The decrements of aging lead to a gradual wear and tear and deterioration of body systems; in particular, muscle mass and strength decline with age. Conditions which older adults are predisposed to, such as poor vision, unsteady gait, impaired balance, chronic disease, frailty, and effects of polypharmacy, increase their risk for falls. Falls are the third leading cause of chronic disability among older adults and pose a serious threat to safety and independence (Dadgari et al., 2016). According to the Center for Disease Control and Prevention (CDC, 2017), more than one in four older adults fall each year. Falls have a significant impact on the health care system, accounting for thousands of visits to urgent care and emergency departments and they impose a hefty financial burden related to injuries and restorative care. Due to the personal consequences of falling and the burden of falls on the health care system, mitigation of risk and reduction of harm are important objectives for health care providers.

Background and significance

Aging is associated with decreased muscle mass, strength, and bone density, which increases the risk for balance instability, falls, and injury (CDC, 2017). It is estimated that older adults lose approximately 1% of their muscle mass and 3% of muscle strength each year (Brown et al., 2016).

One in three adults aged 65 years and older and one half of adults older than 80 years report falling in the past year (Flaherty & Resnick, 2019). Today, there are approximately 50 million older adults, and the annual expenditure on fall-related complications paid by Medicare is estimated to be 51 billion dollars (CDC, 2017). The Census Bureau (2017) estimates that by the year 2040, the number of older adults will reach 80 million. As this cohort of adults age, costs are expected to rise, further increasing the economic burden on the health care system and stretching operational capacities of rehabilitation centers and long-term care facilities.

Complications resulting from a fall, such as hip fracture, subdural hematoma, or traumatic brain injury, are the leading cause of death from injury in the older adult (Flaherty & Resnick, 2019). The seriousness of these injuries illustrates why falls often precipitate loss of independence, debilitation, decline in functional status, and/or death. Furthermore, the experience of falling can impose a heavy psychological burden on older adults. The fear of falling can limit physical activity and can escalate functional decline, further increasing the risk of falling (CDC, 2017). A fall not only affects the older adult but has the potential to place additional burdens on caregivers, family members, and the care team.

It is reported that those who maintain an exercise regimen are less likely to experience low bone density, falls, and fractures when compared with those who are sedentary (Santos et al., 2017). There is a lack of data demonstrating which exercise regimens are safe and effective for older adults to participate in as a means of fall prevention. Health care providers, in turn, may be hesitant to prescribe exercise programs for this population. They may also struggle with how to individualize the prescription to prevent harm and how to account for the unique characteristics and abilities of each patient (Dadgari et al., 2016).

Prevalence and incidence

The true incidence and prevalence of falls is not known because falls that do not result in injury are underreported. The incidence of falls is estimated to be one in three to four per year up to age 80 years, after which the incidence increases to one in two (Flaherty & Resnick, 2019). The most recent data available from the CDC estimates the prevalence of falls among older adults at 29% per year (2017). As the population of older adults grows, it is expected that both the incidence and the prevalence will continue to rise.

Purpose statement

Improved strength and mobility may provide benefits to older adults at risk for falling. Physiologic changes of aging and reduced mobility lead to a higher incidence of falls in the older person, often resulting in serious injury or death. The purpose of this article was to review the available evidence and determine whether a community- or home-based exercise program is effective at reducing the occurrence of falls in community-dwelling older adults.

Methods

This systematic review was conducted with attention to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. PubMed and Cumulative Index to Nursing and Allied Health Literature were queried to evaluate the effectiveness of home- and community-based exercise programs as a fall-prevention intervention for community-dwelling older adults. The following keywords and MeSH terms were used: exercise OR physical activity, home, fall*. The search was limited to the highest available level of evidence, randomized controlled trials, and meta-analyses published no earlier than 2014. The search was not limited to English-only; articles in all journals were considered and the reference lists of key articles were reviewed. The final search yielded 198 articles for review.

Selection criteria

Review of article titles and abstracts were performed to eliminate duplicates and those that did not meet the inclusion criteria. Figure 1 illustrates the article selection process. Initial results were narrowed to remove duplicates and articles that did not fit inclusion criteria or were incongruous based on screening of the title or abstract. Only full-text articles were included. Studies were narrowed from 198 initial records screened to a final of 15 studies that met the criteria. The articles selected were reviewed by the authors and evaluated for their contribution to the body of knowledge surrounding this topic. A total of 14 randomized controlled trials and one quasi-experimental intervention study were identified. These included participants aged 60 years or older who live independently in the community and who engaged in either a home-based or community-based exercise intervention with measured fall rates as an outcome. Studies were excluded if the participants had debilitating physical or cognitive conditions, if exercise was not medically appropriate, or if the participants did not live independently. Examples of exclusions include participants with Parkinson disease, physical or cognitive limitations after a stroke, dementia, and those residing in a care facility (Table 1).

Figure 1. Search strategy tree.

Table 1. Search Strategy

Data extraction

This review includes 15 studies with a total of 7,694 participants. All studies included the rate of falls and many also included the rate of falls that caused injury. Table 2, "Data Analysis" (see Supplemental Digital Content 1, http://links.lww.com/JAANP/A125), demonstrates study characteristics and Table 3, "Study Analysis" (see Supplemental Digital Content 2, http://links.lww.com/JAANP/A126), details relevant outcomes of the studies.

Data analysis

Once all studies were selected, each author used the Critical Appraisal Skills Program (CASP) tool to appraise each randomized control trial. The CASP tool is a reliable instrument for assessing the quality of RCTs. The three major topics considered when evaluating these studies include evaluating the validity, believability, and generalizability of the study. All authors reviewed the studies that met inclusion criteria and agreed that each article was done with sufficient rigor to report accurate and valuable data.

Quality of evidence

Given the high level of evidence of the trials included in this review, the authors determined that the results are believable, generalizable, and valid (see Table 4, Supplemental Digital Content 3, http://links.lww.com/JAANP/A127). Each study was assessed for strengths, biases, and limitations. All the studies reported statistical significance with 95% to 99% confidence intervals. The rigor of each trial was reviewed, and each was found to satisfactorily account for differences among participants.

Sample characteristics

Participants were from different areas of the world including the United Kingdom, Germany, France, Thailand, Egypt, Iran, Finland, Australia, Canada, and the United States. Recruitment methods varied among the studies; eight studies recruited participants from primary care providers, clinics, or emergency departments; four recruited participants by targeted mail and advertisements; two recruited participants from fall prevention clinics; and two studies recruited participants from other related studies. All participants were at least 60 years old and were disproportionately female. It is unclear if this gender discrepancy represents differing life expectancies, living independence, overall health status, or willingness to participate. All participants were community dwelling and were without serious medical or cognitive conditions that would make exercising harmful or difficult. Many of the participants had sustained a fall before the study. All participants were able to partake in the intervention independently.

Outcomes

Fall reduction

Although 11 studies reported a reduction in the total number of falls among participants in the exercise group, only seven studies found this reduction to be statistically significant (Liu-Ambrose et al, 2019; Barker et al., 2019; Siegrist et al., 2016; Mohammed et al., 2019; Dadgari et al., 2016; Iliffe et al, 2014; Li, et al, 2018). Five of the trials used the Otago Exercise Program (OEP) as part of their intervention. The trial by Iliffe et al. (2014) compared the OEP to a weekly group exercise class plus walking and usual care, finding that the group exercise class plus walking was superior to both OEP and usual care alone. All the statistically significant studies included a health care professional in the recruitment process and only one study (Barker et al., 2019) did not include in-person exercise instruction. Instead, Barker et al. involved regular telephonic coaching and goal setting with linkage to community support services. The remaining eight trials did not find a statistically significant reduction in the rate of falls among participants in the intervention group (Arkkukangas et al., 2019; Boongird et al., 2017; El-Khoury et al., 2015; Gallo et al., 2018; Gill et al., 2016; Patil et al., 2015; Suttanon et. al., 2018; Voulukelatos et al., 2015), but three trials found a reduction in falls at large (Boongird et al., 2017; El-Khoury et al., 2015; Patil et al., 2015) and two trials found a statistically significant reduction in injurious falls (El-Khoury et al., 2015; Patil et al., 2015). Overall, unsuccessful trials were more likely to have a hands-off approach, in which the participants completed self-paced walking programs (Boongird et al., 2017; Voukelatos et al., 2015) or were provided with written educational material with instruction for exercises to complete independently (Suttanon et. al., 2018).

Safety and injurious falls

Injurious falls were reported in six studies (Barker et al., 2019; El- Khoury et al., 2015; Gill et al., 2016; Li et al., 2018; Patil et al., 2015; Siegrist et al., 2016). A wide variety of injuries were accounted for, including bruises, skin injuries, sprains/strains and fractures, and other nonspecified injuries. Of the six studies that reported falls with injuries, five demonstrated that falls with injuries were reduced among participants in the exercise intervention group compared with usual care (Barker et al., 2019; El- Khoury et al., 2015; Gill et al., 2016; Patil et al., 2015; Siegrist et al., 2016), whereas one study failed to demonstrate this (Li et al., 2018). The remaining studies did not report injuries as a result of falls. Eight studies (Barker et al., 2019; Liu-Ambrose et al, 2019; Siegrist et al., 2016; Suttanon et. al., 2018; Boongird et al., 2017; Arkkukangas et al., 2019; Iliffe et al, 2014; Li, et al, 2018) reported injuries and adverse events occurring during the exercise intervention, and noninjurious falls reported were reported in only one study (Li, et al., 2018).

Synthesis and critique

In this review, 11 studies found a reduction in falls among the participants in the intervention group, and this reduction was statistically significant in seven studies. Three additional studies found a statistically significant reduction in falls with injury. The studies that found exercise to be an effective fall reduction intervention were more likely to involve formal instruction led by a physical therapist or health care provider, with methods centering on regular group exercise as well as instruction-based sessions within the home. This indicates the value that exercise with regular instruction lends to reducing fall risk in older adults. Many unsuccessful trials used recruitment methods that did not include a health care provider, such as newspaper advertisements, mail in questionnaires, and targeted mass mailings. These unsuccessful trials were more likely to include self-paced walking (Voukelatos et al., 2015), weekly group exercises plus walking (Patil et al., 2015), and motivational interviewing (Arkkukangiaset et al, 2019). Individual factors such as personal motivation, access to transportation, caregiving roles, and family support likely influenced participation in these programs. A noteworthy finding of this analysis is that exercise was found to be a safe intervention for older adults. This finding, paired with the value of instruction-based exercise for reducing fall risk, should encourage health care practitioners to actively engage older patients in discussions about exercise and make referrals to community-based programs as appropriate.

Implications for practice

Reducing the number of falls that cause injury is a worthwhile goal for older persons because it will allow individuals to maintain their independence. The evidence shows that exercise can be an effective fall prevention measure for community-dwelling older adults. This review also highlights the value of the health care provider's participation, recommendation, and referral to a formal program as a strategy for fall prevention and harm reduction. Health care providers should familiarize themselves with community-based resources for exercise and fall reduction to recommend appropriate exercise programs for at-risk patients, and they should encourage patients to perform similar exercises at home when possible. Additionally, providers should be involved in program instruction and follow-up to reduce falls in this population.

Limitations

A common limitation of these studies is the inability to blind the intervention group, practitioners, and data collectors. There were two studies that maintained double blinding until after group allocation and completion of baseline assessments. Nine studies maintained blinding of the data collector and/or the interventionist throughout the trial. Four studies did not blind participants or data collectors or interventionists at any point. Other potential limitations include the reliance on participants self-reporting their fall occurrences and their compliance with the exercise program. Attrition was also a limiting factor in many trials because many participants failed to follow-up due to illness, difficulty with travel, or scheduling conflicts. Only one study (Boongird et al., 2017) reported the presence of visual and hearing deficits, which are common among older people and may influence ability to participate in group exercise sessions. One study recruited relatively active adults, whereas seven studies recruited participants with known balance issues or a history of falls. In all the statistically significant studies, there was a mean age of less than 75 years. Other factors that may have influenced the ability of participants to be consistent with the program include familial support, personal motivation, professional obligations, and whether the participant acted as a caregiver for a partner. Seven studies were not controlled for those participants who may have been more physically active at the start of the program because they targeted participants with a previous fall or at high risk of falling.

This review highlights the benefits of exercise in preventing falls, preserving physical independence, and reducing fall-related harm among older persons living in the community. Evidence may be strengthened by further research that aims to determine the safest and most beneficial types of fall preventive exercises to guide providers' efforts to recommend reliable programs for their at-risk patients.

Acknowledgements:The authors would like to thank NYU Meyers Director of the Adult-Gerontology Primary Care NP Program Dr. Leslie-Faith Morritt Taub, PhD, ANP-C, GNP-BC, CDE, CBSM, FAANP, for her assistance with the outline of the design for this article, editorial assistance, and guidance in preparing this article.

References