Frailty is considered highly prevalent in aging populations and is associated with increased risk of adverse health outcomes, including mortality, institutionalization, falls, and hospitalization (Bharathi & Baby). There is no universal definition of frailty, but there is a general consensus that frailty begins as a consequence of decline in multiple physiological systems, particularly the neuromuscular, neuroendocrine, and immunological systems (Bharathi & Baby). This decline results in susceptibility to a sudden change in health status (Strandberg & Pitkälä, 2007). Geriatricians typically define frailty as a biologic syndrome of decreased reserve and resistance to stressors, resulting from cumulative declines across multiple physiologic systems, and causing vulnerability to adverse outcomes (Fried et al., 2001).
The ability to engage in activities of daily living such as walking, bathing, dressing, and eating is crucial for an independent life. Frailty is associated with decreased muscle strength, balance, and functional ability in older adults (Strandberg & Pitkälä, 2007), and may present an obstacle to these activities and personal independence. However, the beneficial effects of exercise on the functional capacity of older people has been supported by evidence (King, Rejeski, & Buchner, 1998). Strengthening exercises have been shown to improve physical function and pre-clinical disability in older adults who have impairments in physical performance (Binder et al., 2002). Older adults may not have the money or access to transportation that would facilitate access to the equipment, usually available in a physical therapy clinic, which has been shown to help them achieve increases in strength (Brock, Manton, & Woodbury, 1990). Moreover, the concept of exercising with younger people in public facilities may make them feel embarrassed or self-conscious. Hence, their ability and willingness to perform regular exercise away from home may be constrained.
The advantages of home-based exercise are convenience and low cost, but adherence and sufficient exercise progression are a concern with this type of exercise. Thus, the purpose of the present review is to systematically examine the effectiveness of home-based exercises on improvement in the functional performance of frail older adults.Methods This literature review was conducted via a systematic search for all randomized controlled trials (RCTs) that evaluated the effect of home-based exercise on functional performance in frail older adults.
Publications were accessed through electronic databases, including PubMed, the Web of Science, and Cochrane. The first phase was conducted without any limitations on the year of publication, and included the search terms “frailty,” “frail older adult,” “functional performance,” and “home based exercise.” The second phrase of research limited selections to RCTs conducted within the past 20 years, or those that had been published beginning in January 1997.
Article titles and abstracts were examined to select pieces meeting the inclusion criteria. These criteria are outlined as follows: -“Frail older adults” were defined for the purposes of this research as demonstrating perceptible physical limitations associated with frailty. Participants in each study were living at home, able to follow an exercise program, and currently not involved in any exercise program.-Only original RCTs published in English were considered. All selected RCTs examine, as a primary outcome, the effects of the effects of the home-based exercise on functional ability in the domain of frailty.
Each of these studies implemented home-based exercise programs for the intervention group, while the control group maintained their routine lifestyle and received no treatment. -Non-RCTs, cohort studies, and studies that accompanied exercise with other intervention like diet or hormonal supplements were excluded.Quality Assessment The methodological quality of each RCT was rated using the Physiotherapy Evidence Database (PEDro) scale (Cohen, 1988). The PEDro scale is an instrument for assessing the methodological quality of RCTs in physical therapy and exercise studies. Scores range from 0 to 10, with higher scores indicating greater methodological quality.
Responses to items 2 through 11 are summed to create a total score, while item 1 relates to external validity. The reliability of this scale was evaluated with acceptable good results in intraclass correlation coefficients 0.56–0.91 (Maher, Sherrington, Herbert, Moseley, & Elkins, 2003). The assessment of the studies according to the PEDro scale presented in Table 1 (Appendix A).Types of Outcomes The outcome measures for this review are measures of functional performance.
The first of these is the 6-minute walk (6MW) test, which measures the maximum distance that an individual can walk in 6 minutes (Enright et al., 2003). The 6MW test is generally considered to be a valid and reliable measure and is commonly used to assess functional status. The second is the physical performance test (PPT), which objectively assesses the performance of everyday activities that challenge balance, mobility, coordination, and fine and gross motor function. In this test, seven items are timed and scored, with the sum of all scores ranging between 6 and 28 (Weiner, Duncan, Chandler, & Studenski, 1992). Third, the timed up and go (TUG) test measures the time it takes a subject to stand up from an armchair, walk a distance of 3 meters, turn, walk back to the chair, and sit down. The test is used as a short assessment of basic mobility for frail older adults (Podsiadlo & Richardson, 1991).
Fourth, summary of disability score which is used to provide data on function in terms of activities of daily living, eight activities of daily living, ranged from 0 (no disability) to 16 (total disability)(Landis & Koch, 1977). Finally, some RCTs used handhold or isokinetic dynamometers in their assessments, which are considered reliable measures of muscle strength (Bohannon, 1986; Feiring, Ellenbecker, & Derscheid, 1990).ResultsStudy Selection All titles and abstracts were screened for eligibility, and only studies considered to be at low risk of methodological bias were considered. Figure 1 (Appendix B) summarizes the steps taken to identify appropriate articles for the review. A search of relevant databases yielded 39 results, and after removal of duplicates, nine studies were considered potential candidates and were screened for relevant content. In the next phase, five out of the nine studies were excluded based on the inclusion criteria. Two trials were excluded for not meeting the study design parameters, one was excluded for not meeting the type of training characteristics, and two were excluded because their participation criteria did not meet the needs of this review. Finally, four studies met the inclusion criteria and were included in this research.
Methodological Quality PEDro scores ranged from 6 to 8 points, with a mean score of 6.5 (Table 1). All selected studies scored 6, except for one that received a score of 8 (Latham et al., 2003). These results indicate the high quality of the selected trials.
All included studies specified criteria for participant eligibility. In all studies, the subjects were randomly allocated to groups, and only one used allocation concealment (Latham et al., 2003). All studies showed similarities at baseline. None of the trials blinded participants or therapists, but three had blinded assessors (Chandler, Duncan, Kochersberger, & Studenski, 1998; Gill et al.
, 2002; Latham et al., 2003). Three studies met the intention-to-treat analysis criteria (Gill et al., 2002; Latham et al.
, 2003; Vestergaard, Kronborg, & Puggaard, 2008). All studies applied statistical analysis to group differences, and reported point estimates and measurements of variability. No studies were excluded on the basis of methodological quality.Participants and Study Characteristics Two studies were conducted in the US (Chandler et al., 1998; Gill et al.
, 2002), one in Denmark (Vestergaard et al., 2008), and one in Australia (Latham et al., 2003).The RCTS included a combined sample population of 589 older adults (66.2% women), with a mean age of 80.3±2.
4 years old. All participants were community-dwelling older adults. A total of 349 were recruited from primary care units, while 240 were recruited from geriatric rehabilitation units (inpatient or day ward). The identification of frail older adults was based on Winograd’s frailty scale (Latham et al., 2003), tests of physical abilities (Gill et al.
, 2002; Vestergaard et al., 2008) , and the inability to descend stairs step over step without holding the railing. Poor stair descent has been associated with increased risk for falls (Chandler et al., 1998; Studenski et al.
, 1994). The intervention lasted 6 months in one trial (Gill et al., 2002), 5 months in another (Vestergaard et al., 2008), and 10 weeks in the remaining two (Chandler et al., 1998; Latham et al.
, 2003). The frequency of the training programs was reported as 3 times per week in all RCTs. Only one trail reported the length of each session to be 30 minutes videotaped exercise (Vestergaard et al.
, 2008). Three of the four selected studies provided a detailed description of the intensity of the exercise intervention (Chandler et al., 1998; Latham et al., 2003; Vestergaard et al., 2008). One of the RCTs used ankle weights as the core of the training program (Latham et al., 2003).
The participants completed a series of three sets of eight repetitions of each exercise, with a final goal set at 60–80% of their 1-RM (initial goal set at 3 –40% 1RM, accomplished goal 51±13 % 1-RM) (Latham et al., 2003). The other three RCTs used thera-bands as the foundation of the training program. Lower and upper body exercises consisted of resisted hip extension and abduction, knee flexion and extension, ankle dorsiflexion, toe raises, chair rises, and stair stepping and arm exercises.
The intervention targeted both upper and lower extremities in two RCTs (Gill et al., 2002; Vestergaard et al., 2008), and only the lower extremities in the remaining two (Chandler et al., 1998; Latham et al., 2003). Table 2 summarizes the characteristics of each study (Appendix C).
Outcome MeasurementsMobility Several aspects of mobility were measured in three of the four trials (Chandler et al., 1998; Latham et al., 2003; Vestergaard et al.
, 2008). Individualized standard measurements of mobility were used, including measures of walking ability using several variations of walking tests. Results showed no changes in the walking ability of intervention groups (Chandler et al.
, 1998; Latham et al., 2003). Additionally, walking speed did not improve (Vestergaard et al., 2008). Latham et al. (2003) also used the TUG test to assess for changes in mobility, but the intervention had no effect on this measurement. Functional ability Functional ability was measured in two trials (Latham et al.
, 2003; Vestergaard et al., 2008). Vestergaard et al. (2008) used the physical performance test (PPT) to assess the performance of everyday activities, and the intervention group showed significant improvement. By contrast, Latham et al. (2003) used the Barthal Index, and found no significant improvement in the intervention group.
On the other hand functions of activities of daily living was reported and measured in one trail using the disability score and found a significant improvement with the intervention (Gill et al., 2002).Muscle strength Three out of the four included RCTs measured both upper and lower body muscle strength (Chandler et al., 1998; Latham et al., 2003; Vestergaard et al., 2008). Chandler et al. (1998) reported improved lower body strength, while Latham et al.
(2003) noted no significant improvement. Vestergaard et al. (2008) found no significant improvement in either upper or lower body strength. Discussion This systematic review summarizes four RCTs that recruited a combined sample of 589 participants to examine the potential effects of home-based physical exercise interventions on the functional and physical performance of frail older adults. In executing this review, strict criteria were used to define frailty in older people. Selected studies not only included the term “frail” in their titles or abstracts, but were required to use the term in a way that was consistent with the parameters of this research. All included trials were assessed to have sufficient methodological quality for inclusion in this review. The evidence presented by these studies suggests that home-based physical exercise has positive effects on the functional and physical outcomes measured.
Two of the trials applied only lower extremity strength training (Chandler et al., 1998; Latham et al., 2003). Although one of them showed significant improvements in strength and mobility skill performances such as gait, transfer, stooping, and stair climbing, increased strength did not improve performance in the 6MW test. The initial level of frailty could be a confounding factor and affect the impact of improvements in strength on performance (Chandler et al., 1998). However, On the other hand, Latham et al.
(2003) found home-based exercise interventions to have no effect on mobility, functional ability, or strength. These negative results could be explained by the short duration of the program used in the study (only ten weeks), or because the exercise program targeted only one muscle group (quadriceps), hence limiting greater improvements in performance. The contradictory results of studies using only strength training of the lower extremities could be further parsed and potentially resolved by including strength training as part of a more complete program, such as a multi-component exercise program.
In this review, two RCTs included both upper and lower body strengthening exercises (Gill et al., 2002; Vestergaard et al., 2008). One of the two trials included only female participants and used home-based exercise videos that emphasized flexibility and dynamic balance, strengthening exercises for the arms and legs using a thera-band, and aerobic exercises. Improvements were found in handgrip and biceps strength, chair rise, physical performance test results, and mobility-tiredness score in the intervention group (Vestergaard et al.
, 2008). Adherence to the training program was 89%. Participants in this study had difficulty going outdoors and socializing outside of the home, and were therefore more likely to comply with prescribed exercise routines. Gill et al.’s (2002) study, which also utilized both upper and lower body strengthening exercises, required participants to perform progressive, competency-based conditioning exercises of the arms and legs with elastic resistance bands. An improvement in disability score was found in people with moderate frailty, and was maintained at 12 months. No improvement was reported for people with severe frailty (Gill et al., 2002).
The cause of disability in elderly persons is highly complex and is associated with specific risk factors, including impairments in physical ability and cognitive status, and subsequent illnesses (Fried & Guralnik, 1997). This may explain why participants with severe frailty did not benefit from the intervention. Despite the evidence presented by this systematic review that home-based physical exercise has positive effects on frail older adults, it is important to note that there was significant variation among the studies regarding sample size, degree of frailty, types of intervention, and forms of assessment. Furthermore, the lack of change in some outcomes measured in this analysis could indicate the need for exercise programs to be carefully tailored to the populations they are applied to. Conclusion This systematic review demonstrates that home-based exercise is a simple, safe and widely applicable intervention to prevent dependency and decline in frail older populations. Moreover, exercise training specific to frail elderly people and tailored to meet the needs of individuals could improve many aspects of their physical function. These results suggest that multi-component exercise programs that include some type of resistance training may prove especially effective in promoting improved performance.