Effects of elastic band training on body composition and physical performance in older people: A systematic review with meta-analysis

Objectives: This systematic review with meta-analysis aimed to evaluate the effects of elastic band training (EBT) on body composition and physical performance in apparently healthy older people


Introduction
A physically inactive lifestyle during aging has a negative effect on body composition and physical performance, impacting quality of life (Hernandez-Martinez et al., 2023).On the contrary, a physically active lifestyle is a protective factor against alterations in physical performance and body composition (Schilling et al., 2023), improving physical and mental well-being and the perception of quality of life (Valdés-Badilla et al., 2022).Considering that physical function deteriorates with age leading to a higher frequency of health conditions during aging, physical activity, physical exercise and a healthy lifestyle help to slow down this deterioration of physical function as well as medical conditions in older people (Agustí et al., 2023;Asim et al., 2021;Langhammer et al., 2018).
It is recommended to perform muscle strengthening, aerobic and balance exercises with a minimum duration of 150 min of moderate intensity physical activity or 75 min of vigorous intensity per week (Nikitas et al., 2022;World Health Organization, 2022).Leading a physically active lifestyle during old age can lead to significant improvements in postural balance (Thomas et al., 2019), muscle strength (Wickramarachchi et al., 2023), and significantly decreased fall risk and fractures (Thomas et al., 2019).A systematic review by Thomas et al. (2019) analyzed different physical activity strategies in apparently healthy older people, such as endurance training, balance training, pilates, and active exergames, and found significant improvements in all therapies on balance and frailty compared to inactive control groups.Indeed, there is a favorable change in fall prevention through the analysis of strength, balance, fall rate, fitness and/or fear of falling following muscle strength training (Rivera Miranda et al., 2024).A meta-analysis by Labata-Lezaun et al. (2023) on multicomponent training interventions (which involves at least three essential physical qualities or abilities, typically resistance, aerobic capacity, balance, and flexibility) in older people reported significant improvements in physical performance variables such as balance, walking speed, muscle strength of the lower and upper limbs and cardiorespiratory fitness compared to active/inactive control groups.However, a meta-analysis conducted by Stern et al. (2023) compared two types of interventions (high-intensity interval training vs. multicomponent training) and found significant improvements in physical performance in balance and cardiorespiratory fitness in favor of high-intensity interval training.A systematic review by Alzar-Teruel et al. (2022) reported significant improvements in timed up-and-go (TUG), gait speed, and a significant reduction in fat mass in favor of high-intensity interval training compared to active/inactive control groups.In another meta-analysis conducted by el Hadouchi et al. (2022) comparing the effect of strength training vs. power training on physical performance in apparently healthy older people, significant improvements were reported in upper and lower limbs muscle strength and postural balance in favor of power training.
Considering that alterations in body composition lead to decreased physical function in older people, it is important to assess these as a whole (Kim et al., 2017).Interventions using traditional training have positively affected body composition and physical performance (el Hadouchi et al., 2022;Labata-Lezaun et al., 2023;Stern et al., 2023).However, they are not easily accessible to older people and have a high economic cost when intervening in spaces such as gyms or laboratories (Falck et al., 2022).In this way, there are other alternatives as an intervention in older people, such as elastic band training (EBT), which is a low-cost material of great interest since elastic bands allow great adaptability and have proven to be helpful to achieve beneficial effects on body composition and physical performance in older people (Kim et al., 2021;Sanchez-Lastra et al., 2022b;Su et al., 2022).A metaanalysis by Tsai et al. (2022) in older people with sarcopenia reported significant improvements in TUG, maximal isometric handgrip strength, gait speed and appendicular skeletal muscle index in favor of EBT compared to active/inactive control groups.Similar to that reported by Oliveira et al. (2022), a meta-analysis in apparently healthy older people showed significant improvements in 30-second chair stand, TUG, maximal isometric handgrip strength and 6-minute walk test in favor of EBT in comparison to active/inactive control groups.Although there is evidence in systematic reviews and meta-analyses on the effects of EBT in older people, these only include body composition in sarcopenic and obese older people (Ghiotto et al., 2022;Oliveira et al., 2022;Tsai et al., 2022), and physical performance in apparently healthy, sarcopenic and sarcopenic obese older people (Ghiotto et al., 2022;Oliveira et al., 2022;Tsai et al., 2022).However, the effect of EBT on body composition and physical performance in apparently healthy older people has not been meta-analyzed so far in experimental design studies (randomized controlled trials and non-randomized controlled trials).It is likely that individuals with a medical condition may experience a greater margin to improve outcomes compared to already healthy individuals (Qiu et al., 2023).Thus, a systematic review/meta-analysis is required just in healthy older people to see how much improvement EBT can have in this specific cohort.Therefore, this systematic review with meta-analysis aimed to evaluate the available body of published peer-reviewed studies related to the effects of EBT on body composition and physical performance in apparently healthy older people compared to active/ inactive control groups.

Protocol and registration
The Preferred Reporting Items for Systematic Reviews and Metaanalyses (PRISMA) standards were adhered to in this systematic review.PROSPERO (the International Prospective Register of Systematic Reviews; ID code: CRD42024547050) has the protocol registered.

Eligibility criteria
Original peer-reviewed articles published up until May 2024 that were free of restrictions on language or publication date met the inclusion criteria for this systematic review.Conference abstracts, books and book chapters, editorials, letters to the editor, protocol records, reviews, case studies, and trials were among the documents that were not included.Furthermore, studies were included in a systematic review using the PICOS framework (population, intervention, comparator, outcome, and study design) (see Table 1).

Information and database search process
The search process was conducted between October 2023 and May 2024 using six generic databases: PubMed, ProQuest, EBSCOhost, CINAHL Complete, Scopus, and Web of Science (core collection).Medical Subject Headings (MeSH) from the National Library of Medicine of the United States of America used free language terms and hand search of reference lists was also conducted related to EBT, body composition, and physical performance in apparently healthy older people.The search string used was as follows: ("elastic band training" OR "resistance training" OR "training functional" OR "elastic band exercises" OR "resistance exercises" OR "functional exercises") AND ("body composition" OR "body fat" OR "fat-free mass" OR "fat mass" OR "muscle mass") Lack of baseline data and/or follow-ups.

Study design
Experimental design studies (randomized controlled and non-randomized controlled trials) with pre-and postassessments.
AND ("physical function" OR "physical performance" OR "functionality" OR "functional independence" OR "functional dependence" OR "functional mobility" OR "health condition" OR "falls" OR "fall risk" OR "risk of fall" OR "falling risk" "balance" OR "static balance" OR "dynamic balance" OR "walking speed" OR "gait speed" OR "mobility" OR "strength" OR "muscle strength" OR "upper body strength" OR "lower body strength" OR "muscle power") AND ("elderly" OR "older people" OR "older adults" OR "older subject" OR "aging" OR "ageing" OR "aged").Two independent experts were consulted over the included articles and the inclusion and exclusion criteria in order to help find more pertinent studies.We set two requirements that the experts needed to fulfill: (i) possess a PhD. in sports science; and (ii) have peer-reviewed works published in journals with an impact factor, according to Journal Citation Reports®, on physical performance in various population groups and/or physical performance.In order to prevent prejudice in their searches, our search approach was not disclosed to experts.After completing these procedures, on May 30, 2024, we conducted a database search to find pertinent retractions or errata pertaining to the papers that were included.

Studies selection and data collection process
The EndNote reference manager (version X9, Clarivate Analytics, Philadelphia, PA, USA) was used to export the studies.JHM and ICC conducted separate searches, eliminated duplicates, examined titles and abstracts, and examined complete texts.At this point, there were no disparities discovered.The procedure was carried out once again for recommendations made by outside specialists and searches inside reference lists.The entire texts of possibly suitable papers were then examined, and the rationale behind the exclusion of those that did not fit the selection criteria was disclosed.

Methodological quality assessment
TESTEX, a tool created especially for exercise-based intervention studies (Smart et al., 2015), was used to evaluate the methodological quality of the chosen studies.One potential exclusion criterion was TESTEX results (Smart et al., 2015).According to Smart et al. (2015), there is a 15-point rating system (5 points for study quality and 10 points for reporting), where studies that do not reach 60 % are excluded for further analyses in the present review (Smart et al., 2015).Two writers (ICC, THV) carried out this process separately, while a third author (JHM) served as a referee for cases that were on the borderline and needed further validation from another author (PVB) in cases where doubts still remained.

Data synthesis
The following data were obtained and analyzed from the selected studies: (i) author and year of publication; (ii) country of origin; (iii) study design; (iv) initial health status of the sample; (v) number of participants in the intervention and control groups; (vi) mean age of the sample; (vii) activities performed in the EBT and control groups; (viii) training volume (total duration, weekly frequency and time per session); (ix) training intensity; (x) data collection instruments on body composition and physical performance; (xi) main results of the studies before and after the interventions.Specific data were extracted for both body composition and physical performance in mean and standard deviation.

Risk of bias in individual studies
Two independent researchers (JHM and THV) evaluated the risk of bias version 2 (RoB 2) of the included studies, and a third researcher (PVB) analyzed the results.The Cochrane Handbook for Systematic Reviews of Interventions' recommendations for randomized control trials served as the foundation for this evaluation (Sterne et al., 2019).
Based on the randomization procedure, deviations from planned interventions, missing outcome data, outcome assessment, and choice of reported outcome, the risk of bias was classified as "high," "low," or "some concerns" for each study delivering the risk of bias for each study analyzed.Subsequently, the overall risk of bias in the sum of all the studies analyzed is reported by means of a graph in percentages for each procedure analyzed (Sterne et al., 2019).

Summary measures for meta-analysis
The study methodology includes meta-analyses; complete information is accessible at PROSPERO (registration code: CRD42024547050).The standardized mean difference (SMD), a standard statistic that assesses the absolute difference between the mean values in two groups in a randomized controlled trial and non-randomized controlled trial, were computed for each analysis using Comprehensive Meta-analysis Software (RevMan 5.4) where a value p < 0.05 is considered for statistical significance (Verhagen et al., 1998).In each trial, the random-effects model (Der Simonian-Laird approach) was used to calculate and pool the SMD and MD of the body composition and physical performance from preintervention to postintervention, comparing groups (EBT vs.Control groups) (Higgins et al., 2003).The fundamental premise of the random-effects model is that genuine effects (interventions, duration, etc.) vary throughout studies and that samples are selected from populations with varying effect sizes.If at least three studies showed the same results, the data were pooled (Davey et al., 2011).
Heterogeneity between trial results was tested with a Cochran's Q test (Morris, 2008) and I 2 statistic.I 2 values of <25 %, 25-50 %, and >50 % represent small, medium, and large amounts of inconsistency respectively (Higgins et al., 2003).Egger regression tests were performed to detect small study effects and possible publication bias (Higgins and Thompson, 2002).

Certainty of evidence
Studies were categorized as having high, moderate, low, or very low confidence of evidence based on their assessment of the evidence using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) scale (Guyatt et al., 2011).Because studies with experimental designs (randomized controlled trials and non-randomized controlled trials) were included, all analyses began with a high degree of certainty and were downgraded if there were concerns about the risk of bias, consistency, accuracy, precision, directness of results, or risk of publication bias (Guyatt et al., 2011).The studies were evaluated separately by two researchers (JHM, ICC), and any disagreements were settled by agreement with a third author (PVB).

Study selection
Fig. 1 details the search process for the studies.A total of 5916 records were found.Subsequently, duplicates were eliminated, and the studies were filtered by selecting the title, abstract, and keywords, resulting in 3604 references.In the subsequent analysis phase, 3256 articles were excluded because the texts did not meet the search criteria, leaving 348.Subsequently, 48 descriptive studies, 121 intervention studies were excluded in older people who were not apparently healthy, 58 in children or adults, and 110 review studies.After this process, 11 potential studies remained, of which 2 were excluded because they did not have a control group.Only 9 met all the selection criteria (Choi et al., 2020;Flández et al., 2021;Martins et al., 2015;Miranda-Aguilar et al., 2020;Oh et al., 2017;Pourtaghi et al., 2017;Sanchez-Lastra et al., 2022a;Souza et al., 2019;Yang et al., 2015).
Regarding the control groups, 3 studies presented active control groups (Miranda-Aguilar et al., 2020;Sanchez-Lastra et al., 2022a;Souza et al., 2019).In the study of Sanchez-Lastra et al. (2022a), a 12week duration of 3 series of 50 min of static stretching 4 exercises for the upper body and 4 exercises for the lower body with 2 series of 10 to 14 repetitions with a duration of 20 s per exercise at intentions of 7 to 10 (RPE) no pain.In the study of Souza et al. (2019), the control group used weight machines or free weights (Movement®, São Paulo, Brazil) to execute the exercises.In that order, the exercises were 2 sets of 10 to 14 repetitions of the barbell squat, seated rowing, and barbell bench press, with 60 s of rest between each set at an intensity of 8 to 10 (RPE).Three exercises were performed for the lower body (parallel squat 90 • , incline leg press, and gastrocnemius raise) and 3 exercises for the upper body (bench press, open pectoral, and lateral raises) with free weights and varied load machines with intensities ranging from 5 to 7 OMNI-RES scale (Miranda-Aguilar et al., 2020).In contrast, six studies (Choi et al., 2020;Flández et al., 2021;Martins et al., 2015;Oh et al., 2017;Pourtaghi et al., 2017;Yang et al., 2015) presented inactive control groups that did not receive intervention and continued with their activities of daily living.

Body composition
Regarding body composition assessments, significant decreases in total fat mass were observed in favor of EBT compared to the control groups in the individual results of the studies analyzed; however, a meta-   analysis was not possible because 2 studies reported by dual-energy Xray absorptiometry (DEXA) (Flández et al., 2021;Oh et al., 2017) and 1 by Skinfolds (Miranda-Aguilar et al., 2020), also 2 studies present the results in kg (Miranda-Aguilar et al., 2020;Oh et al., 2017) and 1 in percentage (Flández et al., 2021).There was no significant improvement in favor of EBT compared to active/inactive control groups in fat-free mass (SMD = 0.42; 95 % CI = − 0.14 to 0.99; I 2 = 99 %; p = 0.14), as presented in Fig. 4.

Arm curl test
As for muscle strength assessments, significant improvements were observed in arm curl in favor of EBT compared to the control groups; however, it was not possible to meta-analyze because only 1 study analyzed this variable.

Maximal isometric handgrip strength
There was no significant improvement in favor of EBT compared to the active/inactive control groups in maximal isometric handgrip strength (SMD = 0.65; 95 % CI = − 1.56 to 2.86; I 2 = 100 %; p = 0.57), as presented in Fig. 5.

30-second chair stand test
However, in the 30-second chair stand test, a significant increase was reported in favor of EBT compared to the active/inactive control groups (SMD = 3.03; 95 % CI = 0.14 to 5.93; I 2 = 100 %; p = 0.04), as presented in Fig. 6.

Back-scratch test
In upper body flexibility measured by the back-scratch test, no significant improvements were observed in favor of EBT in comparison with the active/inactive control groups (SMD = − 0.90; 95 % CI = − 3.12 to 1.31; I 2 = 99 %; p = 0.42); these results are presented in Fig. 7.

Sit-and-reach test
However, in lower body flexibility measured through the sit-andreach test, significant improvements were reported in favor of EBT compared to the active/inactive control groups (SMD = 2.09; 95 % CI = 0.15 to 4.03; I 2 = 100 %; p = 0.04) as shown in Fig. 8.

One-leg stance and static standing balance
In the balance assessments, no statistically significant differences were observed in the one-leg stance and static standing balance when comparing EBT vs. active/inactive control groups.It was impossible to perform a meta-analysis because only 1 study analyzed these variables.

Timed up-and-go test
However, in the TUG test, a significant decrease in performance time was observed in favor of EBT when compared to active/inactive control groups (SMD = 3.10; 95 % CI = 1.67 to 4.53; I 2 = 98 %; p < 0.0001), as shown in Fig. 9.

Certainty of evidence
The results obtained in the certainty of evidence did not allow definitive recommendations to be made in favor of EBT as an intervention to improve body composition and physical performance in apparently healthy older people, as it was very-low (Table 4).

Discussion
This systematic review with meta-analysis aimed to evaluate the available body of published peer-reviewed studies related to the effects of EBT on body composition and physical performance in apparently healthy older people compared to active/inactive control groups.After reviewing 5916 records, 9 randomized and non-randomized controlled trials involving 477 healthy older people were included.All studies achieved a score equal to or >60 % on the TESTEX methodological scale.In addition, the risk of bias was high in all studies.Six meta-analyses were performed showing significant improvements in 30-second chair stand, sit-and-reach and TUG tests.However, in maximal isometric handgrip strength, back-scratch test, and body composition (only fatfree mass), no significant improvements in favor of EBT were reported.

Body composition
No significant improvements were reported in favor of EBT in fat-free mass compared to the active/inactive control groups.However, individual results show a decrease in body fat percentage in favor of EBT compared to the control groups (Flández et al., 2021).This is similar to what was reported in a meta-analysis by Wu et al. (2023) where only a significant decrease in body fat percentage (WMD = − 2.31, 95 % CI: − 2.76 to − 1.87;I 2 = 27.5 %, Q = 27.59,df = 20, p < 0.001) in favor of multicomponent training compared to active/inactive control groups, finding no significant increase in fat-free mass (WMD = 0.44, 95 % CI: − 0.25 to 1.12; I 2 = 0.0 %, Q = 5.71, df = 15, p > 0.05).However, a meta-analysis by Liu and Lee (2024) reported a significant decrease in body fat percentage (CI: − 262.55-260.11,p < 0.001, I 2 = 100 %) and a significant increase in fat-free mass (p < 0.001, I 2 = 98 %, CI: 0.31-0.71) in favor of EBT compared to inactive control groups in older female with sarcopenic obesity and osteosarcopenia.The differences in the reported studies may be due to the different training volumes carried out in the interventions; it has been reported that higher training volumes and  repetitions close to muscle failure can maximize the increase in muscle mass (Schoenfeld and Grgic, 2017).In this sense, the present systematic review reported volumes of 1 to 4 sets with 10 to 20 repetitions with moderate to vigorous intensities of 5 to 14 on the OMNI-RES perceived exertion scale.Secondly, nutrient intake, specifically protein intake, is key in increasing muscle mass (Stokes et al., 2018).Previous studies have recommended a daily protein intake of ~1.6 g/kg/day and 2.2 g/ kg/day to promote muscle growth (Stokes et al., 2018).However, this variable was not controlled in the studies analyzed, which may explain the different findings in the fat-free mass variable.Thirdly, the different assessment instruments used in the studies may have led to different findings, for example in the meta-analysis by Liu and Lee (2024) the studies used bioelectrical impedance analysis (BIA) and DEXA for the assessment of body composition.Unlike our meta-analysis, where the studies used the DEXA method and 1 study used skinfolds (that could not be meta-analyzed).In this regard, Achamrah et al. (2018) reported a lack of concordance between BIA and DEXA methods at the individual level, regardless of body mass index.Specifically, their study found that in overweight and obese patients (25 kg/m 2 < body mass index < 40 kg/ m 2 ), BIA overestimated fat-free mass by 7.18 kg to 8.28 kg and underestimated body fat percentage by 4.02 kg to 5.67 kg compared to DEXA, which is in line with recent study reporting that BIA underestimated body fat percentage and overestimated fat-free mass in overweight and obese postpartum female compared to DEXA (Ellegård et al., 2016).

Muscle strength
In the maximal isometric handgrip strength test, no significant increases were reported in favor of EBT compared to the active/inactive control groups.This is similar to that reported by Mañas et al. (2021) in a meta-analysis of community-dwelling older people where no significant increase in maximal isometric handgrip strength (Hedges' g = − 0.00; 95 % CI = − 0.17, 0.16; p = 0.93) was reported in favor of unsupervised EBT compared to active/inactive control groups.In contrast, a meta-analysis by Tsai et al. (2022) in older people with sarcopenia reported significant increases in maximal isometric handgrip strength (SMD = 3.20; 95 % CI = 1.14 to 5.27; I 2 = 82 %; p = 0.002) in favor of EBT compared to active/ inactive control groups.Similar to that reported by Zhao et al. (2022) in a meta-analysis in older people with sarcopenia showing significant increases in maximal isometric handgrip strength (Hedges' g = 0.60, 95 % CI = 0.30-0.89;p < 0.05) in favor of EBT compared to inactive control groups.Fritz et al. (2015) reported improvements in maximum isometric strength for the upright row, squat, and back extension in older people through progressive overload with EBT, progressing from an intensity of 6 to 7 on the OMNI-RES scale in the first 3 weeks at an intensity of 8 to 9 on the OMNI-RES scale in the last 5 weeks.It is essential to consider that different intensities of elastic bands can lead to different gains in muscle strength (Lopes et al., 2019).During the beginning of a movement, the resistance generated by the elastic band is low, allowing a resisted execution at high speed; however, as the range of motion increases and depending on the intensity of the elastic band, the resistance considerably increases, forcing the subject to greater effort to maintain muscle power at higher overload ranges (Yang et al., 2015).Based on the findings of our meta-analysis, it is difficult to explain why there were no increases in maximal isometric handgrip strength through EBT.We agree that progressive overload is a key principle to generate muscular adaptations.However, we do not know if this was applied in four of the meta-analyzed studies since this was not reported and may have influenced the subjects' responses to EBT training.
However, in the lower body muscle strength using the 30-second chair stand test, significant improvements were reported in favor of EBT compared to the active/inactive control groups.Results similar to Table 4 GRADE assessment for the certainty of evidence.
those reported in a systematic review by Ghiotto et al. (2022), where specifically Liao et al. (2018) found that EBT led to an improvement in physical performance measured with the 30-second chair stand test in a sample of older people with sarcopenic obesity.It is important to mention that the comparison between different studies on the 30-second chair stand test is limited by the structural characteristics of the chair used (i.e., height, length, presence or absence of armrests) which are very often not reported.Shamay et al. (2014) reported that a lower seat height may make the task more difficult to perform, given the greater range of motion, while a higher seat height may decrease the amount of lower body work required.A meta-analysis by Labata-Lezaun et al. (2023) reported significant increases in 30-second chair stand (SMD = 1.14; 95 % CI = 0.43 to 1.85; I 2 = 86 %; p = 0.002) in favor of multicomponent training compared to active/inactive control groups in apparently healthy older people.Similar to that reported by Wu et al. (2021) in a meta-analysis of apparently healthy older people showing better performance in 30-second chair stand (WMD = − 3.86, 95 % CI − 5.11 to − 2.61; I 2 = 0.0 %, Q = 1.22,df = 3, p = 0.000) in favor of highintensity interval training compared to active/inactive control groups.Based on the studies, the improvements found through EBT can be attributed to increases in lower body muscle strength, specifically increased strength in the knee flexor and extensor musculature, although this was not measured directly in the studies (Brearley and Bishop, 2019;Nakatani et al., 2002).However, moderate correlations (r = 0.52) have been reported between performing the 30-second chair stand test and maximum voluntary isometric contraction in knee extension in older female (Nakatani et al., 2002).Given that these actions are the main ones during the execution of the 30-second chair stand test, the strength gains in the lower body can be seen specifically transferred to the performance of the test (Brearley and Bishop, 2019).Improvements through EBT could be explained by the qualities of training.Specifically, elastic bands allow for different degrees of tension to be applied throughout the range of motion (Galpin et al., 2015).In general, the resistance imposed by the elastic band is low at the beginning of the movement, allowing for high-speed resisted execution, improving the conduction velocity of motor units, which may be associated with a greater recruitment of fast motor units (Melchiorri and Rainoldi, 2011).When range of motion increases, resistance increases, forcing individuals to make an additional effort to maintain muscle power in higher overload ranges (Lopes et al., 2019).This could favor neurological adaptations that enhance the various expressions of muscle strength (Bellar et al., 2011).However, to improve maximal isometric handgrip strength through EBT based on the various studies analyzed, repeated stimulation of the hand and forearm muscles through exercises that demand repeated acts of gripping and pulling may be necessary (Rogers et al., 2002).

Flexibility
In upper body flexibility using the back-scratch test, no significant improvements were reported in favor of EBT compared to active/inactive control groups.These results are similar to that reported by Hai et al. (2022) in a meta-analysis in apparently healthy older people where no significant improvements in back-scratch (g = 0.13, 95 % CI = − 0.06 to 0.32; p > 0.05) were found in favor of active exergame interventions compared to active/inactive control groups.This is probably due to the different intervention protocols with EBT and the diverse selection of exercises, these did not influence the back-scratch test.Specific exercises through EBT for the external and internal rotators of the shoulder joint may be necessary (Kim et al., 2021) since it is well-documented that resistance training can improve range of motion through greater stretch tolerance, greater fascicle length, and changes in pennation angle (Afonso et al., 2021).It is likely that due to the different intervention protocols for the upper and lower body with EBT, these did not influence the back-scratch test, as well as the different intensities used in the interventions by increasing the tension of the elastic bands (Lopes et al., 2019).
However, in lower body flexibility measured by the sit-and-reach test, significant improvements were reported in favor of EBT compared to active/inactive control groups.Similar to that reported by Ko et al. (2023) in a meta-analysis conducted in apparently healthy older people showing significant improvements in sit-and-reach (SMD = 0.38, 95 % CI: 0.07-0.68,p = 0.02, I 2 = 34 %) in favor of a yoga intervention compared to active/inactive control groups.Our findings demonstrate that lower body exercises performed during interventions may be able to increase the range of motion in older people; this may be attributed to a reduction in passive tension and stiffness of the tissues surrounding the joints (Arietaleanizbeaskoa et al., 2022;Domínguez et al., 2016); in fact, these structures tend to limit the range of motion, given that cross-links increase during old age, which makes the connective tissue less flexible in older people (Arietaleanizbeaskoa et al., 2022;Domínguez et al., 2016).

Balance
Another result reported in the present meta-analysis was significant improvements in balance by the TUG in favor of EBT compared to active/inactive control groups.Similar to that reported by Tsai et al. (2022) in a meta-analysis in older people with sarcopenia showing significant improvements in TUG (SMD = − 2.17; 95 % CI = − 2.93 to − 1.42;I 2 = 57 %; p < 0.00001) in favor of EBT compared to active/ inactive control groups.Similarly, a meta-analysis conducted by Lu et al. (2021) in older people with sarcopenia reported improvements in TUG (SMD = − 0.92, 95 % CI: − 1.30 to − 0.55, p < 0.00001, I 2 = 22 %) in favor of resistance training and vibration platform training (SMD = − 0.30, 95 % CI: − 0.60 to 0.00, p = 0.05, I 2 = 0 %) compared to active/ inactive control groups.EBT may have improved the capacity of the lower body extensor muscles during walking (Kerrigan et al., 1998).Additionally, it may have improved the strength of the hip stabilizing muscles, specifically the gluteus medius, and minimums, allowing better balance and improving pelvic gait when walking during the test (Acosta Benito et al., 2022;Kwak et al., 2016).

Dosage
The duration of the interventions ranged from 6 to 72 weeks, with a frequency of 2 to 3 weekly sessions of 18 to 80 min per session and intensities of 5 to 14 in the OMNI-RES.Similar to that reported by Tsai et al. (2022) in a meta-analysis in older people with sarcopenia with interventions ranging from 6 to 36 weeks with a frequency of 2 to 3 sessions per week of 20 to 60 min showing significant improvements in appendicular skeletal muscle mass index (p = 0.01), in maximal isometric handgrip strength (p = 0.02) and TUG (p < 0.00001).Similar to what was reported in the meta-analysis of Mañas et al. (2021) in apparently healthy older people from community homes, unsupervised EBT interventions were carried out for 12 to 52 weeks with a frequency of 2 to 3 sessions per week of 30 to 60 min duration with an intensity of 10 to 15 in RPE finding not significant improvements in maximal isometric handgrip strength (p = 0.93).The increase in maximal isometric handgrip strength in another systematic review (Tsai et al., 2022) may be due to the fact that the participants were not in full health to begin with (sarcopenia) whereas in the review by Mañas et al. (2021) the older people were apparently healthy and the response to improvement in this variable may decrease.In an overview carried out by Hernandez et al. (2024) in apparently healthy older people with interventions through active exergames with durations that ranged between 3 and 20 weeks, 2 to 3 sessions per week of 30 to 60 min per week with moderate to vigorous intensities of 3 to 6 points in 10-point RPE showing significant improvements in TUG (p < 0.0001) and 30-second chair stand test (p = 0.0008).
Our systematic review reported that the certainty of evidence is very low which does not allow us to establish definitive recommendations on using EBT to improve body composition and physical performance in apparently healthy older people.Similarly in a systematic review with meta-analysis on different modalities of resistance training in older people with sarcopenia on physical performance, a low certainty of evidence was presented (Lu et al., 2021).Some systematic reviews of EBT on flexibility, balance, and fragility (Daryanti Saragih et al., 2022;Yeun, 2017) in older people have been published; however, they have not included certainty of evidence in their analyses (Daryanti Saragih et al., 2022;Yeun, 2017).

Limitations and strengths
This systematic review's limitations include: (i) the participants' inability to manage their food consumption and their failure to complete a food record, which could have an impact on body composition and reveal their eating habits and macronutrient and micronutrient intake and the participants' physical performance; (ii) some of the articles analyzed did not report intensity (four did not mention it), which makes it challenging to replicate these interventions; and (iii) the degree of the certainty of evidence was very low, making it challenging to recommend the interventions analyzed.Strengths include: (i) methodological quality over 60 % in the examined studies; (ii) the PRISMA, PROSPERO, TESTEX, RoB 2, and GRADE scales' methodological procedures, (iii) using the core collection of six generic databases (PubMed, ProQuest, EBSCOhost, CINAHL Complete, Scopus, and Web of Science); and (iv) analyzing how EBT affects older people's body composition and physical performance.The findings presented in this systematic review indicate that EBT based therapy may have a positive impact on the physical performance of older people, specifically in 30-second chair stand, sitand-reach and TUG tests.They also demonstrate that the interventions under consideration may enhance this population's general health status and quality of life.Considering the outcomes of our analysis and the contradictions in the reported results on maximal isometric handgrip strength and body composition we suggest creating new studies that will cover additional facets of health status, such as physiological, and/or biochemical measures.These studies should incorporate high-quality methodologies, such as double-blind randomization, supervised control groups, and previously registered research protocols.However, to maximize physical performance and quality of life in older people, interventions should include a more detailed explanation of dosage (i.e., duration, volume, intensity, density) and activities performed.This will allow the replication of the interventions carried out and obtain benefits similar to those reported by studies, optimizing the design of training programs, applying interventions based on scientific evidence.

Conclusion
EBT significantly improved physical performance, mainly in 30-second chair stand, TUG, and sit-and-reach tests in apparently healthy older people.However, the certainty of evidence is very low; therefore, not definitive recommendations can be made for or against EBT.Therefore, more high-quality studies are needed to draw definitive conclusions.Nevertheless, based on the analyzed studies in our systematic review, it is recommended to use elastic bands training to improve physical performance in older people with a frequency of 2 or 3 weekly sessions of 50 to 80 min on alternate days.

Fig. 4 .
Fig. 4. Effect of elastic band training compared to control groups on the following outcome: fat-free mass.The squares indicate the study-specific effect estimate.Bars indicate the width of the corresponding 95 % confidence interval.The diamond is centered on the summary effect estimate, and the width indicates the corresponding 95 % confidence interval.

Fig. 5 .
Fig. 5. Effect of elastic band training compared to control groups on the following outcome: maximal isometric handgrip strength.The squares indicate the studyspecific effect estimate.Bars indicate the width of the corresponding 95 % confidence interval.The diamond is centered on the summary effect estimate, and the width indicates the corresponding 95 % confidence interval.

Fig. 6 .
Fig. 6.Effect of elastic band training compared to control groups on the following outcome: 30-second chair stand.The squares indicate the study-specific effect estimate.Bars indicate the width of the corresponding 95 % confidence interval.The diamond is centered on the summary effect estimate, and the width indicates the corresponding 95 % confidence interval.

Fig. 7 .
Fig. 7. Effect of elastic band training compared to control groups on the following outcome was back-scratch.The squares indicate the study-specific effect estimate.Bars indicate the width of the corresponding 95 % confidence interval.The diamond is centered on the summary effect estimate, and the width indicates the corresponding 95 % confidence interval.

Fig. 8 .
Fig. 8. Effect of elastic band training compared to control groups on the outcome: sit-and-reach.The squares indicate the study-specific effect estimate.Bars indicate the width of the corresponding 95 % confidence interval.The diamond is centered on the summary effect estimate, and the width indicates the corresponding 95 % confidence interval.

Fig. 9 .
Fig. 9. Effect of elastic band training compared to control groups on the outcome: timed up-and-go.The squares indicate the study-specific effect estimate.Bars indicate the width of the corresponding 95 % confidence interval.The diamond is centered on the summary effect estimate, and the width indicates the corresponding 95 % confidence interval.

Fig. 10 .
Fig. 10.Egger's test for assessment of potential publication bias.A: 30-second-chair stand test.B: back-scratch test.C: fat-free mass.D: maximal isometric handgrip strength.E: sit-and-reach test.F: timed up-and-go test.

Table 1
Selection criteria used in the systematic review.

Table 2
Evaluation of the quality of the study according to the TESTEX scale.
a Three points are possible: one point if adherence >85 %, one point if adverse events were reported, and one point if exercise attendance was reported.bTwo points possible: one point if the primary outcome is reported, one point if all other outcomes were reported.#total out of 15 points.TESTEX: tool for assessment of study quality and reporting in exercise(Smart

Table 3
Studies report the effects of elastic band training on older people's body composition and physical performance.