Impact of wearable-assisted walking on sarcopenia and body composition in older adults

Abstract

Background

Physical inactivity constitutes a global public health challenge, particularly among older adults. This study investigates the impact of a wearable device-assisted walking program on body composition, muscle strength, and sarcopenia in this population.

Methods

Eighty healthy older adults (40 men and 40 women), aged 65 and above, were randomly assigned to either an intervention (experimental) group, which participated in a structured walking program utilizing wearable devices, or a non-intervention (control) group, which maintained their habitual activities. The walking program spanned 12 weeks. Key assessments included body composition analysis, handgrip strength measurement, the 5× sit-to-stand test, and the appendicular skeletal muscle index (ASMI).

Results

The intervention group exhibited significant improvements in body composition and muscle strength. Skeletal muscle mass (SMM) increased by 5.5% (95% CI: 4.2–6.8%), handgrip strength improved by 13.1% (95% CI: 10.6–15.6%, p < 0.05), and lower limb function improved by 10.5% (p < 0.05). ASMI proved to be an effective and accessible metric for sarcopenia assessment, demonstrating strong correlations with SMM, handgrip strength, and sit-to-stand performance.

Conclusions

Wearable devices, such as smartwatches and fitness trackers, provide real-time feedback and enable tailored exercise recommendations, offering a practical and personalized strategy for mitigating sarcopenia. This study confirms that wearable-assisted walking programs effectively improve muscle health and functional performance in older adults, reinforcing their potential for promoting healthy aging.

Background

Physical inactivity has emerged as a significant global public health concern, particularly among older adults, who are at increased risk of skeletal muscle loss and diminished functional capacity. Prolonged periods of sedentary behavior, such as sitting or lying down, contribute to sarcopenia – a condition defined by the gradual decline in skeletal muscle mass (SMM) and strength. Sarcopenia increases frailty, falls, and disability among aging populations [1]. Although evidence supports walking as an intervention for sarcopenia, limited studies have specifically examined the effectiveness of wearable-assisted walking programs in improving body composition and muscle function in older adults. Research indicates that after the age of 70, muscle mass decreases at a rate of 0.5–1.0% per year, with this rate accelerating to 25–40% per decade [2]. The World Health Organization (WHO) has identified physical inactivity as the fourth leading risk factor for mortality, recommending that individuals engage in at least 150 min of moderate-intensity physical activity per week to preserve muscle mass, strength, and functional mobility [3]. Furthermore, many studies have underscored the urgent need for effective interventions to combat sarcopenia and promote healthy aging [3–5].

Engagement in regular physical activity is well-established as a fundamental strategy to enhance physical function, mood, and overall health, while simultaneously reducing abdominal fat and preserving muscle strength [6]. However, older adults aged 65 and above frequently demonstrate low levels of physical activity, often adopting increasingly sedentary lifestyles, which exacerbates age-related muscle loss [7]. Walking, a low-impact and accessible form of exercise, is widely acknowledged as one of the most appropriate activities for older adults and has been shown to enhance body composition and increase limb muscle strength [8].

Wearable Technology/Devices and Aging: Wearable technology, including smartwatches (e.g., Apple Watch) and fitness trackers (e.g., Garmin Vivosmart HR), has been extensively utilized in prior research to monitor and promote physical activity among older adults [9–11]. Validated in studies of aging populations, these devices provide real-time feedback on activity levels, enabling users to monitor progress and adjust daily routines. The practicality of these devices for older adults lies in their ability for continuous monitoring without requiring substantial behavioral changes. This feature facilitates the integration of physical activity into their daily lives, especially for those with mobility or cognitive impairments. Additionally, wearable devices can be used to tailor exercise programs to individual needs and capabilities, offering personalized recommendations that enhance engagement and adherence.

Walking as an Intervention: Walking was chosen as the intervention in this study because it is one of the most widely recommended physical activities for older adults, particularly for those at risk of sarcopenia. Unlike resistance exercises, which may require specialized equipment or guidance, walking is an easily accessible and low-impact activity that can be performed by individuals with varying levels of physical fitness. While resistance training is effective for managing sarcopenia [12], walking serves as a viable alternative, especially for older adults who may be unable to engage in more intense exercises due to health constraints. Moreover, walking promotes cardiovascular health, supports joint flexibility, and enhances lower limb muscle strength, all of which are critical factors in preventing the functional decline associated with sarcopenia [1, 12, 13].

Objectives: The primary objectives of this study are to examine how a structured walking program influences body composition, evaluate changes in limb muscle strength, and explore how SMM and limb strength contribute to understanding and managing sarcopenia in older adults. Specifically, the study aims to evaluate the effects of walking, supported by wearable technology, on body composition, muscle mass, and muscle function, thereby providing insights into practical interventions for improving physical health in aging populations. Therefore, this study hypothesizes that a structured walking program with real-time feedback from wearable devices will result in greater improvements in body composition, muscle mass, and functional performance than habitual activity alone.

Methods

This study utilized a quasi-experimental design to evaluate the effects of a wearable device-assisted walking program on body composition, muscle strength, and sarcopenia in older adults. The design included both an intervention group and a control group, structured to emulate realistic physical activity conditions for older adults, thereby enhancing the applicability of the findings to real-world settings.

Inclusion and exclusion criteria

Participants were defined as “healthy older adults” if they met the following criteria: (1) age 65 or older, (2) independent living without major cardiovascular, metabolic, or neuromuscular diseases (e.g., uncontrolled hypertension, insulin-dependent diabetes, or severe arthritis), and (3) no history of severe spinal injury or chronic pain limiting mobility. Mild, stable conditions (e.g., controlled hypertension) were permitted if they did not restrict physical activity. These criteria align with WHO definitions of healthy aging and sarcopenia screening protocols [14, 15].

A preliminary health assessment was conducted to collect basic information, including age, height, weight, and other relevant health markers for inclusion criteria. To enhance the generalizability of the study findings, the demographic information related to socioeconomic status were also collected, such as income and education level, to assess how representative the sample is of the broader older adult population and whether the findings are applicable across different socioeconomic groups.

Exclusion criteria includes individuals with physical or cognitive impairments that would impede safe participation. Cognitive health was assessed during screening using the mini-mental state examination, with scores below 24 indicating potential impairment and necessitating exclusion. This threshold aligns with standard protocols for identifying clinically significant cognitive decline in older adults [14, 16].

Study settings and design

Following the quasi-experimental design, researchers recruited 80 healthy adults aged 65 and older (40 men, 40 women) and randomly assigned them to an intervention group and a control group using a computer-generated randomization sequence after eligibility confirmation and baseline information collection. The RAND function in Microsoft Excel generated the randomization sequence. An independent researcher not involved in the intervention process and outcome assessments carried out group allocation to reduce the risk of selection bias. Also, to reduce potential measurement bias, outcome assessors were blinded to the group allocation throughout the intervention period. In accordance with the WHO definition, “healthy older adults” were those capable of maintaining independent living functions and general cognitive competence [15]. The study adhered to the sample size guidelines for comparing experimental and control groups as outlined by Mills and Gay, which recommend a minimum of 30 participants per group [17]. The total sample of 80 participants was evenly distributed between the intervention (experimental) group and the non-intervention (control) group, with the sample size calculated to ensure a 95% confidence interval (CI) for baseline comparability in terms of age, height, and weight [18].

Both groups utilized wearable devices (smartwatches or fitness trackers), with the experimental group engaging in a structured walking program designed to be accessible and sustainable for older adults. The program focused on moderate-intensity aerobic activity, with the intervention group instructed to walk at a pace of 100 steps per minute for at least 30 min per session, five days per week. This pace was monitored in real-time using wearable devices (Garmin Vivosmart HR and Apple Watch), which provided auditory or vibrational feedback if the step rate fell below the target. During supervised sessions (Monday to Friday), researchers further verified compliance through device data checks. The intensity and duration were tailored to ensure safety and feasibility for most participants, based on their physical capabilities. The program included supervised walking sessions from Monday to Friday and self-managed walking sessions on weekends, allowing participants flexibility in adhering to the prescribed regimen. Participants in both groups logged their steps and caloric expenditure over the 12-week (3-month) period.

Step count data from the wearable devices were automatically synchronized to a centralized platform (Garmin Connect and Apple HealthKit) daily. Researchers verified data accuracy through weekly cross-checks against device logs and participant self-reports. Discrepancies (e.g., missing data or outliers exceeding 3 standard deviations from the mean) were resolved by reviewing device timestamps and participant activity diaries. This dual verification process ensured the reliability of the step count and caloric expenditure data.

All participants provided informed consent in accordance with ethical guidelines, confirming their history of a free orthopedic or cardiovascular disease, and agreed to refrain from using any supplements that could influence muscle mass or the progression of sarcopenia. Ethical approval for the study was granted by the Human Trials Review Board of Tri-Service General Hospital, National Defense Medical College (Approval No.: C202305014).

Research materials

Wearable devices such as the Garmin Vivosmart HR and Apple Watch were selected for their ability to provide accurate and reliable data on daily steps, calories burned, and walking distance [19]. These devices were chosen for their accessibility, ease of use, and demonstrated validity in tracking physical activity levels in older adults. Both devices are well-established in the field of physical activity research and have been shown to offer reliable measurements for this population. The Garmin Vivosmart HR, in particular, is lightweight and designed for continuous wear, ensuring comfort and practicality for older adults. The Apple Watch offers additional advanced features, including heart rate monitoring, which provides a more comprehensive assessment of exercise intensity and participant exertion.

Activity data were collected biweekly, either in person or via mobile communication. Moderate-intensity physical activity was operationally defined as 100 steps per minute [20]. Step count thresholds followed the classification system proposed by Tudor-Locke et al. for older adults: sedentary (< 2500 steps/day), low activity (2500–5000 steps/day), somewhat active (5000–8000 steps/day), active (8000–11000 steps/day), and highly active (> 11000 steps/day) [21].

Measurement instruments

All assessments, including body composition, handgrip strength, and physical performance tests, were conducted between 8:00 AM and 10:00 AM to minimize diurnal variation and ensure measurement consistency across participants.

Body composition assessment

Body composition was assessed utilizing the portable InBody 370 s analyzer, which employs bioelectrical impedance analysis (BIA) for estimation. Participants stood upright on the device while holding hand sensors for approximately 15 s to obtain various metrics, including body weight, total body water (TBW), protein, minerals, body fat mass (BFM), SMM, and appendicular skeletal muscle index (ASMI), among others [22, 23]. Standard body composition reference values for adults aged 65 and older were applied to Body Mass Index (BMI): 18.5–22.9 kg/m² for both genders [24], Percent Body Fat (PBF): 15–25% for men and 20–30% for women [13], Skeletal muscle mass: Calculated as muscle percentage (muscle mass/body weight × 100%) with reference ranges of 70–84% for men and 60–72.5% for women [25], Basal Metabolic Rate (BMR): Approximately 1600–1800 kcal for men and 1200 kcal for women [26], TBW: 50–60% for men and 45–60% for women [27], and Waist-Hip Ratio (WHR): 0.9 for men and 0.7 for women [22].

Muscle strength assessment

Lower limb strength was evaluated using the Jones Senior Fitness Test (SFT), which measured the 5× sit-to-stand performance and 10-meter walking speed [28, 29]. In the sit-to-stand test, participants crossed their arms over their chest and completed five full stands, with completion time recorded. For the walking speed assessment, participants walked 10 m, with the time taken recorded over three trials, and the fastest result was recorded.

Upper limb strength was quantified using a Jamar handheld dynamometer to measure handgrip strength (HGS). Participants sat with their forearms parallel to the floor and elbows positioned at 90°. An exerted maximum grip force with their dominant hand over three trials were scored, and the average score recorded [30].

Sarcopenia assessment

In accordance with the AWGS 2019 guidelines, the evaluation of sarcopenia was conducted in two stages. The first stage involved assessing lower limb function through the 5× sit-to-stand test and 10-meter walking speed. Scores exceeding 12 s for the sit-to-stand test and below 0.8 m/s for walking speed were indicative of potential sarcopenia [14, 16]. The second stage involved measurements of ASMI based on BIA, where values below 7.0 kg/m² for men and 5.7 kg/m² for women indicated the presence of sarcopenia [14].

Control variables

To address cognitive variability, trained personnel delivered comprehensive instructions and demonstrations for the walking exercise. The weekly walking program encouraged participants in the intervention group to sustain a pace of 100 steps per minute for a minimum of 30 min per session. In contrast, the non-intervention group independently monitored their daily step count and caloric expenditure. The 12-week program did not specifically control factors such as weather, social activity, and medication use. Participants continued their normal daily routines outside of the scheduled walking sessions. Furthermore, all participants received dietary guidance and were advised to refrain from high-protein supplements to mitigate potential confounding effects on changes in muscle mass [31]. This approach ensured that any observed alterations in body composition and muscle strength could be predominantly attributed to the walking intervention rather than dietary modifications.

Statistical analysis

Statistical analyses were conducted using SPSS 27.0 (IBM^®, Armonk, NY, USA), with significance levels established at p < 0.05. Homogeneity tests, including independent t-tests, were performed on age, height, and weight for both groups to assess baseline comparability. To examine pre- and post-test outcomes for both groups, a two-factor analysis of covariance (ANCOVA) was employed. This method controlled for baseline covariates – age, sex, BMI, socioeconomic status (income and education level), and baseline values of the dependent variables (e.g., pre-intervention ASMI, SMM) – to adjust for pre-existing group differences, reduce error variance, and isolate the intervention’s effects on outcome variables [32]. ANCOVA was selected because it adjusts for baseline values, reduces error variance, and provides more reliable estimates of the intervention’s effects by accounting for pre-existing differences between the experimental and control groups. This statistical method helps to isolate the impact of the walking program on outcomes related to body composition, muscle strength, and sarcopenia. Additionally, Pearson’s correlation and linear regression analyses were utilized to explore associations among key variables.

Results

Participant homogeneity test

A statistical t-test was performed to examine the homogeneity between the experimental and control groups with respect to age, height, and weight. The results revealed no significant differences in age (t = − 0.085), height (t = − 0.581), and weight (t = − 0.186) for men participants, and in age (t = 0.584), height (t = 0.666), and weight (t = − 0.050) for women participants (p > 0.05). These findings indicate that the two groups were homogeneous at baseline in terms of demographic characteristics across both genders.

Impact of walking intervention on daily steps and caloric expenditure

The 12-week walking intervention resulted in significant increases in daily steps and caloric expenditure on weekdays (p < 0.05). Participants in the experimental group showed a notable improvement in physical activity, consistently achieving between 5,000 and 8,000 steps per day, categorizing them as “somewhat active” to “active.” On weekdays, the experimental group consistently surpassed an average of 8,000 steps per day, while the control group maintained an average of about 2,500 steps, indicating a sedentary lifestyle. Regarding caloric expenditure, the experimental group exhibited a substantial increase, expending between 350 and 377 kcal/day, compared to the control group, which expended between 127 and 144 kcal/day. However, no significant differences were noted in steps or caloric expenditure during weekends (see Table 1). The insignificant change in weekend step counts (p = 0.27) likely stems from decreased adherence to structured activity on non-supervised days, a period when participants may have experienced reduced motivation or lacked scheduled guidance. These findings highlight the positive impact of the intervention on both physical activity levels and caloric expenditure during weekdays.

Table 1.

Weekday step counts and caloric expenditure in the intervention vs. control group

Weekdays	Men, EG (n = 20)	Men, CG (n = 20)	t-value	Women, EG (n = 20)	Women, CG (n = 20)	t-value
	M ± SD	M ± SD		M ± SD	M ± SD
Mon, Steps	8421.33 ± 91.75 (6128.17/2293.16)	2743.17 ± 212.57	22.45*	7654.52 ± 129.47 (5667.11/1987.41)	2526.65 ± 187.16	20.53*
Calorie, kcal	350.01 ± 17.82 (239.27/110.74)	144.51 ± 9.24	16.42*	318.56 ± 13.59 (222.62/95.94)	127.71 ± 8.06	15.72*
Tues, Steps	8645.26 ± 88.39 (6215.41/2429.85)	2635.25 ± 220.37	26.61*	8182.69 ± 121.62 (6067.03/2115.66)	2612.32 ± 164.34	21.44*
Calorie, kcal	355.41 ± 23.58 (243.78/111.63)	135.44 ± 8.22	19.74*	339.87 ± 13.87 (239.51/100.36)	135.49 ± 7.59	16.06*
Wed, Steps	8777.42 ± 82.46 (6338.56/2438.86)	2598.94 ± 227.51	30.92*	7618.91 ± 130.19 (5623.15/1995.76)	2744.43 ± 157.82	18.19*
Calorie, kcal	377.29 ± 29.17 (264.72/112.57)	127.62 ± 7.86	23.15*	317.30 ± 14.35 (220.43/96.87)	143.35 ± 7.31	13.85*
Thurs, Steps	8281.64 ± 95.95 (5927.43/2354.21)	2691.78 ± 218.46	21.12*	7776.25 ± 127.65 (5768.92/2007.51)	2691.15 ± 163.15	19.37*
Calorie, kcal	342.52 ± 16.22 (232.34/110.18)	131.46 ± 8.12	16.23*	326.99 ± 14.24 (227.71/99.28)	139.92 ± 7.45	14.61*
Fri, Steps	8195.47 ± 97.75 (5987.43/2208.34)	2594.38 ± 228.47	20.97*	7392.97 ± 132.62 (5417.01/1975.96)	2615.74 ± 160.78	17.28*
Calorie, kcal	340.15 ± 15.57 (236.94/103.21)	123.26 ± 7.69	15.33*	305.21 ± 13.84 (211.54/93.67)	126.43 ± 7.35	13.23*
Sat, Steps	3356.61 ± 445.33	3261.51 ± 577.56	0.46	3562.81 ± 515.18	3234.11 ± 623.87	1.07
Calorie, kcal	218.56 ± 19.36	162.56 ± 7.07	0.72	172.37 ± 7.12	148.94 ± 7.22	1.39
Sun, Steps	3872.61 ± 482.37	3337.29 ± 605.77	1.91	3437.22 ± 539.26	3327.59 ± 614.59	0.68
Calorie, kcal	182.43 ± 15.75	154.42 ± 6.15	1.63	169.40 ± 6.84	153.61 ± 6.85	0.94

CG Control Group, EG Experimental Group, Mean ± SD, t-values for significance, *p < 0.05

Effect of interventions on body composition

The intervention induced significant alterations in body composition, with notable improvements in SMM, soft lean mass (SLM), BMR, bone mineral content (BMC), and ASMI. In contrast, reductions were observed in TBW, BFM, fat-free mass (FFM), BMI, PBF, and WHR (see Table 2). These findings are further visualized in Fig. 1, which illustrates pre- and post-intervention comparisons of step counts and SMM in both the experimental group (EG) and control group (CG), stratified by gender. These findings provide robust support for Hypothesis 1, demonstrating that the walking intervention positively impacted the body composition of older adults.

Table 2.

Body composition measures for both interventions

Parameters		Men, EG (n = 20)	Men, CG (n = 20)	F-value (p-value)	Women, EG (n = 20)	Women, CG (n = 20)	F-value (p-value)
		M ± SD	M ± SD		M ± SD	M ± SD
TBW (%)	Pre	37.37 ± 4.81	36.91 ± 2.89	333.07* (0.000)	29.82 ± 1.91	30.12 ± 1.95	405.19* (0.000)
	Post	39.81 ± 3.54	36.62 ± 2.15		32.04 ± 2.35	30.08 ± 1.91
Protein (g)	Pre	9.99 ± 1.32	9.74 ± 0.95	570.35* (0.000)	7.95 ± 0.52	7.88 ± 0.45	147.56* (0.000)
	Post	10.24 ± 1.29	9.72 ± 0.94		7.86 ± 1.49	7.85 ± 0.43
Minerals (g)	Pre	3.39 ± 0.47	3.29 ± 0.28	2590.66* (0.000)	2.92 ± 0.24	2.85 ± 0.24	1907.59* (0.000)
	Post	3.43 ± 0.46	3.29 ± 0.27		2.96 ± 0.24	2.84 ± 0.24
BFM (kg)	Pre	20.68 ± 4.17	20.44 ± 4.75	770.75* (0.000)	26.35 ± 5.58	27.62 ± 4.77	1724.05* (0.000)
	Post	18.86 ± 3.62	21.10 ± 4.90		23.27 ± 5.28	27.64 ± 4.66
SLM (kg)	Pre	47.94 ± 6.23	45.74 ± 2.92	2611.58* (0.000)	38.24 ± 2.46	36.99 ± 2.67	364.52* (0.000)
	Post	48.53 ± 6.29	45.15 ± 3.11		39.03 ± 2.46	37.01 ± 2.85
FFM (kg)	Pre	50.75 ± 6.59	50.41 ± 4.24	455.92* (0.000)	40.69 ± 2.62	40.67 ± 2.86	456.57* (0.000)
	Post	47.82 ± 5.97	51.01 ± 4.27		36.83 ± 2.26	40.72 ± 2.73
SMM (kg)	Pre	27.23 ± 4.07	27.41 ± 2.37	475.62* (0.000)	21.54 ± 1.58	21.78 ± 1.80	201.18* (0.000)
	Post	28.72 ± 4.37	27.18 ± 2.33		22.34 ± 1.76	21.60 ± 1.65
BMI (kg/m2)	Pre	25.85 ± 2.01	26.00 ± 1.62	211.03* (0.000)	26.35 ± 2.58	26.57 ± 2.71	777.39* (0.000)
	Post	24.59 ± 1.69	26.30 ± 1.56		24.84 ± 2.72	26.58 ± 2.74
PBF (%)	Pre	28.85 ± 3.96	29.11 ± 4.79	770.75* (0.000)	38.98 ± 5.37	39.33 ± 4.29	1221.60* (0.000)
	Post	26.99 ± 3.83	29.95 ± 5.04		35.21 ± 4.94	38.98 ± 3.78
BMR (kcal)	Pre	1422 ± 79.02	1437.20 ± 73.02	330.47* (0.000)	1249.1 ± 56.50	1255.8 ± 40.52	1452.02* (0.000)
	Post	1526.4 ± 92.10	1432.10 ± 70.45		1267.45 ± 54.93	1257.4 ± 41.71
WHR (W/H)	Pre	0.91 ± 0.06	0.93 ± 0.08	298.10* (0.000)	0.90 ± 0.03	0.92 ± 0.04	154.28* (0.000)
	Post	0.86 ± 0.04	0.94 ± 0.08		0.86 ± 0.04	0.91 ± 0.04
BMC (g)	Pre	2.81 ± 0.38	2.72 ± 0.23	6525.35* (0.000)	2.44 ± 0.18	2.46 ± 0.21	82.71* (0.000)
	Post	2.83 ± 0.38	2.71 ± 0.23		2.46 ± 0.18	2.48 ± 0.21
ASMI (kg/m2)	Pre	7.16 ± 0.53	7.15 ± 0.29	1081.12* (0.000)	6.20 ± 0.32	6.19 ± 0.32	261.70* (0.000)
	Post	7.89 ± 0.54	7.14 ± 0.40		6.75 ± 0.29	6.18 ± 0.25

CG Control Group, EG Experimental Group, TBW Total Body Water, BFM Body Fat Mass, SLM Soft Lean Mass, FFM Fat-Free Mass, SMM Skeletal Muscle Mass, BMI Body Mass Index, PBF Percentage Body Fat, BMR Basal Metabolic Rate, WHR Waist-to-Hip Ratio, BMC Bone Mineral Content, ASMI Appendicular Skeletal Muscle Index, Pre-and post-test values were presented as Mean Values ± Standard Deviations (M ± SD), F-value was presented as F-value (p-value), *p < 0.05

Fig. 1.

Average weekday step counts and SMM pre and post across EG and CG stratified by gender. *Note; SMM: Skeletal Muscle Mass, EG: Experimental Group, CG: Control Group. The EG showed marked increases in both average weekday step counts and SMM after the intervention, especially among males, while the CG showed minimal or no changes. These results further support the impact of wearable-assisted walking on improving activity levels and muscle mass

Effect on upper and lower limb muscle strength

The walking intervention resulted in significant improvements in both upper and lower limb strength. Handgrip strength demonstrated notable increases in both genders. The Upper limb strength, as measured by HGS, demonstrated significant improvements in the experimental group (men: 28.95 kg to 32.75 kg; women: 16.20 kg to 19.85 kg), with F-values of 599.98* (p < 0.05) for men and 221.85* (p < 0.05) for women. These improvements corresponded to a large effect size for HGS (Cohen’s d = 0.85 for men; d = 0.78 for women), indicating substantial practical significance of the intervention. Similarly, significant enhancements were observed in the 5× sit-to-stand time and walking speed. The lower limb strength, assessed through the 5× sit-to-stand test and 10-meter walking speed, exhibited notable enhancements (see Table 3). The sit-to-stand time decreased (men: 8.09 to 7.50 s; women: 9.92 to 8.73 s), while walking speed increased (men: 1.14 m/s to 1.25 m/s; women: 1.03 m/s to 1.14 m/s), thus further substantiating Hypothesis 2; Specifically, the experimental group exhibited a reduction in sit-to-stand time and an increase in walking speed, whereas no significant changes were noted in the control group.

Table 3.

Pre- and Post-Intervention data

Parameters		Men, EG (n = 20)	Men, CG (n = 20)	F-value (p-value)	Women, EG (n = 20)	Women, CG (n = 20)	F-value (p-value)
		M ± SD	M ± SD		M ± SD	M ± SD
HGS (kg)	Pre	28.95 ± 4.83	28.60 ± 5.09	599.98* (0.000)	16.20 ± 2.54	16.95 ± 2.64	221.82* (0.000)
	Post	32.75 ± 4.91	28.1 ± 4.84		19.85 ± 2.67	16.85 ± 2.22
5× Sit-to-Stand (sec)	Pre	8.09 ± 1.04	10.33 ± 1.09	1037.1* (0.000)	9.92 ± 0.73	10.05 ± 0.62	186.97* (0.000)
	Post	7.50 ± 0.91	10.34 ± 0.92		8.73 ± 0.81	10.03 ± 0.63
Walking Speed (m/sec)	Pre	1.14 ± 0.09	1.14 ± 0.09	519.50* (0.000)	1.03 ± 0.07	1.04 ± 0.07	347.21* (0.000)
	Post	1.25 ± 0.10	1.13 ± 0.09		1.14 ± 0.08	1.04 ± 0.06

CG Control Group, EG Experimental Group, HGS Handgrip Strength, Pre-and post-test values were presented as Mean Values ± Standard Deviations (M ± SD). F-value was presented as F-value (p-value), *p < 0.05

Relationship between SMM, HGS, 5× Sit-to-Stand, and ASMI

As depicted in Table 4, the correlation analysis reveals significant positive relationships among SMM, HGS, 5× Sit-to-Stand performance, and ASMI in both genders, underscoring their relevance in the assessment of muscle health and sarcopenia. The SMM exhibits a particularly strong correlation with ASMI (r = 0.954, p < 0.01), along with notable correlations with 5× sit-to-stand performance (r = 0.846, p < 0.01) and HGS (r = 0.722, p < 0.01) in the men cohort. Likewise, ASMI in men demonstrates a robust correlation with HGS (r = 0.747, p < 0.01) and 5× Sit-to-Stand performance (r = 0.915, p < 0.01). In contrast, a lower correlation was observed in the women, while SMM still shows a moderate to strong positive correlation with ASMI (r = 0.555, p < 0.05), 5× Sit-to-Stand performance (r = 0.492, p < 0.05), and HGS (r = 0.502, p < 0.05). These findings suggest that SMM, HGS, and lower limb strength are closely associated with ASMI, reinforcing their significance as critical markers of sarcopenia, with variations in correlation strength observed between genders. This evidence supports the application of these metrics in both clinical and community settings for monitoring muscle health in older adults.

Table 4.

Correlation analysis for predictive variables

Sex	Variables	SMM	HGS	5× Sit-to-Stand	ASMI
Men	SMM	1
	HGS	0.722**	1
	5× Sit-to-Stand	0.846**	0.772**	1
	ASMI	0.954**	0.747**	0.915**	1
Women	SMM	1
	HGS	0.502*	1
	5× Sit-to-Stand	0.492*	0.378	1
	ASMI	0.555*	0.455*	0.785**	1

SMM Skeletal Muscle Mass, HGS Handgrip Strength, ASMI Appendicular Skeletal Mass Index, ^*p < 0.05, ^**p < 0.01

The regression analysis indicates that SMM, HGS, and the 5× Sit-to-Stand test are significant predictors of ASMI in older adults, supporting their role as indicators of sarcopenia (Table 5). The men cohort model exhibits strong explanatory power, with these variables accounting for 95.4% of the variation in ASMI (R² = 0.961, Adjusted R² = 0.954, p < 0.01), suggesting a robust capacity to predict ASMI based on measures of muscle mass and limb strength. In the women cohort, the model accounts for 74.7% of the variation in ASMI (R² = 0.787, Adjusted R² = 0.747, p < 0.01), demonstrating substantial predictive capability, though slightly lower than that observed in men. These findings support Hypothesis 3, emphasizing SMM, HGS, and lower limb function as effective indicators of muscle health and sarcopenia risk in older adults, with minor gender differences noted in predictive accuracy. The observed relationship between step count and SMM improvement is further illustrated in Fig. 2, which highlights the positive association between physical activity and muscle gains across genders and groups. This results underscore the critical importance of assessing muscle mass and functional strength for the early detection of sarcopenia and implementing targeted interventions within aging populations.

Table 5.

Regression analyses for predictive variables

Sex	Variables	β	95% CI	R	R 2	Adjusted R2	p-value
Men	SMM	0.495	0.412	0.981	0.961	0.954	0.000
	HGS	0.311	0.220				0.002
	5× Sit-to-Stand	0.431	0.357				0.000
Women	SMM	0.418	0.302	0.887	0.787	0.747	0.000
	HGS	0.294	0.180				0.004
	5× Sit-to-Stand	0.397	0.285				0.000

Independent Variables: SMM Skeletal Muscle Mass, HGS Handgrip Strength, 5× Sit-to-Stand, Dependent Variable: ASMI Appendicular Skeletal Mass Index, ^*p < 0.05, ^**p < 0.01

Fig. 2.

Average weekday steps against skeletal muscle mass improvements (Δ SMM) across all groups. *Note: A strong positive correlation (r = 0.79) between step counts and SMM gains was particularly evident in the EG. This visual evidence supports the notion that increased physical activity is associated with greater muscle mass improvements in older adults. Independent Variables: SMM: Skeletal Muscle Mass, HGS: Handgrip Strength, 5× Sit-to-Stand, Dependent Variable: ASMI: Appendicular Skeletal Mass Index, * p < 0.05, ** p < 0.01

Skeletal muscle mass, handgrip strength, and 5× sit-to-stand performance emerged as significant predictors of ASMI in both genders, as demonstrated by regression analysis. Specifically, in men, a one-unit increase in SMM corresponded to a 0.495 increase in ASMI (95% CI [0.412, 0.578]), highlighting a strong positive association. Similar patterns were observed in women, albeit with slightly narrower confidence intervals (see Table 5). The exclusion of zero from these intervals reinforces the clinical significance of these associations.

Discussion

This study highlights the significant effects of a 12-week walking exercise intervention on physical activity levels and caloric expenditure in older adults. On average, Men in the experimental group reported an average of 6,119 steps per day, resulting in a caloric expenditure of 243.41 kcal, while women averaged 5,709 steps per day, expending 224.36 kcal. In contrast, the control group, which maintained a routine without additional exercise, averaged 2,500–2,750 steps per day, burning less than 150 kcal. These differences in activity levels positioned the intervention participants in the “somewhat active” to “active” categories, aligning with prior studies that show physical activity can significantly enhance mobility and overall health in older adults [33, 34]. Similar studies have noted that trained older adults who consistently engage in physical activities, such as walking, exhibit higher levels of mobility and overall physical health [35, 36]. Our findings corroborate systematic reviews on wearable technology for older adults, which emphasize that real-time feedback and goal setting (e.g., step targets) significantly enhance physical activity adherence [9–11]. Notably, a meta-analysis by Vijayan et al. (2021) [11] demonstrated that wearable devices increase daily steps by 15–25% in aging populations, whereas we observed a substantial 144% increase in weekday steps (from 2,500 to 8,000 steps/day). This synergy between device-driven monitoring and structured exercise underscores the potential of combined interventions to mitigate sarcopenia progression. Weekday activity levels increased markedly; however, no significant improvements occurred during weekends. The lack of structured sessions, scheduled goals, or external accountability on non-program days may explain this discrepancy. Researchers did not implement specific strategies to promote weekend engagement, and participants walked at their discretion. Future interventions may benefit from incorporating additional support mechanisms on weekends, such as scheduled group walks, digital prompts, or involvement from family members or caregivers, to enhance adherence and ensure more consistent activity patterns throughout the week.

These results support previous findings indicating that increased physical activity is associated with improvements in muscle health and body composition among older adults. The walking intervention also produced substantial improvements in body composition. Significant reductions were observed in BMI, PBF, FFM, BFM, and WHR, while SLM, SMM, BMR, and ASMI saw significant increases. These results mirror findings from previous studies that link physical activity to positive changes in body composition, especially in older populations [37]. Specifically, the observed increases in SMM and decreases in PBF and BFM aligns with studies suggesting that regular physical activity, such as walking, may help replace fat with lean muscle tissue in aging adults [38]. Additionally, the increase in TBW observed among both genders supports existing literature indicating that TBW tends to decrease with aging, which often results in increased adiposity [27]. The improvements in BMI, which is an essential indicator of health in older adults, further underscore the role of physical activity in reducing the risk of functional decline associated with higher BMI [39, 40].

While BMI is a useful measure, it does not distinguish between muscle and fat mass. In this study, the integration of multiple body composition indicators – such as ASMI, SMM, and HGS–enhances the precision of health assessments. This is particularly important since elevated BMI can mask significant variations in fat and muscle mass. Notably, ASMI, a key indicator of sarcopenia, improved in both men and women following the intervention, with men’s average ASMI reaching 7.89 ± 0.54 kg/m² and women’s reaching 6.75 ± 0.29 kg/m², both of which exceeded the AWGS cut-off points of 7.0 kg/m² for men and 5.7 kg/m² for women. These findings align with previous research that supports the use of ASMI as a reliable metric for sarcopenia screening in older adults. The intervention resulted in a 10.2% increase in ASMI among men (from 7.16 ± 0.53 to 7.89 ± 0.54 kg/m²) and an 8.9% increase in women (from 6.20 ± 0.32 to 6.75 ± 0.29 kg/m²), exceeding the AWGS guidelines’ diagnostic thresholds for sarcopenia [14]. Similarly, SMM improved by 5.5% in men (27.23 ± 4.07 to 28.72 ± 4.37 kg) and 3.7% in women (21.54 ± 1.58 to 22.34 ± 1.76 kg). Concurrently, BFM decreased by 8.8% in men and 11.7% in women, demonstrating the intervention’s dual impact on muscle preservation and fat reduction. These changes align with prior studies reporting 5–15% improvements in ASMI following structured physical activity programs [37, 41]. These findings underscore the value of integrating wearable-assisted walking interventions into community or clinic-based health programs. By providing real-time feedback and step goals, such programs can enhance adherence, increase daily activity levels, and reduce the incidence of sarcopenia among older adults.

Regression analysis revealed significant positive correlations between SMM, HGS, and sit-to-stand performance with ASMI in both men (95.4%) and women (74.7%), suggesting that SMM, grip strength, and lower limb strength are critical predictors of muscle health in older adults. This reinforces the utility of ASMI in identifying individuals at risk for sarcopenia and underscores the importance of muscle preservation in aging populations.

However, the improvement in muscle strength observed in this study is particularly notable for upper limb strength, as indicated by the significant increases in handgrip strength. This result is somewhat surprising given that walking primarily targets lower limb activity. One explanation could be the functional relationship between upper and lower limb strength – improvements in lower limb mobility and muscle strength may indirectly enhance overall muscular endurance, including that of the upper limbs. Several interrelated mechanisms may explain the improvement in HGS despite the walking-focused intervention. Initially, rhythmic walking enhances neuromuscular coordination through synchronized activation of core and postural muscles, thereby indirectly supporting upper limb function by improving whole-body kinetic chain efficiency [42, 43]. Furthermore, the arm swing during walking, while not resistance-based, provides repetitive low-intensity engagement of shoulder and forearm muscles, potentially mitigating disuse atrophy in sedentary older adults [40]. Additionally, systemic adaptations, including increased blood flow and mitochondrial efficiency from aerobic exercise, may foster muscle protein synthesis throughout both upper and lower limbs [36]. These mechanisms are consistent with research demonstrating that lower-body training can induce cross-education effects on untrained muscles via neural pathways [44], suggesting that even localized interventions can produce integrative musculoskeletal benefits in aging populations. These findings are in line with previous research indicating that walking, as a dynamic rhythmic activity, promotes both muscle strength and joint flexibility [45].

Regarding lower limb strength, significant improvements were observed in both the 5× sit-to-stand test and walking speed, supporting the notion that the intensity and pace of the walking program may have stimulated muscle growth, as evidenced by increases in SMM and SLM. These improvements align with findings from previous studies suggesting that regular walking enhances SMM, particularly in the lower limbs, and contributes to a higher BMR and caloric expenditure in older adults [46].

Despite the promising findings, several limitations should be acknowledged. Although wearable devices improve activity tracking, variations in device sensitivity and participant adherence may have influenced the accuracy of step count and intervention fidelity. Future studies should consider integrating multi-sensor systems or accelerometers for cross-validation and objective adherence assessment. Additionally, longer-term studies are warranted to assess the sustainability of intervention effects and their impact on functional outcomes and quality of life in aging populations.

Limitations

Several limitations require consideration and suggest avenues for future research. First, uncontrolled factors, including participant fatigue, illness, and intermittent lack of motivation, may have influenced adherence and overall outcomes. Future studies should incorporate strategies to mitigate these influences. Second, the lack of systematic monitoring of seasonal and weather-related variations (e.g., rainfall, extreme temperatures) during the 12-week intervention period may have affected outdoor walking adherence. Future research should account for environmental conditions. Third, the racial, ethnic, and genetic homogeneity of the sample restricts the generalizability of the findings. Future studies should prioritize the inclusion of a more diverse population. Finally, while wearable technology was utilized for step tracking, potential issues with compliance and accuracy may have impacted data reliability. Future research could enhance data reliability by integrating advanced devices with environmental sensors to concurrently track activity and weather conditions. In addition, future studies should consider employing randomized controlled trials with longer follow-up periods to assess the sustainability of intervention effects and determine their long-term impact on functional independence and sarcopenia progression.

Conclusions

The significant findings of this study emphasize that physical activity, particularly walking, is an optimal exercise modality for older adults when customized to appropriate parameters. The structured walking program significantly enhanced body composition and muscle strength in both upper and lower limbs, advocating for its integration into public health policies. Recommendations include: incorporating structured walking plans into community health initiatives, leveraging wearable devices for real-time monitoring to enhance adherence, and training healthcare providers to utilize ASMI and grip strength assessments for early sarcopenia detection. Furthermore, ASMI serves as an effective and accessible metric for sarcopenia evaluation, providing valuable insights into muscular health. The interventions are practical for community-based health assessments, establishing a reliable framework for physical health monitoring in aging populations. Policymakers should prioritize these low-cost, scalable strategies to reduce sarcopenia-related disability burdens in resource-constrained settings. These findings provide a foundation for future research, offering evidence-based guidelines for exercise interventions aimed at mitigating age-related muscle loss. Conclusively, wearable-assisted walking programs represent a scalable and cost-effective intervention strategy to combat sarcopenia in older adults, demonstrating significant improvements in muscle mass, physical strength, and body composition.

Abbreviations

ASMI

Appendicular Skeletal Muscle Index

BIA

Bioelectrical Impedance Analysis

BFM

Body Fat Mass

BMC

Bone Mineral Content

BMI

Body Mass Index

BMR

Basal Metabolic Rate

CG

Control Group (Non-intervention)

EG

Experimental Group (Intervention)

FFM

Fat-Free Mass

HGS

Handgrip Strength

PBF

Percent Body Fat

SFT

Senior Fitness Test

SLM

Soft Lean Mass

SMM

Skeletal Muscle Mass

TBW

Total Body Water

WHR

Waist-Hip Ratio

WHO

World Health Organization

Authors’ contributions

C.E.W. Writing – original draft, Validation, Supervision, Formal analysis, Data curation, Conceptualization, and Y.B.M. Writing – proofreading, review & editing, Visualization, Validation, Supervision, and Statistical validation.

Funding

The authors have received no funding for any research relevant to the present study.

Data availability

Data on the participants were obtained before and after training. All authors confirm the authenticity and availability of the data. All datasets on which this paper’s conclusions are based are available to editors, reviewers, and readers upon request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and received approval from the Human Trials Review Board of Tri-Service General Hospital, National Defense Medical College (Approval No.: C202305014). Informed consent was obtained from all participants following a detailed explanation of the study’s purpose, procedures, and potential risks. To ensure understanding, particularly among older adults with varying literacy levels.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.