Systematic Review of the Effect of Home Modification and Fall Prevention Programs on Falls and the Performance of Community-Dwelling Older Adults

Multifactorial Interventions

Multifactorial approaches to fall prevention combine multiple interventions in one package. The types of interventions included may vary but often incorporate several of the following approaches: home modifications, education on health and safety, medication management, vision management, gait and balance training, and exercise. The disciplines providing these services may also vary but often include occupational therapy, internal medicine, physical therapy, nursing, and social services.

The evidence that multifactorial approaches reduce falls and difficulties with ADLs and IADLs in older adults is strong. Ten Level I RCTs explored multifactorial approaches (Clemson et al., 2004; Close et al., 1999; Davison, Bond, Dawson, Steen, & Kenny, 2005; Day et al., 2002; Gitlin et al., 2006; Hogan et al., 2001; Hornbrook et al., 1994; Logan et al., 2010; Nikolaus & Bach, 2003; Shumway-Cook et al., 2007). Clemson et al. (2004) reported a 31% reduction in falls at 14-mo follow-up for the Stepping On program, which included occupational therapy and incorporated environmental and home safety, balance, strength, vision screening, and medication management. Davison et al. (2005) reported 36% fewer falls after an intervention that combined occupational therapy, physical therapy, and medical management for older adults presenting to an emergency department after a fall. In addition, participants had a shorter duration of hospital admissions during follow-up and better falls efficacy than those in the control group. Groups did not differ, however, on the number of participants falling and the number of hospital admissions.

Nikolaus and Bach (2003) reported 31% fewer falls in the year after an intervention that included a diagnostic home visit and home intervention assessing the home for environmental hazards and providing advice for hazard reduction. In addition, training in the use of technical and mobility aids was provided, and a second home visit took place 3 mo later. Occupational therapy, physical therapy, nursing, and social work provided the intervention. Nikolaus and Bach reported that the intervention was most effective in a subgroup that had fallen 2 or more times before study recruitment. The results indicated that the rate of falls and proportion of frequent fallers was significantly reduced in this subgroup.

Close et al. (1999) found that an occupational therapy home assessment and a medical visit with referrals as appropriate led to a reduction in falls and fall risk, a decrease in the chance of hospital admission, and a slower decline in ADL function at 1-yr follow-up. All participants were recruited to the study after presenting to the emergency department after a fall. Participants in the intervention group were provided with minor equipment and referred to social services for additional adaptations. All participants in the Logan et al. (2010) study were recruited after calling an ambulance after a fall. The multifactorial intervention incorporated strength and balance exercises, home assessment of hazards, and review of medications and blood pressure as provided by occupational therapy, physical therapy, and nursing. Referrals were made to other agencies as needed. The results indicated fewer calls for an ambulance, fewer falls, better performance in ADLs and IADLs, and a significant decrease in fear of falling at follow-up.

Gitlin et al. (2006) evaluated the effectiveness of an individualized program to reduce difficulties in ADLs and IADLs, improve self-efficacy, and reduce fear of falling in community-dwelling older adults >70 yr old with difficulty in one or more ADLs. Five sessions provided by an occupational therapist examined environmental hazards and incorporated problem solving to identify behavioral and environmental contributors to performance difficulties. A single visit by a physical therapist focused on balance and muscle strengthening and fall recovery techniques. Results of the study indicated that participants in the intervention group had less difficulty with ADLs and IADLs, improved self-efficacy, and decreased fear of falling compared with control participants. In addition, those in the intervention group had fewer home hazards and more use of adaptive strategies, with significant improvements persisting at 1-yr follow-up.

Day et al. (2002) studied falls among older adults living at home with good to excellent self-reported health. Using an intervention that incorporated group-based exercise, home hazard management, and vision in various combinations, they reported that exercise alone provided the strongest evidence for a single intervention. The addition of home hazard management, vision management, or both further reduced falls. Shumway-Cook et al. (2007) provided group exercise, fall prevention education, and falls risk assessment to older sedentary adults. Although the intervention group had a 25% decrease in falls at follow-up compared with the control group, the results were not significant. Hogan et al. (2001) found no difference in falls and emergency department visits between groups for an intervention providing an individualized treatment plan plus exercise for older adults who fell within 3 mo before study recruitment. They reported, however, that participants who had reported two or more falls during the prerecruitment period had a significantly longer time between falls and fewer falls.

Physical Activity Interventions

Studies of physical activity interventions in the systematic review included group and individual sessions that incorporated balance retraining, walking, general exercise in sitting and standing, lower-extremity strengthening, use of a workstation format, or tai chi. Mixed but overall positive results were found in studies that measured the impact of physical activity programs on the performance of community-dwelling older adults. Seven RCTs reported a significant decrease in falls and fall risk after physical activity, regardless of the type of exercise incorporated into each study (Buchner et al., 1997; Faber, Bosscher, Chin, Paw, & van Wieringen, 2006; Gardner, 1998; Lord, Ward, Williams, & Strudwick, 1995; Means, Rodell, & O’Sullivan, 2005; Skelton, Dinan, Campbell, & Rutherford, 2005; Voukelatos, Cumming, Lord, & Rissel, 2007). Exercise programs in these studies included functional walking, tai chi, balance, flexibility, lower-limb strengthening, and gait training. Voukelatos et al. (2007) also reported a reduction in fear of falling in a study of tai chi with community-dwelling older adults.

Lin, Wolf, Hwang, Gong, and Chen (2007) examined the effects of several fall prevention programs; the group participating in an exercise program showed improved functional reach and balance scores and a decreased fear of falling compared with either an education group or a home safety assessment and modification group. In a Level I RCT of Finnish older adults living in the community, Luukinen et al. (2007) found no significant difference between control and intervention groups in fall rates after exercise; however, results indicated a significant decrease in the decline of balance skills over a 6-mo period. In another Level I RCT, Lord et al., 2005) compared an extensive intervention combining exercise with strategies for maximizing vision and sensation with a minimal intervention providing brief advice and with a no-intervention control. Although researchers found no difference in fall rates, the participants receiving the most extensive intervention had a decrease in some physiological risk factors related to falls such as knee flexion and sit-to-stand time but no improvement in balance.

Nitz and Choy (2004), in a Level I RCT, compared a fall prevention pamphlet control group with a group participating in a series of workstations of balance activities in addition to receiving a pamphlet. Participants in the workstation group showed improvement in functional motor ability, lateral reach, and functional step compared with control participants, but no difference in falls was found between groups at follow-up. In Hauer et al. (2001), older adults in a geriatric rehabilitation unit were randomized to either an exercise program emphasizing strength, balance, and functional performance or a placebo control group before discharge to the community. Both groups received physical therapy twice a week that did not emphasize strength and balance. The results indicated that those in the exercise intervention group had improved strength, functional motor performance, and balance and reduced fear of falling compared with the control group. No differences were found between groups for falls.

In several Level I RCTs included in the systematic review, results varied with age, fall history, and participant’s activity level. For those over age 80, strengthening, balance retraining that progressed in difficulty, and a walking plan led to fewer falls and fewer injuries when falls did occur (Campbell et al., 1997; Robertson, Devlin, Gardner, & Campbell, 2001). Campbell et al. (1997), however, found no difference in ADL performance at follow-up. Means et al. (2005) reported that participants in an exercise intervention group with a history of multiple falls had fewer falls in follow-up than did the control group with a similar fall history. Morgan, Virnig, Duque, Abdel-Moty, and DeVito (2004) reported that falls among higher functioning older adults increased after an exercise program, whereas lower functioning adults had a decrease in the risk of falls.

Home Assessment and Home Modifications

The home assessment and home modification interventions in this section of the systematic review included hazard identification, structural changes to the inside and outside of the home, and provision of assistive technology and assistive devices. A Level II prospective cohort study (Liu & Lapane, 2009) surveyed older adults at baseline and during a 2-yr follow-up period as part of the second Longitudinal Study of Aging. Liu and Lapane (2009) found that those having residential modification at baseline (e.g., railings, bathroom modifications) were less likely to experience a decline in physical functioning. In a Level I RCT, Cumming et al. (1999) compared occupational therapy home visits for home modifications for older adults before hospital discharge with a no-intervention control group. The occupational therapist facilitated the implementation of the home modifications. Although both groups showed no difference in falls at follow-up, a significant difference was found between groups at follow-up for participants reporting one or more falls in the year before recruitment.

In a Level II nonrandomized controlled trial in Sweden (Petersson, Kottorp, Bergström, & Lilja, 2009), occupational therapists provided assessments for home modifications for older adults with disabilities reporting difficulty in at least one of these areas: getting in and out of the home, mobility indoors, or self-care in the bathroom. The local government provided grants to install the home modifications. Participants in the intervention group reported significantly less difficulty in everyday tasks at follow-up than the comparison group.

In a Level I RCT, Campbell et al. (2005) studied targeted falls prevention provided to older adults ≥75 yr old with severe visual impairment. Home safety assessment and modification provided by an occupational therapist was compared with vitamin D supplementation and a home exercise program provided by a physical therapist, a combination of both interventions, or a social visit. Although the home safety group had significantly fewer falls than the exercise group, strict adherence to exercise was also associated with fewer falls. Neither group reported reduced injuries as a result of the falls. In a Level I RCT by Mann, Ottenbacher, Fraas, Tomita, and Granger (1999), an occupational therapist provided a functional assessment, environmental interventions, and assistive technology to frail older adults living alone. Although both intervention and control groups had a functional decline at 18-mo follow-up, the control group declined significantly more than the control group. Tomita, Mann, Stanton, Tomita, and Sundar (2007; Level I RCT) evaluated the effectiveness of commercially available smart home technology to operate lights, appliances, door, and windows for frail older adults living alone. Although no difference was found in IADL performance between groups at follow-up, those in the intervention group reported a high degree of satisfaction with the technology. Stevens, Holman, Bennett, and de Klerk (2001; Level I RCT) reported no differences between groups when a home modification assessment and information on hazard reduction was provided to healthy older adults by a research nurse.

A Level I RCT (Pighills, Torgerson, Sheldon, Drummond, & Bland, 2011) compared the effectiveness of a home modification assessment by an occupational therapist with that of a nonprofessional assessor. The study also included a no-treatment control group. Both the occupational therapists and the nonprofessional assessors received training in home modification assessment. Outside agencies provided the home modifications for the older adults. Although Pighills et al. (2011) found no effect on fear of falling, those in the occupational therapy assessment group had fewer falls than did those in the control group. No difference was found when comparing falls for the trained assessor and control groups, and no difference was found in ADLs at follow-up between either occupational therapy or assessor groups versus the control group.