], 0.79 [95% CI, 0.68 to 0.91]; I2=87.2%) but no statistically significant effect on people experiencing a fall (24 RCTs; n=12,490; relative risk [RR], 0.95 [95% CI, 0.89 to 1.01]; I2=56.4%), people experiencing a fall-related injury (16 RCTs; n=9,445; RR 0.94, [95% CI, 0.85 to 1.03]; I2=34.3%) or mortality (23 RCTs; n=9,721; RR, 0.96 [95% CI, 0.79 to 1.17]; I2=0%) at 6 to 36 months of followup. Small numbers of the multifactorial studies reported no statistically significant effect on fall-related injuries, fall-related fractures, people experiencing fall-related fractures, activities of daily living (ADL), quality of life (QOL), hospitalization and institutionalization, but were underpowered for these outcomes. Meta-analysis of exercise trials showed an 11 percent reduction in people experiencing a fall (15 RCTs; n=4,926; RR, 0.89 [95% CI, 0.81 to 0.97]; I2=43.9%), a 13 percent nonstatistically significant reduction in falls (14 RCTs, n=4,663; IRR, 0.87 [95% CI, 0.75 to 1.00]; I2=57.3%), a 19 percent reduction in injurious falls (10 RCTs, n=4,622; IRR, 0.81 [95% CI, 0.73 to 0.90]; I2=0.0%), and a qualitative reduction in people experiencing an injurious fall, with estimates ranging from 0.61 to 0.90 (5 RCTs, n=2,776) and no individual study reaching statistical significance. There was no effect on mortality (11 RCTs, n=4,263; RR, 0.93 [95% CI, 0.71 to 1.22; I2=0%) at 12 to 60 months of followup. Hospitalizations, institutionalizations, QOL and instrumental activities of daily living (IADL) outcomes were reported in a few exercise trials showing no statistically significant effect. Seven heterogeneous trials (n=7,531) of different vitamin D formulations (with or without calcium), dosing schedules, and varying baseline fall risk show mixed results at 9 to 36 months of followup. The single trial of annual high-dose cholecalciferol (500,000 IU) showed an increase in falls, people experiencing a fall, and injuries, while one trial of calcitriol showed a reduction in falls and people experiencing a fall; the remaining five trials showed no statistical difference in falls, people experiencing a fall, or injuries. A single study reported no difference in QOL, and no studies reported on the outcomes of hospitalizations, institutionalizations, or ADL/IADL for vitamin D interventions. Three environment intervention trials (n=2,175) reported mixed results at 12 to 18 months of followup: one trial showed a 46 percent reduction in falls (IRR, 0.54 [95% CI, 0.36 to 0.83]), while two trials showed no statistically significant effect on falls. None of the trials reporting people experiencing a fall, injuries, QOL, or ADLs showed any statistically significant differences between the intervention group and control; and no environment trials reported mortality, hospitalization, institutionalization, or harms outcomes. Two underpowered medication management RCTs (n=266) showed no difference in fall-related outcomes or mortality. Two cognitive behavioral intervention trials (n= 886) showed mixed results in people experiencing a fall, a nonstatistically significant reductions in falls, mixed results for injuries, and no difference in mortality. Six trials (n=1,770) examined the effectiveness of multiple interventions, with one to two trials testing each of the following combinations of interventions compared to control: exercise+environment, exercise+psychological, exercise+knowledge+fall-risk assessment, exercise+vitamin D, and knowledge+environment. One trial of knowledge+environment (n=310), one trial of exercise+environment+vision (n=272), and one trial of exercise+psychological (n=378) interventions reported fewer falls and/or people experiencing a fall by 20 to 46 percent. Other multiple intervention combinations showed no statistically significant difference in falls, fallers, or injuries with the exception of a single exercise+vitamin D trial, which showed a large, statistically significant reduction in injurious falls (IRR, 0.38 [95% CI, 0.17 to 0.81]) despite no difference in falls or people experiencing a fall. In the 62 included trials, 65 percent of RCTs were conducted in high-risk populations. Multifactorial interventions were more likely to recruit high-risk populations (73%, 19 of 26), while other intervention types were equally or more likely to include average risk populations. Definitions of high risk were variably defined but most often included history of falls as at least one criterion. Harms were not consistently reported for any intervention type. Some exercise trials and multifactorial intervention trials with exercise components reported largely minor adverse effects associated with muscle soreness. One high-dose vitamin D trial reported an increase in falls outcomes, which has not been replicated in other trials; otherwise, vitamin D trials reported similar adverse events in the vitamin D and control groups.

Limitations: Our search was limited to English-language literature. We excluded trials specifically recruiting participants with neurologic conditions (e.g., Parkinson’s disease) and other specific diagnoses (e.g., vitamin D insufficiency, osteoporosis) so our findings may not be applicable to these populations. Our review protocol prioritized hard health outcomes (falls, fallers, injuries) and did not include changes in balance, endurance or walking speeds nor did it include falls efficacy or fear of falling.

Conclusions: The current evidence base demonstrates that exercise is associated with fewer people experiencing a fall and a reduced number of injurious falls in average- and high-risk older adults. Multifactorial interventions appear to reduce falls but not people experiencing a fall or injuries; trials are clinically and statistically heterogeneous. No specific effective exercise or multifactorial protocol has been replicated in larger population trials. Vitamin D, environment, and medication management interventions have either single trials showing no statistically significant effect or a few trials reporting mixed results. Single trials of cognitive behavioral, knowledge+environment, and exercise+environment+vision interventions showed moderate effectiveness in reducing falls and/or people experiencing a fall.

Publication types

  • Review

Grants and funding

Prepared for: Agency for Healthcare Research and Quality, U.S. Department of Health and Human Services, 5600 Fishers Lane, Rockville, MD 20857; www.ahrq.govContract No. HHSA-290-2012-00015-I, Task Order No. 6Prepared by: Kaiser Permanente Research Affiliates Evidence-based Practice Center Kaiser Permanente Center for Health Research Portland, OR