In this exploratory trial, we recruited older adults aged 60-80 years from the University Social Responsibility program at National Taipei University of Nursing and Health Sciences University (Taipei City), and used a randomized parallel design. A total of 32 participants were randomly assigned to either the experimental (n=16) or control group (n=16) through simple randomization using a random number table, with an allocation ratio of 1:1. The personnel (research assistant) responsible for enrolling participants and those (principal investigator) managing the random allocation sequence were distinct, ensuring that the allocation process remained blinded and minimizing the potential for allocation bias. A pre- and postintervention design was implemented. The intervention consisted of a 70-minute exergame-based mat training session conducted twice weekly for 10 consecutive weeks. Each session included a 10-minute warm-up, approximately 45 minutes of primary training, and a 15-minute cooldown and stretching routine (Table 1). Participants completed the International Physical Activity Questionnaire (IPAQ) and the World Health Organization Quality of Life questionnaire (WHOQOL-BREF, Taiwanese Version) before and after the intervention. In addition, the Senior Functional Test and AFAscan fitness assessment (Scanleader, Scanleader Technologies Co. Ltd.) were conducted. Participants in the control group were instructed to maintain their usual daily activities without any additional structured exercise intervention during the study period. A no-intervention control condition was selected to provide a natural reference for functional fitness and quality-of-life changes without training. This approach allowed us to better isolate the effects of the exergame-based mat training and is consistent with previous exercise trials involving older adults.

A randomized parallel-controlled clinical trial was conducted. A total of 32 eligible participants were randomly assigned to 2 groups: control group and experimental group. The participants completed questionnaires (the IPAQ and WHOQOL-BREF), along with physical fitness assessments (Senior Fitness Test and AFAscan Fitness Assessment), before and after the intervention. Finally, 30 participants completed all experimental procedures required for data analysis.

The mat-based exercise system (Stampede, SHENNONA Co., Ltd) facilitates exercise training by using electronic light and sound effects to guide participants through various exercises, with intensity progressively increasing from moderate to high levels. This system facilitates customized movement programming tailored to specific training objectives, and its architectural diagram is presented in Figure 1A. The training program is divided into 2 phases: the first phase focuses on fundamental movements, whereas the second phase introduces more advanced exercises. With progress, additional exercise equipment is incorporated, along with increased sets, duration, movement variations, and training intensity. Training sessions can be conducted in a simple progressive manner or a game-competitive format (Figure 1B). Throughout the training, participants wore a heart rate monitor (ALATHEC Obeat1) that recorded their exercise intensity and heart rate fluctuations (Figure 2B). Exercise intensity was indicated by lighting colors, corresponding to specific percentages of the maximum heart rate (HRmax) as follows: blue (0%-50% HRmax), cyan (50%-60% HRmax), green (60%-70% HRmax), yellow (70%-80% HRmax), orange (80%-90% HRmax), and red (>90% HRmax). However, the exercise intensity and frequency were adjusted based on participants’ adaptation. This study encompassed 5 different lighting modes, namely all light on, circle on, 4 corners, 2 lines, and diamond (Figure 2A). Each mode was paired with different training protocols and movement patterns, supplemented by training equipment, such as medicine balls, resistance bands, kettlebells, and balance pads. This combination enhanced engagement while simultaneously promoting functional fitness, including muscular strength, muscular endurance, aerobic capacity, flexibility, agility, balance, and core strength. Detailed training protocols are provided in Multimedia Appendix 1. All participants achieved a 100% attendance rate, and adverse events were observed by qualified specialists. All training sessions were supervised and instructed by certified fitness instructors.

InBody 270 (Biospace) was used as the body composition analyzer. It is a noninvasive bioelectrical impedance analysis device, which estimates body composition by transmitting a low-level electrical current through the body and measuring resistance. It provides accurate assessments of BMI, body fat percentage, obesity analysis, skeletal muscle mass, and visceral fat levels, serving as a comprehensive tool for evaluating body composition. The detailed procedures have been described previously [15].

In older adults, various components of functional fitness were evaluated using standardized tests. Aerobic endurance was assessed through the 2-minute step test, whereas upper and lower limb muscular endurance were measured using the arm curl test and chair stand test, respectively. Lower limb flexibility was evaluated with the sit-and-reach test, and shoulder flexibility was assessed using the back-scratch test. In addition, static balance was measured using the one-leg standing test, and dynamic balance was evaluated using the 8-foot up-and-go test. Detailed procedures, devices, and manipulation have been described previously [16]. These senior functional tests are cost-effective and convenient for assessing functional fitness in older adults. They are suitable for independently living individuals aged 60-90 years and above, encompassing numerous fitness levels, ranging from frailty to high fitness [17]. All certified assessors, independent from the research team, conducted the Senior Fitness Assessments and AFAscan fitness assessment in a single-blind manner.

The AFAscan leader facility was used to assess various components of physical fitness, including lower body explosiveness, core strength, flexibility, agility, balance, and endurance. It used gyroscopes (positioned on the chest and both thighs) and a force platform to precisely and objectively measure all physical performance parameters.

Physical activity levels were measured using the IPAQ, which evaluates walking, moderate-intensity, and vigorous-intensity activities over the past 7 days. The IPAQ has shown good test-retest reliability (Spearman ρ=0.76-0.94) and acceptable internal consistency (Cronbach α=0.80), with evidence of concurrent validity when compared against accelerometer data [18,19]. The Taiwanese version of the IPAQ was used, specifically the long-form Taiwan Physical Activity Survey Questionnaire, which consists of 26 items encompassing four domains: work, household activities, transportation, and leisure activities [20].

In addition, the Taiwanese version of the WHOQOL-BREF—a short version of the World Health Organization Quality of Life Assessment questionnaire—was used. This version consists of 28 items, encompassing 4 domains: physical health, psychological well-being, social relationships, and environmental factors. The WHOQOL-BREF has demonstrated acceptable internal consistency, with Cronbach α ranging from 0.66 to 0.84 across domains, and satisfactory construct validity when tested in different populations [21,22]. This shortened and adapted version serves as a valid alternative to the long-form WHOQOL questionnaire used in Taiwan [22].

The inclusion criteria were as follows: (1) older adults (aged 60‐80 years) with no musculoskeletal injuries in the past month and capable of participating in training activities, (2) completion of the Physical Activity Readiness Questionnaire (PAR-Q) and eligibility for moderate-intensity exercise intervention, and (3) ability to adhere to the exercise frequency required by the study and independently commute to the training site. The exclusion criteria were as follows: (1) individuals with musculoskeletal disorders, joint diseases, unstable cardiovascular conditions, or cognitive impairments, as well as those with psychiatric disorders based on questionnaire evaluation, and (2) individuals whose PAR-Q responses indicate potential risks or concerns.

Descriptive statistics and chi-square tests were used to analyze the anthropometric and sociodemographic characteristics of all participants. The normality of all dependent variables was evaluated using Kolmogorov-Smirnov tests to ensure the appropriate selection of parametric and nonparametric analyses. To examine significant between- and within-group differences, nonparametric methods (Wilcoxon signed-rank test and Mann-Whitney U test) and parametric methods (mixed 2-way analysis of variance, unpaired 2-tailed t tests, and paired t tests) were applied. Two participants withdrew before completing posttest assessments and thus provided only baseline data. Because no postintervention outcome data were available for these individuals, no imputation was applied, and analyses were performed using complete cases. Consequently, a total of 15 participants in the experimental group and 15 participants in the control group were included in the final statistical analysis. All statistical analyses were performed using SPSS software (version 22; IBM Corp), with statistical significance set at P<.05. Effect sizes (d) were calculated as Cohen d and interpreted according to conventional benchmarks: small (0.2), medium (0.5), and large (0.8) [23]. The 95% CIs were reported alongside all effect size estimates to enhance interpretability.

This study was reviewed and approved by the Fu Jen Catholic University Institutional Review Board (IRB # C112161) per ethical guidelines for human research. Before participation, all participants were informed about the study procedures and precautions, and written informed consent was obtained. All acquired data will be analyzed and disseminated in deidentified format to ensure participant confidentiality. Study participants will receive individualized physical fitness assessment reports and monetary compensation consisting of US $32.77 (exchange rate: US $1=30.5 TWD) for transportation expenses. To ensure participant safety, all training sessions were supervised by qualified specialists, who continuously monitored for potential and unexpected adverse events. Any adverse event would have been documented immediately, managed with appropriate first aid, and referred for medical evaluation if necessary.

The experimental procedure is illustrated in Figure 3. This study recruited 32 participants, with 16 allocated to each of the experimental and control groups. However, one participant from each group withdrew from the study because of personal reasons. Finally, 15 participants in both groups were analyzed for primary outcomes. The participants in the experimental group were paired and engaged in the training activities through alternating and cooperative interactions. Heart rate monitoring during the training indicated fluctuations characteristic of high-intensity interval training (Figure 2B). In addition, no significant differences were observed in sex, age, height, weight, systolic, diastolic pressure, and resting heart rate (t₂₈=0.065‐1.121; all P>.05) before the intervention (Table 2).

Pre- and postintervention results of the Physical Activity Scale indicated a significant main effect of time on total physical activity levels (Table 3). After the intervention, activity levels significantly increased in the experimental group, with significant within-group differences. Therefore, the mat exercise program effectively enhanced overall physical activity in older adults. For vigorous physical activity, the main effect of time (F_1,28=13.9; P=.001) and interaction effect (F_1,28=11.9; P=.002) were observed. The experimental group demonstrated a significant within-group increase in total metabolic equivalents (MET) (t₁₄=2.88; P=.01; mean difference [MD] 1266, 95% CI 325-2206; d=0.74) and vigorous activity (t₁₄=7.37; P<.001; MD 950, 95% CI 674-1227; d=1.90), indicating that the intervention effectively promoted physical activity. In addition, after the intervention, the experimental group demonstrated significantly higher vigorous activity levels than the control group (t₂₈=1.97; P=.04; MD 439, 95% CI 28-914; d=0.72), supporting the effectiveness of the exergame-based mat program. However, no significant main effects or interaction effects were observed for moderate-intensity activity, walking, or sedentary behavior (F_1,28=0.583‐1.889; P>.05). Thus, the intervention exerted a relatively lower impact on moderate- and low-intensity physical activities.

The WHOQOL-BREF assessment evaluated overall health and health satisfaction, along with four domains: physical health, psychological health, social relationships, and environment (Table 4). Concerning physical health, the experimental group demonstrated significant differences both within (P=.02) and between (P=.01; Hodges-Lehmann median difference=1.2; 95% CI 0.6-2.3; r=0.53) groups. Specifically, significant within-group improvements were observed in pain and discomfort (P=.02), dependence on medical aids (P=.002), mobility (P=.008), and daily activities (P=.002). In addition, daily activities (P=.002) and work capacity (P=.006) showed a significant between-group difference in the experimental group, compared with the control group. Concerning psychological health, the experimental group exhibited a significantly higher overall score than the control group (P=.02; Hodges-Lehmann median difference=2.7, 95% CI 1.3-4; r=0.52). Significant between-group differences were observed in self-esteem (P=.02) and negative feelings (P=.001), with negative feelings showing significant differences within (P=.02).

Concerning social relationships, the experimental group demonstrated significantly higher overall scores than the control group (P=.02; Hodges-Lehmann median difference 2, 95% CI 1-3; r=0.50). Furthermore, the experimental group exhibited significantly higher scores in social support than the control group (P=.003). In contrast, regarding the environment, no significant within- or between-group differences were observed in overall scores or individual items, indicating that the intervention did not substantially affect this aspect of quality of life.

Table 5 summarizes the results for systolic blood pressure, diastolic blood pressure, and heart rate. No significant interaction effects were observed between group and time (F_1,28=0.583‐1.889; all P>.05). In addition, no significant main effect of group was observed (F_1,28=1.505‐2.181; all P>.05), except for a significant main effect of time on diastolic blood pressure (F_1,28=4.350; P=.04). In the control group, systolic blood pressure and heart rate did not change, whereas diastolic blood pressure significantly reduced after the intervention. In the experimental group, systolic blood pressure, diastolic blood pressure, and heart rate showed no significant between- or within-group differences.

For muscular endurance of both upper and lower limbs, a significant interaction effect was observed between time and group (F_1,28=18.1, P<.001; F_1,28=10.91, P=.003, respectively). Specifically, upper limb muscular endurance in the experimental group was significantly higher in the posttest compared with the pretest (t₁₄=4.92; P<.001; MD 5.07, 95% CI 2.80-7.27; d=1.27), and significantly higher compared with the control group (t₂₈=2.92; P=.007; MD 5.33, 95% CI 1.59-9.07; d=1.06). Moreover, the gain score for upper limb endurance was significantly higher in the experimental group than in the control group (t₂₈=4.26; P<.001; MD 5.4, 95% CI 2.80-7.99; d=1.55). Similarly, lower limb muscular endurance was significantly higher in the experimental group after the intervention compared with before the intervention (t₁₄=3.24; P=.006; MD 3.2, 95% CI 1.08-5.31; d=1.55) and the control group (postintervention values; t₂₈=2.70; P=.01; MD 4.73, 95% CI 1.15-8.32; d=0.98). The gain score for lower limb endurance was significantly higher in the experimental group than in the control group (t₂₈=3.31; P=.003; MD 4.80, 95% CI 1.83-7.77; d=1.21).

No significant interaction effect was observed between time and group for upper limb flexibility (F_1,28=0.284; P=.59). No significant between- or within-group differences were observed for upper limb flexibility after the intervention. However, a significant interaction effect was observed for lower limb flexibility (F_1,28=20.71; P<.001). However, lower limb flexibility in the experimental group was significantly higher after the intervention than before the intervention (t₁₄=4.74; P<.001; MD 4.41, 95% CI 2.41-6.40; d=1.04). Postintervention values were significantly higher than those in the control group (t₂₈=2.85, P=.008, MD 6.47, 95% CI 1.82-11.1; d=1.04). In addition, the gain score for lower limb flexibility was significantly higher in the experimental group than in the control group (t₂₈=4.55; P<.001; MD 6.45, 95% CI 3.54-9.36; d=1.66).

Regarding balance performance, no significant interaction effect was observed for static balance between time and group (F_1,28=1.308; P=.29), with no significant between- or within-group differences. Nonetheless, a significant interaction effect was observed for dynamic balance (F_1,28=6.646; P=.02). Despite no significant within-group difference in the experimental group before and after the intervention, the postintervention time was significantly lower than that in the control group (t₂₈=−2.29; P=.03; MD -0.72, 95% CI −1.36 to −.07; d=0.84). Furthermore, the gain score for dynamic balance was significantly lower in the experimental group than in the control group (t₂₈=−2.58; P=.02; MD −0.77, 95% CI −1.38 to −0.16; d=0.94).

For aerobic endurance, a significant interaction effect was observed between time and group (F_1,28=14.027; P=.001). In the experimental group, postintervention aerobic endurance was significantly higher than preintervention endurance (t₁₄=4.58; P<.001; MD 12, 95% CI 6.38-17.6; d=1.18). Moreover, postintervention values were significantly higher than those in the control group (t₂₈=2.13; P=.04; MD 9.73, 95% CI 0.37-19.1; d=0.78). In addition, the gain score for aerobic endurance was significantly higher in the experimental group than in the control group (t₂₈=3.75; P=.001; MD 12.3, 95% CI 5.59-19.1; d=1.66).

No significant interaction effect was observed for explosiveness, along with no significant between- or within-group differences (F_1,28=0.169; P=.68) (Table 6). For core strength, a significant interaction effect was observed between time and group (F_1,28=14.027; P=.001). In the experimental group, postintervention core strength was significantly higher than preintervention core strength (t₁₄=3.77; P=.002; MD 8.14, 95% CI 3.47-12.8; d=1). Moreover, postintervention values were significantly higher than those in the control group (t₂₈=5.18; P<.001; MD 13.1, 95% CI 7.90-18.2; d=1.89). In addition, the gain score for core strength was significantly higher in the experimental group than in the control group (t₂₈=2.49; P=.02; MD 6.69, 95% CI 1.15-12.2; d=1.04).

Significant interaction effects were observed for flexibility, agility, balance, and endurance (F_1,28=5.27‐34.078; all P<.05). Postintervention fitness characteristics in flexibility (t₂₈=2.10; P=.05; MD 13.3, 95% CI 0.34-26.3; d=0.77), agility (t₂₈=3.93; P=.001; MD 32.6, 95% CI 15.6-49.6; d=1.44), balance (t₂₈=3.06; P=.005; MD 15.1, 95% CI 5.01-25.3; d=1.12), and endurance (t₂₈=5.34; P<.001; MD 27, 95% CI 16.6-37.4; d=1.95) were significantly improved in the experimental group, compared with the control group. Moreover, significant within-group differences were observed in the experimental group, with significantly higher gain scores than in the control group.

Regarding body composition (Table 7), no significant interaction effects were observed for body weight, BMI, skeletal muscle mass, body fat mass, or body fat percentage (F_1,28=0.179‐0.347; all P>.05) between the control and experimental groups. In addition, no significant differences were observed in gain scores between groups. After the intervention, the experimental group exhibited an increasing trend in lower limb muscle mass, whereas the control group exhibited a minor decline in lower limb muscle strength. However, these changes were not statistically significant in terms of gain scores. Therefore, the intervention exerted a limited impact on the overall body composition, segmental muscles, and fat distribution.

The magnitude of correlations was interpreted according to Cohen conventions: small (0.10-0.29), moderate (0.30-0.49), and large (≥0.50). Positive values indicate direct associations, while negative values indicate inverse associations. Pearson correlation analysis of physical fitness assessments in older adults (Table 8) suggested a significantly moderate positive correlation between upper and lower limb muscular endurance (r=0.502; P<.05). In addition, a significantly low positive correlation between upper limb muscular endurance and lower limb flexibility was observed (r=0.377; P<.01). However, no significant correlations were observed between upper limb muscular endurance and either static, dynamic balance, or aerobic endurance. Lower limb muscular endurance demonstrated a significantly low to moderate positive correlation with both cardiorespiratory endurance (r=0.488; P<.01) and lower limb flexibility (r=0.386; P<.01), whereas a significantly moderate negative correlation with dynamic balance (r=−0.6; P<.01) was observed. In addition, lower limb flexibility exhibited a significantly moderate positive correlation with static balance (r=0.402; P<.01). Therefore, the intervention effectively enhanced multiple aspects of physical fitness, underscoring the interplay among fitness components.

Integrating exergames, motion-sensing technologies, and sports-based innovations is essential for the future development of physical fitness interventions. Well-structured training programs not only promote exercise participation and overall health across diverse populations but also enhance the regularity and acceptability of fitness behaviors. Accordingly, the present 10-week mat exercise program was designed based on an exergame strategy, incorporating a variety of exercise equipment to achieve multiple training objectives under coach supervision. The sessions used a 2-person exergame-based mat system that featured turn-taking, cooperative, and competitive game-based activities. In terms of functional fitness, significant improvements were observed in muscular endurance, lower limb flexibility, static and dynamic balance, core strength, agility, and aerobic endurance. Moreover, physical activity levels and quality of life—particularly aspects related to physical health, psychological well-being, and social relationships—were also enhanced as a result of this novel exergame-based mat training intervention.

Traditional community-based exercise programs, including aerobic, resistance, balance, and stretching training, have been shown to improve daily physical activity and aerobic capacity in older adults [24]. However, for those with mild disabilities, such interventions primarily enhance specific fitness components such as lower limb strength and flexibility without significantly affecting overall activity levels [25]. Alternative approaches, such as stationary cycling-based high-intensity interval training (HIIT), effectively increase moderate-to-vigorous activity and energy expenditure [26]. Similarly, motion-based exergames like the Nintendo Wii and Xbox Kinect increase energy expenditure and improve upper limb endurance, physical activity, and psychological well-being, particularly among sedentary or mobility-limited older adults [27,28]. These findings suggest that program design, intervention duration, and participant health status critically shape outcomes. Building on this evidence, our exergame-based mat training program integrated exergaming elements, progressive intensity, HIIT mode, and group-based sessions to provide a comprehensive physical activity experience, yielding significant gains in vigorous activity and functional fitness.

Regarding quality of life, a study [29] on Kinect-based exergames reported significant improvements in physical and psychological health, aligning with our findings. Both the exercise mat program and Kinect training enhanced mobility, reduced pain, and improved daily functioning. Psychologically, the exercise mat training boosted self-esteem and reduced negative emotions. Multicomponent exercise training has also been shown to enhance executive functioning in older adults, contributing to improved quality of life [30]. In addition, exergaming programs significantly improve self-efficacy in older adults, particularly by fostering confidence in maintaining long-term exercise participation [31]. This underscores the role of gamification and social interaction in overcoming psychological barriers and building confidence in physical activity. A systematic review further confirmed the effectiveness of exergames in promoting mental health and socialization among older adults, though identifying the most effective types of exergames remains an area for future research [32]. The exergame-based mat training in this study was designed to achieve multiple fitness goals by incorporating enjoyable, electronic elements to evoke positive emotions. Group-based sessions fostered social interaction and reduced loneliness, while the progressive intensity design introduced appropriate challenges to enhance self-efficacy. Together, these factors contributed to improved physical function and psychological adaptability in older adults.

Our findings also align with meta-analyses demonstrating that active video games improve balance, lower limb strength, and cardiorespiratory fitness in older adults, though effects on body composition and flexibility remain limited [33,34]. In the present study, the exergame-based mat training yielded similar results, with significant improvements in balance, flexibility, muscular strength, and aerobic capacity, despite no significant changes in body composition. Older adults without a history of falls exhibited superior performance, highlighting the role of static balance, lower limb strength, and flexibility in reducing fall risk in this population [35]. Previous studies consistently highlight the critical role of lower limb strength in older adults. Greater lower limb strength has been linked to enhanced static and dynamic balance [36,37], greater functional independence and self-care ability [38], as well as higher quality of life [39]. Together, these findings emphasize that maintaining lower limb strength is fundamental not only for mobility and fall prevention but also for preserving autonomy and overall well-being in later life. Our findings suggest that older adults who participated in the exergame-based mat training demonstrated functional improvements across functional fitness indicators, which may have implications for fall prevention, independence, and overall quality of life.

Exergame interventions provide multifaceted benefits for older adults, including improvements in strength, balance, flexibility, and cardiorespiratory endurance [40]. Although our mat-based exergame training did not reduce body fat, it effectively enhanced upper limb and core strength, aerobic endurance, agility, and balance, thereby reducing fall risk, consistent with prior mat-based studies [41]. Similar to Bacha et al [42], who reported superior cardiorespiratory fitness gains with Kinect-based exergames compared to conventional therapy, our findings further underscore the potential of exergame-based approaches for improving aerobic capacity and overall function. Given evidence linking physical fitness to fall frequency [43], the observed improvements in balance and muscular strength highlight clinically meaningful outcomes. Moreover, parallels with HIIT research indicate that the present exergame training program may facilitate favorable physiological adaptations, such as improved metabolic and cardiovascular health, while mitigating age-related decline [44-46].

This study has several limitations. Long-term quantitative tracking using accelerometry was not implemented, and participants’ dietary caloric intake was neither controlled nor documented, limiting adjustments for physical activity and body composition effects. Upper limb flexibility requires further exploration through targeted exercises and equipment design. In addition, the benefits observed cannot be conclusively attributed to the program, and the lack of an attention-matched intervention for the control group may have introduced bias in self-reported outcomes, such as physical activity and quality of life. However, the inclusion of objective measures (eg, mat-based kinetics and standardized timed tests) provides supportive evidence for the observed effects. As an exploratory trial, the exergame-based training program demonstrated diverse benefits in physical fitness, physical activity, and quality of life. Future research should refine training intensity and modes, include appropriate comparator groups, and further investigate the program’s physiological and therapeutic impacts to enhance its applicability for diverse older adult populations.

The implementation of exergame-based mat training positively influenced functional fitness in older adults. These gains are likely to translate into improved performance of daily physical activities, including enhanced walking capacity, prolonged standing tolerance, greater movement flexibility, and increased independence in activities of daily living. Beyond physical benefits, participation in exergame-based mat training positively influenced psychological well-being by boosting self-confidence and self-esteem while promoting greater social participation. In addition, the integration of group-based exergame and interactive experiences enhanced the willingness and acceptance of older adults to engage in training. These findings highlight the feasibility and benefits of the program in improving functional fitness and overall quality of life among older adults.