What type of exercises should older adults engage in to improve fall efficacy and physical fitness related to falling?

Shin, Sohee; Wuensche, Burkhard; Sohee Shin; Burkhard Wuensche

doi:10.12965/jer.2346276.138

J Exerc Rehabil > Volume 19(4);2023 > Article

Shin and Wuensche: What type of exercises should older adults engage in to improve fall efficacy and physical fitness related to falling?

Review Article

Journal of Exercise Rehabilitation 2023; 19(4): 198-207.

Published online: August 22, 2023

DOI: https://doi.org/10.12965/jer.2346276.138

What type of exercises should older adults engage in to improve fall efficacy and physical fitness related to falling?

Sohee Shin^1,^*, Burkhard Wuensche²

¹School of Exercise and Sport Science, University of Ulsan, Ulsan, Korea

²Department of Computer Science, University of Auckland, Auckland, New Zealand

^*Corresponding author: Sohee Shin, https://orcid.org/0000-0002-7801-1987, School of Exercise and Sport Science, University of Ulsan, 93 Daehak-ro, Nam-gu, Ulsan 44610, Korea, Email: soheeshin@ulsan.ac.kr

Received June 14, 2023 Accepted July 18, 2023

(open-access):

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

This study aimed to examine the effects of exercise interventions developed over the past 10 years to prevent falls among older adults. Cochrane, PubMed, and Embase databases were systematically searched on November 3, 2022, using the following keywords: “elderly” or “aged” and “fall prevention exercise” and “randomized controlled trial.” A total of 918 articles were retrieved, and finally, 18 studies were included in the meta-analysis. main conclusions were as follows: fall prevention exercises showed moderately positive effects on balance, gait, mobility, physical function, lower limb power, and strength, but low effects on proprioception, vision, and reaction speed. The effect sizes were highest when the intervention period was 1–24 weeks, time was 31–60 min, and frequency was thrice per week. Place of intervention (home, community, laboratory) and age (>75 years, <75 years) showed similarly moderate effect sizes. A combined program that includes balance, co-ordination, and resistance exercises is appropriate to improve fall-related fitness and fall efficacy in community-dwelling older individuals.

Keywords: Meta-analysis, Fall-related physical fitness, Community-dwelling older adults, Fall prevention, Exercise

INTRODUCTION

Fall is a serious health problem for older individuals because it is difficult to recover from physical damage or functional disability after a fall, the degree of injury is more severe, and may even cause death (Rubenstein, 2006). When older adults experience a fall, independent activities become difficult owing to their fear of falling, and their level of physical activity decreases. This leads to a vicious cycle that increases the risk of falling (Jefferis et al., 2014). In recent years, the risk of falls has emerged as an important healthcare problem in older adults because when the number of fall increases, medical costs increase. The importance of intervention programs for preventing falls for older individuals has been recognized, and various types of programs have been implemented. Studies on this are conducted widely because earlier, it was believed that falls were unavoidable as they occurred without a specific cause; however, recently, the perception has changed that falls are preventable by removing or alleviating their risk factors (Rubenstein and Josephson, 2006).

Various intrinsic and extrinsic factors affect the risk of fall, and multiple fall prevention programs have been designed to reduce this risk. Several systematic literature reviews related to fall prevention programs that support the effectiveness of multifactorial interventions in preventing falls among older adults have been published. These interventions included exercise, education, environmental modification, medication, mobility aids, vision, and psychological management (Lee and Yu, 2020). Chang et al. (2004) reported three findings through a meta-analysis: (a) multifactorial interventions significantly reduced fall rates in both high-risk and healthy groups compared with usual care; (b) exercise and environmental modifications are the key components of effective multifactorial fall prevention interventions; and (c) key intervention components focusing on exercise and environmental modification are recommended among the multifactorial fall prevention interventions. However, exercise is reported as a representative factor in lowering the risk of falls in older adults in the meta-analysis; in contrast, there are few studies that reviewed the literature only for exercise programs. In addition, exercise guidance is provided to older individuals for fall prevention, and there are studies that statistically reviewed whether such exercises have a positive effect on the improvement of fall-related physical fitness and physical function. Recently, many exercise programs for preventing falls have been developed, and it is important to review their effects. The novelty of this study is to comprehensively review the existing literature and suggest of the type and the way (frequency, place, time) of exercise to increase fall efficacy and fall-related physical fitness in the older adults.

Therefore, the purpose of this study was to conduct a meta-analysis of fall prevention exercise interventions developed over the past 10 years, and to find out which exercise interventions are effective for fall-related physical fitness and fall efficacy in community-dwelling older individuals.

MATERIALS AND METHODS

Reporting standards

The review was conducted in accordance with the PRISMA (preferred reporting items for systematic reviews and meta-analysis) guidelines (Moher et al., 2009) and the recommendations of the Cochrane collaboration (Higgins and Thompson, 2002).

Search strategy

Search criteria were established to systematically select articles based on the population intervention comparison outcomes framework. First, study participants were community-dwelling older individuals aged over 60 years (Population). Second, the intervention method was limited to fall prevention exercises (Intervention). Third, research with an experimental design in the form of a comparison of results through pre- and post-tests were selected (Comparison). Fourth, parameters related to physical fitness and fall efficacy were selected as a dependent variable of the intervention (Outcome). Studies presenting specific statistical values for effect size (ES) calculation were selected. To identify the literature, Cochrane, Embase, and PubMed databases were search on November 3, 2022, using the following keywords: “elderly” or “aged” and “fall prevention exercise” and “randomized controlled trial (RCT).” A total of 918 papers were retrieved. Among the extracted papers, 283 duplicates and 396 systematic reviews, reports without text, and reports published before 2012 were removed. Next, 125 irrelevant articles, 56 unclear, and 40 in which the outcomes or contents were not related to the purpose of this study were removed. Finally, 18 studies were analyzed (Fig. 1).

Data extraction and coding

Two independent researchers with doctoral degrees in health science-related fields extracted the following data from the selected RCTs: participant details (sample size, age, and sex), intervention characteristics (methods, frequency, etc.), outcomes of interest, and results. Coding was conducted after writing a book to analyze the characteristics of the literature. Coding items were author (year of publication), participants, number of cases in the group, age, intervention period, and outcome variables (statistical value for each physical fitness variable). Outcomes were classified in two steps. In step 1, the parameters were divided into factors and were categorized as follows: balance, mobility, cognition, depression, fall efficacy scale (FES), gait, power reaction time, range of motion (ROM), strength, vision, and proprioception. In step 2, parameters with similar results were categorized as same factors to minimize the amount of information lost when analyzing each parameter.

Analysis methods

The coded data were analyzed using Comprehensive Meta-Analysis 3.0. First, the effect of each outcome parameter was calculated as Cohen d. ES was categorized as follows: <0.40, small; 0.40–0.70, moderate; >0.70, large. Second, Q and I² were calculated to verify the heterogeneity of the ES. Cochran Q test is a traditional test for heterogeneity in meta-analyses. It is based on a chi-squared distribution; large produces a probability that exhibits greater variability across studies, rather than within participants of a study. If the P-value for Cochran Q was <0.1, it was possible to determine whether there was heterogeneity in the ES. Mathematically, I² is expressed as I²=τ2/(σ2+τ2), where τ2 denotes the between-trial heterogeneity, σ2 denotes some common sampling error across trials, and σ2+τ2 is the total variation in the meta-analysis. The heterogeneity of I² was interpreted as follows: 0%–25%, low; 25%–49%, medium; 50%–74%, high; and 75%–100%, quite high (Higgins and Thompson, 2002). Fixed-effects or random-effects model was used if homogeneity or heterogeneity was determined, respectively. A quality assessment in this study was also independently conducted by two researchers using the Cochrane risk of bias for RCTs, which assessed selection, allocation, performance, detection, attrition, and reporting biases; the scores were categorized as low-risk, unclear, and high-risk. Finally, publication bias of the results was verified using Funnel plot and Egger regression analysis.

RESULTS

Homogeneity test and total ES

The results of the homogeneity test related to the effect of the fall intervention exercises on the parameters are shown in Table 1. The statistical significance of the Q value was less than 0.000, and I² was 79.98; therefore, the heterogeneity was greater than moderate. Consequently, it was assumed that the articles included in this study were not homogeneous, as analyzed using the random-effects model. According to Cohen d coefficient, the ES was moderate (0.54).

Risk of bias

A summary of the quality assessment results is presented in Fig. 2. The quality assessment revealed that three studies had a high risk of performance bias, and that several studies reported unclear blinding of participants and personnel and selective reporting.

ES for each physical fitness factor and FES

Table 2 shows the effect of fall prevention exercises on FES and physical fitness, and the results of the subgroup analysis of participants’ age, intervention period, frequency, time, and place. First, the ES of the FES, an index representing fall efficacy, was 0.64, which was statistically significant. Regarding physical fitness variables, mobility (ES=0.74) and gait variables (ES=0.66) showed high ES of 0.7; balance (ES=0.58), center of pressure (ES=0.61), and physical function (ES=0.61) had moderate ES; and lower limb muscle power (ES=0.65) and muscle strength (ES=0.68) had high ES. Upper limb strength (ES=0.52) and upper limb power (ES= 0.24) showed medium to small effects. Awareness of mobility (ES= 0.45) and physical function (ES=0.38) showed moderate effects. In contrast, cognition (ES=0.27), reaction time (ES=0.29), ROM (ES=0.34), depression (ES=0.24), vision (ES=0.23), and proprioception (ES=0.30) had small effects.

ES for each age, intervention period, frequency, and time parameter

As a result of the subgroup analysis by classifying the age of participants in the intervention program based on the age of 75 years, both groups of <75 and ≥75 years showed moderate ES of 0.6 (ES=0.57, 0.59). According to intervention period, the groups with 1–12 and 13–24 weeks’ period showed moderate ES of 0.6 (ES=0.59, 0.55); however, the group with 25–48 weeks showed a small ES of 0.3. For the intervention time, the ES of 31–60 min was the largest (0.69), followed by 61–90 min (0.47), and ≤30 min (0.28). Regarding the frequency of interventions, the ES of thrice per week was the largest (0.69). Subgroup analysis by intervention location showed that community (ES=0.56), laboratory (ES=0.53), and home (ES=0.56) had moderate ESs. Characteristics of included studies were shown in the Table 3.

Publication bias assessment

Fig. 3 shows the results of Funnel plot and Egger regression analyses of the publication bias of the meta-analysis. Egger’s regression intercept significance test was not statistically significant (intercept=−0.63, standard error=1.04; P=0.55). Therefore, there was no significant publication bias.

DISCUSSION

Fall risk factors can be divided into two categories: (a) extrinsic such as lighting intensity, height, slope of indoor and outdoor floors, aspects related to the surrounding environment such as bathroom floors; and (b) intrinsic such as physical factors of muscle weakness, gait disorders, and drug intake (Hill-Westmoreland et al., 2002). Falls occur or recur because of a lack of mobility or activity due to aging, such as lower- and upper limb muscle weakness, balance and gait disorders, visual acuity loss, and sensory deterioration. If these risk factors are reduced in advance, falls can be prevented. This meta-analysis examined the effects of fall prevention exercise programs conducted over the last 10 years on fall-related physical fitness and their fall efficacy.

It was found that exercise interventions had positive effects on FES. When older adults experience a fall, their physical activity decreases because of their fear of falling, making it difficult for them to live independently. In contrast, people who are capable of physical activity and have a high level of fall-related physical fitness have high “fall efficacy,” which means that they are not afraid of falling.

In older individuals who participated in the fall prevention exercise programs, the amount of physical activity and level of fall-}related physical fitness increased by actively performing the exercises, leading to an increase in fall efficacy. People with high fall efficacy experienced fewer falls than others. However, Merom et al. (2016) reported that because the participants gained confidence due to the intervention, their exposure to walking increased and, consequently, they experienced more falls. In other words, regular exercise reduces the risk of falling, but frequent physical activity can increase the risk of falls (Hwang et al., 2016). It is necessary to review follow-up studies on the long-term effects of increased physical activity through various interventions on fall risk.

Overall, balance, gait, mobility, physical function, lower limb power, and strength had moderate or high positive effects; however, proprioception, vision, and reaction speed had no significant effects. A meta-analysis of various fall prevention exercise programs showed an overall moderate ES of 0.54 (P<0.05). Various intervention programs have been implemented; therefore, integrating intervention programs into a few factors is challenging. Among them, the intervention program which applied an integrated exercise centered on balance, coordination, and resistance exercises had a significant positive effect on fall-related physical fitness (ES: combined exercise, 0.76; visual feedback of movement performance, 0.66; task-oriented training, 1.04; motor imagery training+task- oriented training, 1.31; proprioceptive neuromuscular facilitation, 0.68; resistance exercise, 0.91; and fall prevention exercise based on the Otago program, 1.12; Wii, 0.79). In contrast, the impact of exercise intervention programs including cognition and dance on fall-related physical fitness was not significant (ES: dual-task training, 0.36; cognitive training, 0.22; social dancing, 0.23; trunk motion visual feedback, 0.09). However, the number of interventions that included cognitive function or dance in this study was small, and most of the physical fitness items related to falls were related to balance and lower extremity muscle strength; therefore, the direct ES may be small. Many studies have reported that cognitive function significantly influences falls (Chua et al., 2019).

Falls during walking are the primary cause of accidental kinematic injuries in older individuals (Zhuang et al., 2014). Voukelatos et al. (2015) reported that a walking program delivered through mailed printed materials and phone calls did not reduce falls in older people but increased their walking behavior and physical activity levels. Simple walking can help increase physical activity; however, a specific exercise program needs to be prepared to prevent falls. According to Mesquita et al. (2015) exercises for fall prevention should provide a moderate or high challenge to balance. Moreover, this study found that a combined exercise centered on balance, resistance, and coordination was effective in improving fall-related physical fitness. Strength gains and balance improvements were associated with positive changes in spatiotemporal measures and joint kinematics, and participants who could generate more muscle force to achieve a quicker push-off speed during the final instants of the stance phase were also able to move their center of gravity further away from their base of support (Zhuang et al., 2014).

The ES were highest when the intervention periods were 1–12 weeks and 13–24 weeks, intervention time was 31–60 min, and frequency was thrice per week. Even in the same older individual, physical function, presence or absence of diseases, physical fitness level, and fall risk level are different, and the necessary conditions or programs should differ according to the characteristics of the group. Social dance programs and Tai Chi are unsuitable for seniors at a high risk of falling (Merom et al., 2016). Several exercise interventions delivered to high-risk groups have also been reported to cause excessive falls in the intervention group (Ebrahim et al., 1997; Merom et al., 2016; Sherrington et al., 2014). In other words, the results of this subgroup analysis correspond to community-dwelling older population, and individualized exercise programs should be prepared for those who are frail or have specific diseases.

Place of intervention and age showed similar moderate ES. In other words, there was a similar positive effect regardless of the place of intervention in the community, home, or laboratory and age (under or over 75 years) in community-dwelling older people. The development of a sustainable, engaging, and motivating program that can prevent falls in the long-term is important, although the implementation of a temporary fall prevention program is also significant. To date, most exercise intervention programs have been community-centered, and it is clear that they are effective in preventing falls. However, the importance of home-based exercise rather than community-based ones has recently been emphasized owing to the coronavirus disease 2019 pandemic. Indeed, Schoene et al. (2013) reported that home-based exercise is effective in reducing falls. Evidence suggests that training at home increases the long-term adherence of older adults to physical activity (Schoene et al., 2013). Considering fall prevention from a motivational and long-term perspective, it is necessary to develop a program that appropriately combines the community and home.

CONCLUSION

This meta-analysis examined the effects of a fall prevention exercise program conducted over the last 10 years on fall-related fitness and their efficacy. Various exercise programs had moderate-to-positive effects on physical fitness (overall ES=0.54). The exercise interventions had a positive effect on FES. Overall, the programs had moderate or higher positive effects on balance, gait, mobility, physical function, lower limb power, and strength, but no significant effect on proprioception, vision, and reaction speed. The ES were highest when the intervention period was 1–12 weeks and 13–24 weeks, the time was 31–60 min, and the frequency was thrice a week. Place of intervention (home, community, laboratory) and age (over or under 75 years) showed similar moderate ES. A program that combines balance, coordination, and resistance exercises is appropriate for improving fall-related fitness and fall efficacy. In addition, fall prevention exercise programs should be developed according to the physical function, physical fitness, and fall risk of older adults. In the future, it will be necessary to examine how these fall prevention programs affect fall risk in the long-term through follow-up studies.

Notes

CONFLICT OF INTEREST

No potential conflict of interest relevant to this article was reported.

ACKNOWLEDGMENTS

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2022-0136).

REFERENCES

Anson E, Ma L, Meetam T, Thompson E, Rathore R, Dean V, Jeka J. Trunk motion visual feedback during walking improves dynamic balance in older adults: assessor blinded randomized controlled trial. Gait Posture. 2018;62:342–348.

Chang JT, Morton SC, Rubenstein LZ, Mojica WA, Maglione M, Suttorp MJ, Maglione M, Suttorp MJ, Roth EA, Shekelle PG. Interventions for the prevention of falls in older adults: systematic review and meta-analysis of randomised clinical trials. BMJ. 2004;328:680

Chua CHM, Jiang Y, Lim S, Wu VX, Wang W. Effectiveness of cognitive behaviour therapy-based multicomponent interventions on fear of falling among community-dwelling older adults: a systematic review and meta-analysis. J Adv Nurs. 2019;75:3299–3315.

Ebrahim S, Thompson PW, Baskaran V, Evans K. Randomized placebo-controlled trial of brisk walking in the prevention of postmenopausal osteoporosis. Age Ageing. 1997;26:253–260.

Eggenberger P, Theill N, Holenstein S, Schumacher V, de Bruin ED. Multicomponent physical exercise with simultaneous cognitive training to enhance dual-task walking of older adults: a secondary analysis of a 6-month randomized controlled trial with 1-year follow-up. Clin Interv Aging. 2015;10:1711–1732.

Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21:1539–1558.

Hill-Westmoreland EE, Soeken K, Spellbring AM. A meta-analysis of fall prevention programs for the elderly: how effective are they? Nurs Res. 2002;51:1–8.

Hwang HF, Chen SJ, Lee-Hsieh J, Chien DK, Chen CY, Lin MR. Effects of home-based Tai Chi and lower extremity training and self-practice on falls and functional outcomes in older fallers from the emergency department-a randomized controlled trial. J Am Geriatr Soc. 2016;64:518–525.

Jefferis BJ, Iliffe S, Kendrick D, Kerse N, Trost S, Lennon LT, Ash S, Sartini C, Morris RW, Wannamethee SG, Whincup PH. How are falls and fear of falling associated with objectively measured physical activity in a cohort of community-dwelling older men? BMC Geriatr. 2014;14:114

Kwok BC, Pua YH. Effects of WiiActive exercises on fear of falling and functional outcomes in community-dwelling older adults: a randomised control trial. Age Ageing. 2016;45:621–627.

Lee SH, Yu S. Effectiveness of multifactorial interventions in preventing falls among older adults in the community: a systematic review and meta-analysis. Int J Nurs Stud. 2020;106:103564

Li F, Harmer P, Fitzgerald K, Eckstrom E, Akers L, Chou LS, Pidgeon D, Voit J, Winters-Stone K. Effectiveness of a therapeutic Tai Ji Quan intervention vs a multimodal exercise intervention to prevent falls among older adults at high risk of falling: a randomized clinical trial. JAMA Intern Med. 2018;178:1301–1310.

Lytras D, Sykaras E, Iakovidis P, Komisopoulos C, Chasapis G, Mouratidou C. Effects of a modified Otago exercise program delivered through outpatient physical therapy to community-dwelling older adult fallers in Greece during the COVID-19 pandemic: a controlled, randomized, multicenter trial. Eur Geriatr Med. 2022;13:893–906.

Marques EA, Figueiredo P, Harris TB, Wanderley FA, Carvalho J. Are resistance and aerobic exercise training equally effective at improving knee muscle strength and balance in older women? Arch Gerontol Geriatr. 2017;68:106–112.

Merom D, Mathieu E, Cerin E, Morton RL, Simpson JM, Rissel C, Anstey KJ, Sherrington C, Lord SR, Cumming RG. Social dancing and incidence of falls in older adults: a cluster randomised controlled trial. PLoS Med. 2016;13:e1002112

Mesquita LS, de Carvalho FT, Freire LS, Neto OP, Zângaro RA. Effects of two exercise protocols on postural balance of elderly women: a randomized controlled trial. BMC Geriatr. 2015;15:61

Moher D, Liberati A, Tetzlaff J, Altman DG. Reprint--preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Phys Ther. 2009;89:873–880.

Oh DS, Choi JD. Effects of motor imagery training on balance and gait in older adults: a randomized controlled pilot study. Int J Environ Res Public Health. 2021;18:650

Page MJ, Moher D, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, Chou R, Glanville J, Grimshaw JM, Hróbjartsson A, Lalu MM, Li T, Loder EW, Mayo-Wilson E, McDonald S, McGuinness LA, Stewart LA, Thomas J, Tricco AC, Welch VA, Whiting P, McKenzie JE. PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ. 2021;372:n160

Reed-Jones RJ, Dorgo S, Hitchings MK, Bader JO. Vision and agility training in community dwelling older adults: incorporating visual training into programs for fall prevention. Gait Posture. 2012;35:585–589.

Rosado H, Bravo J, Raimundo A, Carvalho J, Marmeleira J, Pereira C. Effects of two 24-week multimodal exercise programs on reaction time, mobility, and dual-task performance in community-dwelling older adults at risk of falling: a randomized controlled trial. BMC Public Health. 2021;21:Suppl 2. 408

Rubenstein LZ. Falls in older people: epidemiology, risk factors and strategies for prevention. Age Ageing. 2006;35:Suppl 2. ii37–ii41.

Rubenstein LZ, Josephson KR. Falls and their prevention in elderly people: what does the evidence show? Med Clin North Am. 2006;90:807–824.

Schoene D, Lord SR, Delbaere K, Severino C, Davies TA, Smith ST. A randomized controlled pilot study of home-based step training in older people using videogame technology. PLoS One. 2013;8:e57734

Schwenk M, Grewal GS, Honarvar B, Schwenk S, Mohler J, Khalsa DS, Najafi B. Interactive balance training integrating sensor-based visual feedback of movement performance: a pilot study in older adults. J Neuroeng Rehabil. 2014;11:164

Sherrington C, Lord SR, Vogler CM, Close JC, Howard K, Dean CM, Heller GZ, Clemson L, O’Rourke SD, Ramsay E, Barraclough E, Herbert RD, Cumming R. A post-hospital home exercise program improved mobility but increased falls in older people: a randomised controlled trial. PLoS One. 2014;9:e104412

Smith-Ray RL, Makowski-Woidan B, Hughes SL. A randomized trial to measure the impact of a community-based cognitive training intervention on balance and gait in cognitively intact Black older adults. Health Educ Behav. 2014;41:1 Suppl. 62s–69s.

Tiedemann A, O’Rourke S, Sesto R, Sherrington C. A 12-week Iyengar yoga program improved balance and mobility in older community-dwelling people: a pilot randomized controlled trial. J Gerontol A Biol Sci Med Sci. 2013;68:1068–1075.

Voukelatos A, Merom D, Sherrington C, Rissel C, Cumming RG, Lord SR. The impact of a home-based walking programme on falls in older people: the Easy Steps randomised controlled trial. Age Ageing. 2015;44:377–383.

Wollesen B, Schulz S, Seydell L, Delbaere K. Does dual task training improve walking performance of older adults with concern of falling? BMC Geriatr. 2017;17:213

Zhuang J, Huang L, Wu Y, Zhang Y. The effectiveness of a combined exercise intervention on physical fitness factors related to falls in community-dwelling older adults. Clin Interv Aging. 2014;9:131–140.

Fig. 1

Elimination process of the articles selected for review (Page et al., 2021).

Fig. 2

Risk of bias summary and risk of bias graphy: a summary of the results of quality assessement.

Fig. 3

Funnel plot of standard error by standard differences in means.

Table 1

Meta-analysis results – overall effect sizes and heterogeneity

Model	Effect size and 95% CI					Test of null (2-tail)		Heterogeneity

	k	ES	SE	LL	UL	Z	P	Q	df	P	I²
Fixed	179	0.40	0.01	0.38	0.42	36.46	0.00	888.93	178.00	0.00	79.98

Random	179	0.54	0.03	0.49	0.59	19.47	0.00

CI, confidence interval; ES, effect size; SE, standard error; LL, lower limits; UL, upper limits; df, degrees of freedom.

Table 2

Subgroup analysis results - effect sizes and heterogeneity

Parameter	Effect size and 95% CI			Test of null (2-tail)

	Point estimate	Standard error	Variance	Z	P
Falls efficacy scale	0.64	0.22	0.05	2.92	0.00

Balance	0.58	0.10	0.01	5.69	0.00

Center of pressure	0.61	0.11	0.01	5.66	0.00

Gait	0.66	0.09	0.01	7.63	0.00

Mobility	0.74	0.10	0.01	7.72	0.00

Physical function	0.61	0.10	0.01	6.14	0.00

Lower power	0.65	0.14	0.02	4.51	0.00

Lower strength	0.68	0.09	0.01	7.79	0.00

Upper power	0.24	0.13	0.02	1.87	0.06

Upper strength	0.52	0.24	0.06	2.14	0.03

Cognition	0.27	0.07	0.01	3.78	0.00

Awareness of mobility	0.45	0.08	0.01	5.47	0.00

Awareness of physical function	0.38	0.09	0.01	4.45	0.00

Reaction time	0.29	0.05	0.00	5.36	0.00

Range of motion	0.34	0.13	0.02	2.71	0.01

Depression	0.24	0.04	0.00	5.52	0.00

Vision	0.23	0.26	0.07	0.90	0.37

Proprioception	0.30	0.07	0.00	4.41	0.00

Age of participants
≤74 years old	0.57	0.04	0.00	13.99	0.00
75 years old ~	0.59	0.04	0.00	13.31	0.00

Intervention period
1–12 weeks	0.59	0.04	0.00	15.94	0.00
13–24 weeks	0.55	0.05	0.00	10.31	0.00
25–48 weeks	0.30	0.05	0.00	6.04	0.00

Intervention time
≤30 min	0.28	0.06	0.00	4.78	0.00
31–60 min	0.59	0.03	0.00	17.71	0.00
61–90 min	0.47	0.05	0.00	9.30	0.00

Intervention times/wk
1	0.53	0.08	0.01	6.66	0.00
2	0.45	0.04	0.00	11.76	0.00
3	0.69	0.05	0.00	13.33	0.00
4	0.40	0.10	0.01	3.96	0.00

Intervention place
Community	0.56	0.03	0.00	16.23	0.00
Laboratory	0.53	0.06	0.00	8.30	0.00
Home	0.56	0.11	0.01	5.26	0.00

CI, confidence interval.

Table 3

Characteristics of included studies

Study	Region	No. (women)/age (yr)	Details of the intervention	Intervention component	Device	Place	Periods and frequency	Effect size
Anson et al. (2018)	USA	20 (13)/75.8±6.5	Trunk motion visual feedback Subjects walked on a treadmill approximately 24 inches in front of a 27” TV. The experimental group was instructed to keep the cursor in the center of the bullseye.	Balance, cognition	VFB device	Lab.	4 Weeks, 3 times, 30 min	0.09
Eggenberger et al. (2015)	Switzerland	20 (14)/77.3±6.3	DAnCe: virtual reality video game dancing The program DANCE included virtual reality video game dancing as a simultaneous cognitive-physical training. This training component combines an attention-demanding cognitive action with a simultaneous motor coordination aspect.	Balance, T, resistance, cognition	Virtual reality video game	Lab.	26 Weeks, 2 times, 60 min	0.43
Eggenberger et al. (2015)	Switzerland	22 (16)/78.5±5.1	Treadmill memory The program MEMORY comprised treadmill walking with verbal memory exercise as a simultaneous cognitive– physical training.	Balance, Resistance, Cognition	Treadmill	Lab.	26 Weeks, 2 times, 60 min	0.59
Hwang et al. (2016)	Taiwan	228 (153)/72±8.1	Home-based Tai Chi Yang-style TCC with 18 movements was taught individually each week for 24 consecutive weeks.	Taichi	No	Home	24 Weeks, 1 time, 60 min	0.19
Kwok and Pua (2016)	Singapore	40 (36)/70.5±6.7	Nintendo Wii Wii balance board and resistance band, which included cardiovascular training, resistance band strengthening, calisthenics and balance training.	Balance, Resistance, Endurance	Nintendo Wii	Lab.	12 Weeks, 1 time, 60 min	0.79
Li et al. (2018)	USA	224 (146)/77.5±5.6	Tai Ji Quan Moving for Better Balance, involved practice of a core of eight therapeutically modified exercise forms with built-in variations and a subroutine of integrated therapeutic movement exercises.	Taichi	No	Community	24 Weeks, 2 times, 60 min	0.64
Li et al. (2018)	USA	223 (143)/77.8±5.3	Multimodal The training protocol involved aerobic conditioning, strength, balance, and flexibility activities.	Balance, Resistance, Flexibility	No	Community	24 Weeks, 2 times, 60 min	0.61
Lytras et al. (2022)	Greece	75 (68)/70±67–74	Fall prevention exercise program based on the OEP The exercise program included five exercise groups: general warm-up exercises, lower limb muscle resistance exercises, exercises to improve dynamic and static balance, range of motion exercises and recovery exercises. All of the aforementioned exercises were performed according to the recommendations of the ACSM.	Otago	No	Hospital, Home	24 Weeks, 3 times, 45 min	1.12
Marques et al. (2017)	USA	23 (23)/67±5	Resistance exercise (RE) group The RE protocol aimed to develop muscle mass and strength in the following muscle groups: quadriceps, hamstrings, gluteal, trunk and arms, and abdominal muscles using variable resistance machines. The intensity of the training stimulus was initially set at 50% to 60% of one-repetition maximum (1RM), and then progressed to 80% of 1RM.	Resistance	No	Community	32 Weeks, 3 times, 60 min	0.91
Marques et al. (2017)	USA	24 (24)/70±5	Aerobic training The other exercise group participated in a training program design to improve aerobic capacity. The exercise intensity was initially set at 50% to 60% of the subject’s heart rate reserve for the first 2 months, and after increased to 85%.	Endurance	No	Community	32 Weeks, 3 times, 60 min	0.48
Merom et al. (2016)	Australia	279 (231)	Social dancing Participants in the 12 intervention villages were offered one of two major social dancing styles: Folk dancing (five villages), which included dances from the United Kingdom, United States, France, Italy, Israel, and Greece; or ballroom dancing (seven villages), which included dances such as Rock and Roll, Foxtrot, Waltz, Salsa, and Rumba.	Dance	No	Community	48 Weeks, 2 times, 60 min	0.23
Mesquita et al. (2015)	Brazil	20 (20)/68.5±5.4	PNF The PNF diagonal patterns of movement were selected considering all the basic facilitation procedures, including resistance, manual pressure, traction, stretch and approximation reflexes, and visual and verbal stimulation. The patterns were consistent with the three specific principles of PNF: rhythmic initiation, sustentation and relaxation, and reversal of antagonists.	PNF	No	Lab.	4 Weeks, 3 times, 50 min	0.68
Oh and Choi (2021)	Korea	11 (7)/79.9±5.6	MITG+Task-oriented training The subjects then performed motor imagery training for by freely imagining movements that protected their body and prevented injuries in the event of falls in diverse real-world environments, such as in the bathroom, kitchen, on the stairs, and around obstacles that are frequently associated with falls. And the task-oriented training implemented in this study entailed balance training centered on activities of daily living in the real-life environment of the senior citizen center.	Balance, Image training	No	Community	6 Weeks, 3 times, 40 min	1.31
Oh and Choi (2021)	Korea	11 (8)/78.7±2.62	TOTG The task-oriented training implemented in this study entailed balance training centered on activities of daily living in the real-life environment of the senior citizen center.	Balance	No	Community	6 Weeks, 3 times, 40 min	1.04
Reed-Jones et al. (2012)	USA	16/67.5±5.9	ACSM program+agility+visual training (visual group) was conducted using the Nintendo Wii Fit Plus with Wii Balance Board. The games played involved directing the body’s center of pressure while dodging objects presented by the video game. Video games were back projected onto a rear projection screen while participants stood one meter from the screen on the balance board. The visual angle of display was 60°, therefore objects appeared as if coming toward the participant from their center and moving to their periphery.	Balance, resistance, flexibility, endurance, agility, visual training (cognition)	No	Community	12 Weeks, 2 times, 90 min	0.42
Reed-Jones et al. (2012)	USA	16/67.5±5.9	ACSM program+agility (agility group)	Balance, Resistance, Flexibility, Endurance, Agility	No	Community	12 Weeks, 2 times, 90 min	0.60
Rosado et al. (2021)	Portugal	16 (15)/74.7±5.5	Multimodal exercises +WBV The psychomotor intervention program included exercises simultaneously promoting motor stimulation (e.g., agility, mobility, body awareness) and cognitive stimulation (e.g., problem-solving, cognitive inhibition, or RT training under single and DT conditions). The combined exercise program included the psychomotor intervention program+WBV program.	Balance, Agility, Cognition, Resistance	WBV	Lab.	24 Weeks, 3 times, 70 min	0.52
Rosado et al. (2021)	Portugal	16 (14)/74.3±5.4	Multimodal The psychomotor intervention program included exercises simultaneously promoting motor stimulation (e.g., agility, mobility, body awareness) and cognitive stimulation (e.g., problem-solving, cognitive inhibition, or RT training under single and DT conditions).	Balance, agility, cognition	No	Lab.	24 Weeks, 3 times, 70 min	0.40
Schoene et al. (2013)	Australia	15/77.5±4.5	DDR Intervention group participants were provided with a computerized step pad system connected to their TVs and played a step game.	Balance, agility, cognition	DDR	Home	8 Weeks, 2.5 times, 17.5 min	0.40
Schwenk et al. (2014)	USA	17 (10)/84.3±7.3	Visual feedback of movement performance Balance training including weight shifting and virtual obstacle crossing tasks with visual/auditory real-time joint movement feedback using wearable sensors.	Balance, Cognition	Interactive game-based virtual interface	Community	4 Weeks, 2 times, 45 min	0.66
Smith-Ray et al. (2014)	USA	23 (19)/73.26±7.15	Cognitive training intervention The Posit Science program carefully targets EF domains, such as visuospatial working memory, speed of processing, and inhibition, through three different games that are simple to learn and play: Road Tour, Jewel Diver, and Sweep Seeker. The Posit Science program is self-driven and adapts to the individual’s performance by increasing or decreasing task difficulty so that each participant continues to be challenged and engaged throughout the intervention.	Cognition	Computer based	Community, home	10 Weeks, 2 times, 60 min	0.22
Tiedemann et al. (2013)	Australia	27 (22)/67.7±7.2	Iyengar Yoga An experienced and Iyengar-certified yoga instructor conducted the classes, and in conjunction with the lead researcher, developed the 12-week program of yoga postures to be practiced and adhered to.	Balance, flexibility	No	Community, home	12 Weeks, 4 times, 45 min	0.40
Wollesen et al. (2017)	Germany	50/71	Dual-task training A progressive DT training including task-managing strategies was compared to a non-training control group.	Balance, agility, cognition	No	Community	12 Weeks, 1 time, 60 min	0.36
Zhuang et al. (2014)	China	22 (15)/66.3±4.89	Combined exercise intervention Each class started with a 5-min warm-up, followed by 15 min of balance exercises, 15 min of muscle-strength training, 15 min of 8-form Yang-style TCC, and ending with 10 min of flexibility/stretching and cool-down.	Balance, resistance, Tai Chi, flexibility	No	Community	12 Weeks, 3 times, 60 min	0.76

VFB, trunk motion visual feedback; TCC, Tai Chi Chuan; OEP, Otago exercise program; ACSM, American College of Sports Medicine; PNF, poprioceptive neuromuscular facilitation; MITG, motor imagery training; TOTG, task-oriented training; WBV, whole-body vibration program; RT, resistance training; DT, dual task; EF, executive function; DDR, exergame dance revolution.