Each year, 1 in every 3 adults older than 65 years is involved in an accidental fall.1 Falls can precipitate physical injuries, both fatal and nonfatal,1 as well as psychosocial implications such as a reduction in confidence, physical activity, and community participation.2,3 Falls in this demographic cost the US health care system $28.2 billion in 2010 alone.1 A recent study found that the cost of a fall-related injury was 1.85 times more than implementing a falls-prevention program.4 As health care systems expand and adapt to meet the needs of an increasing population of older adults, treatment and prevention of falls have emerged as a central focus of physical therapy (PT) for this age group.5
The Affordable Care Act6 has stipulations that would add coverage and eliminate cost sharing by patients for preventative services that are appropriate and are recommended by the US Preventative Services Task Force (USPSTF).6,7 The USPSTF7 recommends interventions including exercise, physical therapy, and/or vitamin D supplementation for at-risk community dwellers older than 65 years based on a history of falls, history of mobility problems, or poor performance on the Timed Up and Go (TUG).7 Review of the current literature is consistent with the findings of the USPSTF.7 Recent meta-analyses and systematic reviews have concluded that exercise as a single intervention is effective in reducing risk and rate of falls by modifying intrinsic risk factors.2,8,9 In fact, Sherrington et al2 reported that up to 42% of falls could be prevented by well-designed exercise interventions. Physical therapists, as experts in exercise and movement, are in a pivotal position to provide this preventative service to at-risk populations.10,11
Effective programs generally focus on strength, flexibility, balance, and reaction time components,12 and studies suggest that exercise may increase balance confidence13 and participation in life roles.14 In addition, increased socialization, support, and sense of personal accomplishment may be achieved through exercise in a group format.15,16 Socialization and self-efficacy are key motivators for older adults to initiate and maintain exercise programs17; thus, group-based exercise may also contribute to a higher adherence rate.
Parameters of exercise programs that may be most effective for falls prevention have been outlined by current research, which defines the ideal intensity, duration, and mode of intervention. Evidence from meta-analyses and systematic reviews suggests that exercise programs should provide a moderate to high challenge of balance,18 including activities that reduce the base of support, enhance control of movement of the center of gravity, and reduce the need for upper limb support in standing.2 Exercise programs should be carried out over a longer time period (minimum 3 months) and include a minimal effective dosage of 50 hours, which often equates to 2 times per week over a 6-month period.2 In addition, interventions supervised and delivered by a physical therapist have been shown to produce the most positive outcomes related to falls prevention.5,12 Finally, to achieve a lasting falls-prevention effect, the research advocates for ongoing exercise intervention to maintain the benefits of exercise.2
The high dosage and level of supervision necessary for effective exercise interventions for at-risk individuals have significant fiscal and administrative restraints, thus making adequate individualized therapy difficult to administer within the consumer's and health care provider's financial barriers. When third-party payers are involved, individualized PT treatment sessions can range from $60.00 to $90.00, while billable group therapy is only $19.74 per session.19 Recent research has also demonstrated that group-based exercise can reduce direct medical costs for individuals,20,21 while it also may provide better allocation of economic resources.22 Therefore, group-based exercise may be a viable option for falls prevention,2,3,5,8,23,24 with possible added benefits of convenience, peer support, social interaction, and potential cost savings to health care systems and consumers.20 To our knowledge, there has been no systematic review describing the literature on a physical therapist–led group-based exercise compared with individualized clinic-based or home-based exercise programs. The purpose of this systematic review was to synthesize the evidence from randomized controlled or clinical trials that compare the physical therapist–administered group-based exercise with traditional individual PT interventions on patient outcomes in older adults.
This systematic review used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines throughout the research and reporting process. The PRISMA statement includes a 27-item checklist designed to improve the reporting of systematic reviews and meta-analyses of randomized trials.25 It is not designed to be a quality-assessment instrument.25
A computer-based search strategy used the following search terms (MeSH terms) in PubMed: (PT modalities or exercise) and (postural balance or accidental falls/prevention control) and (“exercise program” [TiAb] or “group program” [TiAb]). The following limits were applied to this strategy: English, date, human, clinical trial, and randomized controlled trial (RCT). A similar search strategy was applied to CINAHL. Because “group exercise” is not a specific term, computerized search results may have omitted many relevant articles; therefore, a systematic hand search of the references among articles selected for review was also performed.
Studies were reviewed if they met the following inclusion criteria: (1) RCT or clinical trial focusing on the (2) comparison of physical therapist–developed or supervised group exercise to a control with no intervention or to traditional individual PT intervention in (3) ambulatory men or women, aged 65 years or older, who (4) lived in the community and/or in an institution and evaluated effectiveness through the (5) use of outcome measure(s) related to functional balance and/or quality of life. Studies also had to be published in English between December 1, 2001, and June 7, 2012.
A stepwise, hierarchical review was performed. Two authors acted as reviewers in each step to prevent bias and improve internal validity. Two authors (A.D., E.R.) reviewed titles and abstracts and 2 authors (A.W., J.T.M.) reviewed the full texts that were selected from the abstract review. All studies meeting the inclusion criteria were selected for the review. In situations where disagreement concerning whether to include an article occurred, the authors discussed the disagreements and came to a consensus. If consensus was not reached between the 2 reviewing authors, a third author made the final decision.
Data and information were extracted from each study regarding study population, intervention techniques, and outcome measures. The following outcome measures were accepted on the basis of their reported reliability and validity: measures of balance including the Berg Balance Scale,26–28 Clinical Test of Sensory Integration and Balance (CTSIB),29 postural sway measured by the Sway Meter,30 single-leg stance,31 tandem stance,32 and coordinated stability test,33 measures of functional mobility including TUG/8 Foot Up and Go test34,35 and gait speed (preferred and fast),36 and measures of quality of life including 15D measure of health-related quality of life,37 36-item Short-Form Health Survey (SF-36)38 and Osteoporosis Assessment Questionnaire (OPAQ).39 Self-reported data of fear of falling, fall rate, and fall-related injuries were also accepted.
To help demonstrate the true magnitude of the measured intervention, effect sizes were calculated from the standardized mean change scores (the difference between the pretest and the posttest) when possible. If standardized mean change score was unavailable, effect size was calculated from the postintervention scores for all studies that provided sufficient data necessary for such calculation. Effect size is the standardized mean difference between 2 comparative groups that measure continuous data and is calculated by subtracting the mean of the intervention group by the mean of the control group and dividing by the within-group or pooled standard deviation of the 2 groups.40 In essence, effect size is a measure of the true treatment effect, regardless of statistical significance, and is less influenced by sample size.40 Authors of the published research studies were contacted if sufficient data were not available in the original published article. The following interpretation of effect size strength was used: less than 0.20 (trivial), between 0.20 and 0.50 (small), between 0.50 and 0.80 (medium), and greater than 0.80 (large).40 When effect sizes were unable to be calculated because the needed data could not be obtained, the measure of difference between groups as reported in the original research article was used to describe the difference between the control group and the intervention group.
For all studies where possible, statistical significance and the effect size permitted a qualitative determination of which treatment group benefited in the given study. If the effect size was trivial and/or there was no statistical significance, the determination was made that there was no difference between groups. Where effect sizes were small or greater, the group that displayed the benefit was identified.
After all articles were selected, the risk of bias for each study was assessed using the PEDro scale. The PEDro scale assesses 10 criteria relating to internal validity and statistical reporting of RCTs.41 Two authors (J.L.M., A.W.) independently assessed the quality of each article that met our inclusion criteria, and independently rated each study and compared results. The authors each used a standardized table outlining the 10 criteria of the PEDro scale to rate each article. If there were differences, the authors discussed these differences and reached a consensus. Previous systematic reviews of RCT have designated a score of 6 or greater on the PEDro scale to indicate a high-quality study.42,43 Based on the Oxford Center for evidence-based medicine, the highest level of evidence lies in the systematic review of RCT; only RCTs were included in the present systematic review.43 Kappa scores were calculated for each item and for the overall agreement between authors.
The systematic search through PubMed and CINAHL yielded 173 potential titles. Of these, 45 full-text articles were identified on the basis of inclusion criteria. A hand search was performed of the references of these articles, identifying 15 additional full-text articles following abstract review. Therefore, 60 full texts were assessed in detail against inclusion criteria, 10 of which met the inclusion criteria.36,44–52Figure 1 provides a detailed description of our study selection process. The kappa value, P value, and 95% confidence interval (CI) indicating agreement in title review resulted in 0.96 (P< .0; 95% CI = 0.85-0.98), abstract review 0.91 (P< .01; 95% CI = 0.73-0.91), and full-text review 0.63 (P< .01; 95% CI = 0.31-0.84). The characteristics of each of the 10 studies included in this review can be found in Tables 1 and 2 and are divided according to the type of control used.
The PEDro scores for each study are detailed in Table 3 and are as follows: 5 studies36,45,48,49,50 scored 7/10, 3 studies46,47,52 scored 6/10, and 2 studies44,52 scored a 5/10. As it is inherent to studies of this nature, all articles failed to meet PEDro items 5 and 6, which indicate blinding of all subjects and blinding of all therapists who administered the therapy, respectively. Of the individual items on the PEDro, criteria 1, 2, 8, and 9 had 100% agreement between reviewers. Items 5, 6, 7, and 11 had 90% agreement, item 9 had 60% agreement, and items 3 and 4 had 40 and 50% agreement, respectively.
Six studies compared group-based exercise with no-exercise intervention. Of these studies, 3 compared group-based exercise coupled with an education component to education alone,44,50,52 while 3 studies compared group-based exercise to a control group that received no exercise intervention.48,49,51 Three studies compared group-based exercise with a physical therapist–prescribed home exercise program (HEP).36,46,47 One study compared group-based exercise, vision correction, and home hazard modifications, combinations of each and a control group.45 There were no studies meeting the inclusion criteria that compared group-based exercise with outpatient PT interventions. For the purpose of this study, the 10 included articles were categorized into 2 main groups: (1) those who compared group-based exercise to a nonexercise control group44,45,48–52 and (2) those who compared group-based exercise to an HEP.36,46,47 The following results are broken down on the basis of these groups accordingly. The results and calculated effect sizes for all reported outcome measures are reported in Tables 4–6.
Reported Patient Outcomes
Four studies44,45,48,50 compared group-based exercise to a control group with no intervention and measured self-reported falls. According to the PEDro score, 4 of the studies were of high quality45,48–50 and one was of lower quality.44 All of these studies reported a significant decrease in the number of falls compared to the control group (P< .02; risk ratio 0.60-0.82),44,45 with effect sizes ranging from trivial to small (0.19-0.25).48,50 Only 1 study,47 considered a high-quality study according to the PEDro score, compared group-based exercise to a physical therapist–prescribed HEP and evaluated the effects on fall rate. This study found no significant decrease in fall rate in either group (P = .87). The effect size could not be calculated.
Two studies that compared group exercise to a control44,50 reported on injury rate associated with falls. Means et al50 reported a significant difference in the injury rates associated with accidental falls, with a trivial effect size favoring the intervention group (0.17), and Barnett et al44 demonstrated a protective effect of the intervention group, which was not significant (0.66, 95% CI: 0.38-1.15).
Four studies compared group-based exercise to a control group with no intervention.44,45,48,51 According to the PEDro score, 2 of these studies were of high quality45,48 and 2 were of lower quality.44,51 One of these studies used the Berg Balance Test (BBT) as an outcome measure for balance.48 This study found significant improvement on the BBT score in the intervention group compared with the control group (P< .01) with a small effect size (0.49).48 This same study also used the CTSIB as an outcome measure for balance48 and found significant differences in the number of patients with improvement (P ≤ 0.001) in comparison to the control group for conditions 5 (eyes closed, on stable surface) and 6 (eyes open, visual conflict, unstable surface).48 Both Barnett et al44 and Day et al45 assessed balance via the coordinated stability test. Both studies found that the exercise group performed significantly better than the control group in the coordinated stability test (P< .01). Effect sizes were not able to be calculated. In addition, Barnett et al44 tested postural sway and found an improvement in the intervention group over the control for all conditions (P values ranged from <.01 to .05) with an effect size ranging from trivial (0.17) to small (0.26). Sung51 used the single-leg stance as an outcome measure for balance and found a significant improvement in the exercise group compared to the control (P = .02). Effect size was found to be small (0.32).51
One study compared group-based exercise to a physical therapist–prescribed HEP and was rated as high-quality on the PEDro scale.46 This study used the BBT, single-leg stance, and tandem stance as measures of balance. This study reported significant differences within both groups for the BBT, single-leg stance, and the tandem stance between baseline and after 8 weeks of intervention. However, this study did not provide the measures necessary to calculate significance or effect size to determine between-group differences.46
Of the studies that compared group-based exercise to a control group with no intervention, 3 assessed physical performance parameters.44,45,48 According to the PEDro score, 2 of these studies were of high quality45,48 while one was of lower quality.44 Madureira et al48 utilized the TUG and observed a significant difference (P< .01) favoring the intervention group over the control with a trivial effect size (0.12). Barnett et al44 measured gait speed and found no differences between groups (P = 1.00), with no effect (0.00).
Three studies36,46,47 compared group-based exercise to a physical therapist–prescribed HEP and were rated as high-quality on the PEDro scale. Two of these studies assessed functional mobility via the TUG46,47 and one measured gait speed.36 All 3 studies36,46,47 observed significant improvements in both the intervention and control groups (P< .03); however, none of the studies demonstrated significant differences between the 2 groups (P> .05). The effect size was calculated for 2 studies36,47 and was found to be trivial (0.00-0.10) but was unable to be calculated for the third study.46
Health-Related Quality of Life
Three studies44,48,52 compared group-based exercise to a control group with no intervention and evaluated the effects on quality of life. According to the PEDro score, 2 of these studies were of high quality,48,52 while one was of lower quality.44 Vaapio et al52 found significant differences in changes between the intervention and control groups for men in depression (P = .01), distress (P = .02), usual activities (P = .05), and sexual activity (P = .05). For women, significant changes were found between groups for usual activities (P = .00) and discomfort/symptoms (P = .04) as measured on the 15D Instrument.37 The significance was in favor of the intervention group; however, effect size could not be calculated. Madureira et al48 measured quality of life specific to osteoporosis, using the OPAQ. They found significant score differences (P< 0.01) in all domains (well-being, physical function, psychological status, symptoms, and social interaction) in favor of the intervention group, with effect sizes ranging from small (0.46) to moderate (0.76).48 However, Barnett et al44 found no statistical significance between groups for any SF-36 categories and the effect sizes were trivial (0.01-0.09).
Only Helbostad et al36 compared group-based exercise with a physical therapist–prescribed HEP that evaluated quality of life. According to the PEDro score, this study was rated as high-quality. This study assessed quality of life using the SF-36 and found significant improvements in the role emotional category favoring the intervention group (P< 0.01) with a trivial effect size (0.17 at 3 months). A nonsignificant finding was demonstrated in favor of the control group at 9 months postintervention with a small effect size (−0.20) for the role physical. All other findings did not demonstrate benefit for one group over the other.
Fear of Falling
Two studies44,49 compared group-based exercise to a control group with no interventions. According to the PEDro score, one of these studies was of high quality,49 while one was of low quality.44 Madureira et al49 reported a significant decrease (P< 0.01) in fear of falls as a psychological component of the OPAQ in the intervention group compared to the control group, with a moderate effect size (0.65), while Barnett et al44 demonstrated a decrease in the number of participants reporting a fear of falling in both groups, which did not achieve significance. Donat and Ozcan46 compared group-based exercise with a physical therapist–prescribed HEP and also found no significant difference between the 2 groups (P> .05) in regard to fear of falling. Effect size was unable to be calculated.
This systematic review sought to synthesize the evidence from randomized controlled or clinical trials that compared the effectiveness of falls prevention in physical therapist–administered group-based exercise to traditional individual PT interventions, based on patient outcomes for older adults. Ten studies that evaluated the effects of group-based exercise on functional balance and/or quality of life in older adults from December 2001 to June 2012 were included in this review. Of these, 8 studies were of high methodological quality based on parameters set forth by Maher et al.42
The results of the studies that compared group-based exercise to a nonexercise control group found significant benefits in the exercise group in the following outcome categories: fall rate,44,45,48,50 functional mobility,48 balance,44,45,48,49,52 health-related quality of life,49,52 and fear of falling.49 One of these studies evaluated fear of falling and found no difference between groups.44 These results provide preliminary support for the use of group-based therapy as an effective method to reduce the risk of falls and improve balance and quality of life in the older adults as opposed to no intervention. There were 2 outcomes that demonstrated contradicting results, postural sway44,45 and TUG.45,48 In regard to postural sway, Barnett et al44 found improvement in the intervention group, while Day et al 45 did not. These differences could be contributed to the specific exercises provided, or the dosage. The exercise group in the study by Barnett et al44 met for 37 one-hour sessions throughout the span of 1 year, while the exercise group in the study by Day et al45 met only for 1-hour sessions one time per week for 15 weeks. In addition, it appears that the exercises performed in the study by Barnett et al44 were of higher intensity, although this is not explicitly described in either study. Both studies were supplemented by a daily HEP. In regard to the contradictory findings for the TUG, Day et al45 found no improvement in the exercise group, while Madureira et al48 did. The difference in results could again be contributed to the simple variance in dosage. The exercise group in the study by Madureira et al45 met one time per week for 1-hour sessions for a total of 40 weeks, while Day et al45 had the same parameters but met only for 15 weeks. This demonstrates the importance of a specified dosage for meeting the minimal threshold for change, as discussed in a meta-analysis by Sherrington et al.2 As previously mentioned, the authors of this study concluded that exercise specific to falls prevention should be performed 2 hours per week lasting at least 6 months for effective falls prevention.2
The results of studies that compared group-based exercise to an HEP found either improvement or no between-group differences for functional mobility,36,47 balance,46 and quality of life.36 Only 1 of these studies assessed fall rate and found no improvement.47 The lack of significant improvement in fall rate may again be secondary to the dosage and/or specific exercises performed. The authors of this study also discuss the possibility that the study population already had a low percentage of falls; therefore, showing improvement that is statistically significant is difficult.47 The results of the 3 studies that compared group-based exercise to an HEP are significantly less positive than those that compared group-based exercise to a control group. This may be because the group-based exercise class in 2 studies was performed 2 times per week in addition to the HEP.36,47 The authors suggest potential diminishing returns to explain why the group-based exercise group did not see further improvement over the HEP group. The third study that compared group-based exercise to an HEP46 prescribed identical interventions and dosages for both groups. In addition, the HEP group met with a physical therapist every 2 weeks to progress exercises. This lacks external validity, as a patient who is given an HEP typically does not have twice-monthly checkups with his or her physical therapist. It is not a surprise that the authors found no between-group differences for any of their outcome measures. Overall, the authors of these 3 studies36,46,47 found improvements in multiple balance measures in both groups, even though there were very few between-group differences, demonstrating the effectiveness of the specific exercise program itself, which may be used to guide future home-based and group-based exercise programs targeting this population.
In regard to the specific balance outcomes, significant improvements were found in many valid and reliable outcome measures, including the BBT,46, 49 CTSIB,48 coordinated stability test,44,45 single-leg and tandem stance,46,51 and postural sway;44 improvements for group-based exercise were comparable to HEP results and significantly greater than control or educational interventions for many of these measures. Functional mobility gains via the TUG47,48 and preferred and fast gait speed36 were found to be significant and comparable to HEP groups in 3 high-quality studies. As a whole, the included studies provide preliminary evidence supporting group-based exercise as an effective alternative to home-based exercise programs, with similar results in terms of physical performance and balance outcomes among older adults.
Multiple quality-of-life measures were also used in a number of included studies to evaluate the psychological and psychosocial effects that group-based exercise is purported to elicit.15 These studies demonstrated an improvement in overall feelings of well-being, including reduced feelings of isolation,49 increased self-esteem,51 improved mental health36 and depression,52 and decreased distress.52 The included studies provided conflicting evidence on whether group-based exercise can result in a decreased fear of falling, with only 1 study49 reporting an improvement as a result of the group-based exercise intervention. Two other studies44,46 assessed this outcome, neither of which found any significant improvement. This correlates with the results of a study by Liu-Ambrose et al,15 which concluded that fear of falling does not necessarily correlate to changes in fall risk factors and physical abilities.
A handful of the selected articles also discussed group-based exercise versus home-based exercise in terms of adherence and satisfaction. Helbostad et al36 reported higher satisfaction in the group program compared to the HEP. Madureira et al49 noted that higher adherence rates are typically achieved when patients belong to a social group with similar characteristics, and that social interaction promotes adherence to both group-based and HEPs. Lord et al53 further reiterated this stance by suggesting that group activities may facilitate long-term compliance to exercise programs, while also increasing enjoyment and social interaction. In contrast, Donat and Ozcan46 discussed the possibility that HEPs give participants the power to self-select the time and days they perform the exercises based on their personal schedule, which may allow for a higher adherence rate compared to group-based programs. Proposed factors that may limit participation in a group exercise class include physical limitations, access to transportation, and personal time schedule conflicts.46
Six of the included studies supplemented the group exercise class with a home program.36,44,45,49,50,52 All of these studies demonstrated positive results, indicating a possible benefit that may be attributable to the increased dosage associated with the addition of an HEP. Supplementation with an HEP provides an externally valid means to maintain the benefits of exercise and continue exercise as a long-term activity, which has been shown to have the most beneficial effects of falls prevention.2 It has also been shown that a home program can improve adherence to the group-based exercise class.49
This systematic review is limited by a search strategy that was restricted to articles published in English from December 2001 to June 2012. One full-text article could not be obtained and was not reviewed. In addition, statistical analysis was not possible because of the heterogeneity of outcome measures used among the included studies. The wide variety of methods used to assess balance, functional mobility, and quality-of-life concepts in the included studies made comparison across studies difficult. If the same outcome measures were utilized by researchers, it would allow for easier comparison of results between studies and further meta-analyses that would result in stronger, more conclusive findings.
The terms associated with group-based exercise in the literature are also extremely varied and difficult to collect using a standardized search strategy, thereby limiting the applicability and numbers of articles returned in a database search. A standardized term such as “group-based exercise” is recommended for use in future research to avoid confusion and facilitate the systematic sharing of accurate information.
Based on the information from current literature, there is evidence to suggest that group-based exercise can be utilized as a valid option for falls prevention in older adults with comparable results to a physical therapist–prescribed HEP, and with greater results than education alone or no intervention. Of course, clinical decision-making with evidence-based recommendations for designing an effective group-based exercise program may include the following: a similar group of individuals in terms of disease/impairment/age, an easily accessible setting, a physical therapist–developed program with a supplemental HEP, and a long-term or cyclic time frame to maintain benefits.
Additional high-quality evidence is needed to further investigate these findings, as well as to explore the benefits of group-based exercise compared to traditional outpatient PT intervention. In light of the recent shift toward preventative care, future research should also be performed to determine the feasibility and cost-effectiveness of implementing group-based exercise programs for at-risk older adults as opposed to traditional outpatient services, HEP, or educational interventions.
2. Sherrington C, Tiedemann A, Fairhall N, Close JCT, Lord SR. Exercise to prevent falls in older adults: an updated meta-analysis and best practice recommendations. NSW Public Health Bull. 2011;22(4):78–83.
3. Gillespie LD, Robertson MC, Gillespie WJ, Lamb SE, Gates S, Cumming RG, Rowe BH. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev. 2009;15(2):CD007146.
4. Hektoen LF, Aas E, Lurås H. Cost-effectiveness in fall prevention for older women. Scand J Public Health. 2009; 37(6):584–9.
5. Shubert T. Evidence-Based exercise prescription for balance
and prevention: a current review of the literature. J Geriatr Phys Ther. 2011;34(3):100–108.
8. Sherrington C, Whitney JC, Lord SR, Herbert RD, Cumming RG, Close JC. Effective exercise for the prevention of falls: a systematic review and meta-analysis. J Am Geriatr Soc. 2008;56(12):2234–2243.
9. Chang J, et al. Interventions for the prevention of falls in older adults: a systematic review and meta-analysis of randomised controlled trials. BMJ. 2004;328(7441):680.doi: 10.1136/bmj.328.7441.680.
10. Dean E. Physical therapy in the 21st
century (Part I): toward practice informed by epidemiology and the crisis of lifestyle conditions. Physiother Theory Pract. 2009;25(5-6):330–53.
11. Dean E. Physical therapy in the 21st
century (Part II): evidence-based practice within the context of evidence-informed practice. Physiotherapy Theory Pract. 2009;25(5–6):354–68.
12. Carter ND, Kannus P, Khan KM. Exercise in the prevention of falls in older people: a systematic literature review examining the rationale and the evidence. Sports Med. 2001;31(6):427–38.
13. Büla CJ, Monod S, Hoskovec C, Rochat S. Interventions aiming at balance
confidence improvements in older adults: an updated review. Gerontol. 2012;58(3):276–286.
14. Fairhall N, Sherrington C, Clemson L, Cameron ID. Do exercise interventions designed to prevent falls affect participation in life roles? A systematic review and meta-analysis. Age Ageing. 2011;40(6):666–674.
15. Liu-Ambrose T, Khan KM, Eng JJ, Lord SR, McKay HA. Balance
Confidence improves with resistance or agility training. Increase is not correlated with objective changes in fall risk and physical abilities. Gerontol. 2004;50(6):373–382.
16. Lord SR, Ward JA, Williams P, Strudwick M. The effect of a 12-month exercise trial on balance
, strength, and falls in older women: a randomized controlled trial. J Am Geriatr Soc. 1995;43(11):1198–206.
17. Phillips EM, Schneider JC, Mercer GR. Motivating elders to initiate and maintain exercise. Arch Phys Med Rehabil. 2004;85(7 Suppl 3):S52–S57.
18. Province MA, Hadley EC, Hornbrook MC, et al. The effects of exercise on falls in elderly
persons. A preplanned meta-analysis of the FICSIT Trials. Frailty Injuries: Cooperative Studies of Intervention Techniques. JAMA. 1995;273(17):1341–1347.
20. Bakker C, Hidding A, van der Linden S, van Doorslaer E. Cost effectiveness of group physical therapy compared to individualized therapy for ankylosing spondylitis. A randomized controlled trial. J Rheumatol. 1994;21(2):264–268.
21. Ackermann R, Cheadle A, Sandhu N, Madsen L, Wagner EH, LoGerfo JP. Community exercise program use and changes in healthcare costs for older adults. Am J Prev Med. 2003;25(3): 232–7.
22. Aprile I, et al. Group rehabilitation versus individual rehabilitation following knee and hip replacement: a pilot study with randomized, single-blind, cross-over design. Eur J Phys Rehabil Med. 2011;47(4):551–9.
23. English C, Hillier S. Circuit class therapy for improving mobility after stroke: a systematic review. J Rehabil Med. 2012;43(7):565–571.
24. Gardner MM, Robertson MC, Campbell AJ. Exercise in preventing falls and fall related injuries in older people: a review of randomized controlled trials. Br J Sports Med. 2010;34(1):7–17.
25. Moher D, Liberati A, Tetzlaff J, Altman DG. PRISMA group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339(b2535).
26. Berg K, Wood-Dauphine S, Williams J, Maki B. Measuring balance
in the elderly
: validation of an instrument. Can J Public Health. 1992;83(suppl 2):S7–S11.
27. Berg K, Wood-Dauphine S, Williams J. The Balance
Scale: reliability assessment with elderly
residents and patients with acute stroke. Scand J Rehabil Med. 1995;27(1):27–36.
28. Berg K, Maki B, Williams J, Holliday PJ, Wood-Dauphine S. Clinical and laboratory measures of postural balance
in an elderly
population. Arch Phys Med Rehabil. 1992;73(11):1073–80.
29. Whitney SL, Poole JL, Cass SP. A review of balance
instruments for older adults. Am J Occup Ther. 1998;52(8):666–671.
30. Sturnieks DL, Arnold R, Lord SR. Validity and reliability of the sway meter device for measuring postural sway. BMC Geri. 2011;11(63).
31. Flansbjer UB, Blom J, Brogardh C. The reproducibility of berg balance
scale and the single-leg stance in chronic stroke and the relationship between the two sets. PMR. 2012;4(3):165–70.
32. Murphy MA, Olson SL, Protas EJ, Overby AR. Screening for falls in community dwelling elderly
. J Aging Phys Activity. 2003;11:64–78.
33. Vogler CM, Sherrington C, Ogle SJ, Lord SR. Reducing risk of falling in older people discharged from hospital: a randomized controlled trial comparing seated exercises, weight bearing exercises and social visits. Arch Phys Med Rehab. 2009;90(8):1317–1324.
34. Hughes C, Osman C, Woods AK. Relationship among performance on stair ambulation, functional reach, and timed up and go tests in older adults. Issues Ageing. 1998;21:18–22.
35. Lin MR, Hwang HF, Hu MH, Wu HD, Wang YW, Huang FC. Psychometric comparisons of the timed up and go, one-leg stand, functional reach, and tinneti balance
measures in community-dwelling older people. J Am Geriatr Soc. 2004;52(8):1343–1348.
36. Helbostad JL, Sletvold O, Moe-Nilssen R. Home training with and without additional group training in physically frail old people living at home: effect on health-related quality of life and ambulation. Clin Rehab. 2004;18:498–509.
38. McDowell I, Newell C. Measuring health – A guide to rating scales and questionnaires, 2nd Ed. Oxford: Oxford University Press. 1996.
39. Randell AG, Bhalerao N, Nguyen TV, Sambrook PN, Eisman JA, Silverman SL. Quality of life in osteoporosis: reliability, consistency, and validity of the Osteoporosis Assessment Questionnaire. J Rheumatol. 1998;25(6):1171–1179.
40. Cook, Clinimetrics corner: effect sizes in describing data. J Man Manip Ther. 2006;16:E54–E57.
41. De Morton NA. The PEDro scale is a valid measure of the methodological quality of clinical trials: a demographic study. Aust J Physiother. 2009; 55(2):129–133.
42. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating the quality of randomized controlled trials. Phys Ther. 2008;82(8):712–721.
43. Sackett D. Evidence based medicine. Seminars in Perinatology. 1997;21(1):3–5.
44. Barnett A, Smith B, Lord S, Williams M, Baumand A. Community-based group exercise
and reduces falls in at-risk older people: a randomised controlled trial. Age Ageing. 2003;32(4):407–414.
45. Day L, Fildes B, Gordon I, Fitzharris M, Flamer H, Lord S. Randomised factorial trial of falls prevention among older people living in their own homes. BMJ. 2002; 325(7356):128–131.
46. Donat H, Ozcan A. Comparison of the effectiveness of two programmes on older adults at risk of falling: unsupervised home exercise and supervised group exercise
. Clin Rehab. 2007;21(3):273–283.
47. Helbostad JL, Sletvold O, Moe-Nilssen R. Effects of home exercises and group training on functional abilities in home-dwelling older persons with mobility and balance
problems. A randomized study. Aging Clin Exp Res. 2004;16(2):113–121.
48. Madureira MM, Takayam L, Gallinaro AL, Caparbo VF, Costa RA, Pereira RM. Balance
training program is highly effective in improving functional status and reducing the risk of falls in elderly
women with osteoporosis: a randomized controlled trial. Osteoporos Int. 2007;18(4):419–425.
49. Madureira MM, Bonfa E, Takayama L, Pereira RM. A 12-month randomized controlled trial of balance
training in elderly
women with osteoporosis: improvement of quality of life. Maturitas. 2010;66(2):206–211.
50. Means KM, Rodell DE, O'Sullivan PS. Balance
, mobility, and falls among community-dwelling elderly
persons: effects of a rehabilitation exercise program. Am J Phys Med Rehabil. 2005;84(4):238–250.
51. Sung K. The effects of 16-week group exercise
program on physical function and mental health of elderly
Korean women in long-term assisted living facility. J Cardiovasc Nurs. 2009;24(5):344–351.
52. Vaapio S, Salminen M, Vahlberg T, Sjosten N, Isoaho R, Aarnio P, Kivela SL. Effects of risk-based multifactorial fall prevention on health-related quality of life among the community-dwelling aged: a randomized controlled trial. Health Qual Life Outcomes. 2007;5(20). doi: 10.1186/1477-7525-5-20.
53. Lord SR, Castell S, Corcoran J, Dayhew J, et al. The effect of group exercise
on physical functioning and falls in frail older people living in retirement villages: a randomized controlled trial. J Am Ger Soc. 2003;51(12):1685–92.
Keywords:Copyright © 2013 the Section on Geriatrics of the American Physical Therapy Association
accidental falls; balance; elderly; group exercise