This randomized prospective study lasted 4 weeks. The participants were voluntarily recruited and divided into an intervention and a control group. Taiwan’s Ministry of Education has formulated health-related physical fitness standards for adolescents aged 7 to 23 years [19]. Its safety and reliability have been verified [19], and therefore, we selected a 1600-meter running test. Before the start of the study, the participants were provided with a comprehensive explanation of the research content, and those who were willing to participate provided written informed consent. In addition, all the data were collected, analyzed, and stored anonymously. The study was approved by the Trial Committee of the Tri-Service General Hospital in Taiwan (approval number: C202005175; trial registration number: NCT05227040).

The study period was from January to April 2021. The inclusion criteria were as follows: (1) students who were over 20 years old and did not have chronic diseases; (2) students who were able to complete the 1600-meter outdoor run before and after the intervention; (3) students who understood and agreed with the research purpose and signed the consent form; and (4) students who used wearable devices correctly.

This randomized prospective study was conducted with 1 experimental group and 1 control group. A total of 95 participants were recruited for this study. However, 12 participants were excluded from the pretest because their CFI could not be calculated (Based on the Zephyr BioPatch Monitoring Device for Human Performance manual, the values of the HR confidence value lower than 20% indicated data were unreliable, and they were automatically deleted) [20]. Randomization was conducted using the simple random sampling method. After 4 weeks, during the posttest running test, the CFI of 3 participants could not be fully calculated, and hence, these participants were excluded. Finally, a total of 80 participants in the intervention group and control group completed the study (Figure 1).

Participants in the RFA group were required to exercise for 30 minutes 3 times per week (in adventure mode) (Figure 2) for 4 weeks. The initial exercise intensity was set according to the instructions given by the virtual coach during the first game and was gradually adjusted according to the game instructions. The research team continued to track the RFA group subjects and encouraged the completion of 4 weeks of physical activity training.

In contrast, no intervention was introduced to the control group participants, who maintained their regular habits during the study period.

Before the pretest and posttest, the participants’ basic demographic information, including age, sex, height, weight, neck circumference, waist circumference, smoking status (yes or no), alcohol consumption status (yes: 2 to 3 times a week or at least once a week; or no), milk intake (yes or no), whether the participants consumed 3 servings of vegetables and 2 servings of fruit per week, exercise habits (4 to 6 times a week, 1 to 3 times a week, or none), sleep quality (CPSQI), and mood disorder (BSRS-5 score), was collected.

Prior to the 1600-meter outdoor run, participants were required to wear the Zephyr BioPatch loaded with the Zephyr BioHarness 3.0 module. In addition, the Ambu BlueSensor T [32] was attached to the participants’ apical area (Figure 3).

After the participants completed the 1600-meter run, the BioHarness 3.0 module was removed from the BioPatch, and the physiological data were extracted. The data, which were binned per second, were then calculated and saved in Excel sheets (Microsoft Corporation). Due to the large amount of data, the method described by Hsaio et al was adopted for data processing [16]. Subsequently, a member of the research team further reviewed the physiological data collected by the BioHarness 3.0 module and calculated the CFI and CMPB values. CFI is an established method for monitoring heart function and workload during physical activity [16]. Its advantages include being noninvasive, wireless, and the ability for real-time monitoring through mobile devices [16].

The definitions of the BioHarness 3.0 parameters are listed below [16,18]:

1. CFI peak acceleration (CFI_PA): It was the peak acceleration per second divided by the number of heartbeats per second, both of which were estimated and measured by BioHarness 3.0.

2. Heart rate: It is defined as the number of heartbeats per second measured by the sensors during exercise. With BioHarness 3.0, the heart rate data were recorded in beats per minute (bpm), and the result was the average value of the data over 1 second.

3. CFI_PA while walking (CFI_PA_walk): This was the peak acceleration per second divided by the number of heartbeats per second while the participant was walking.

4. CMPB while walking (CMPB_walk): It was the movement in centimeters per second divided by the number of heartbeats per second while the participant was walking.

5. CFI_PA while running (CFI_PA_run): This was the peak acceleration per second divided by the number of heartbeats per second while the participant was running.

6. CMPB while running (CMPB_run): This was the average value of the CMPB_run calculated using the start of the run to the end of the run (movement in centimeters per second divided by the number of heartbeats per second while the participant is running).

The Windows version of the SPSS 28.0 software (IBM Corp) was used for the statistical analysis. According to descriptive statistics, categorical data are expressed in numbers and percentages, whereas continuous variables are expressed as averages and SDs. We conducted independent sample t tests and chi-square tests to confirm group differences.

Subsequently, the paired t test was conducted to analyze the differences between the completion time for the 1600-meter run in seconds, CFI, CMPB, CPSQI, and mood disorders between the pretest and posttest. We determined that the data may not be normally distributed, and therefore, we additionally used the Wilcoxon signed rank test for analysis.

Alternatively, generalized estimating equations (GEEs) were adopted to predict important factors influencing the running time. The significance level was set at 2-tailed P<.05.

The study was completed by 80 participants aged 20 to 36 years (mean 23.2, SD 2.96 years), and both groups were composed of 40 individuals. The RFA group comprised 16 males and 24 females, with an average age of 23.75 (SD 3.58) years; the control group included 30 males and 10 females, with an average age of 22.65 (SD 2.08) years. None of the participants were smokers; 51 participants (64%) exercised 1 to 3 times a week.

In addition, there were no significant differences in the participants’ age, BMI, neck circumference, waist circumference, smoking status, and alcohol consumption status; there were also no differences regarding whether they drank milk, consumed 3 servings of vegetables and 2 servings of fruit per week, and the exercise habits between the RFA and the control groups. The detailed demographics of the participants are listed in Table 1.

We used linear regression to select the variables with a significance level at P<.05. A prediction model was established using GEEs, and the BioHarness 3.0 parameters were introduced to identify factors that affected the time taken by the 80 participants to complete the run. The explanatory power of the model was 66.3% (R²=0.663). The results indicated that sex, exercise habits, alcohol consumption status, CMPB_ run, age, and BMI were important factors that affected the time taken for completing the run. If variables such as the group, sex, age, sleep quality, mood disorder, BMI, neck circumference, and waist circumference were controlled, those who exercised 4 to 6 times per week took 48.02 seconds less to complete the run (95% CI –85.65 to –10.38), and compared to nondrinkers, those who consumed alcohol 4 to 6 times per week took 45.83 seconds more to complete the run (95% CI 6.64 to 85.01). Furthermore, each additional unit of CMPB_ run corresponded to a decrease of 0.89 seconds in run time (95% CI –1.37 to –0.4) (Table 3).

To our knowledge, this is the first study assessing the effect of a Nintendo RFA exergame on running completion time of adults after an intervention. As a study that incorporated a popular exergame (RFA) into the intervention implemented among students, in addition to assessing physical fitness and changes in the time it took to complete a run, this study also investigated the changes in the participants’ sleep quality and psychological factors related to mood disorder.

In the RFA group, after 4 weeks of intervention, the completion time for the 1600-meter run decreased significantly for male and female participants, which was consistent with the findings of a previous study [11]. For example, a meta-analysis of studies conducted with children and adolescents found that previous exergames, such as Wii Fit and Dance Revolution, demonstrated similar physiological benefits [11].

Another 3-group randomized controlled trial of obese teenagers aged between 15 and 19 years reported that those who were encouraged to cooperatively play Wii Fit for 40 to 60 minutes per day lost substantially more weight and had an elevated self-identity [35,36]. In addition, similar exergames on Xbox Kinect (Microsoft Corp) exhibited a positive effect on physical fitness and emotional cognition [37] and were proven to enhance the dynamic balance of the lower limbs [38]. Both these benefits were attributed to the fact that Xbox Kinect required users to actively move in a given space by making the tasks dynamic [39] and introduced training programs that promoted visual feedback and dynamic balance [40].

Furthermore, a study provided preliminary evidence that Wii Fit could encourage people to exercise, thereby alleviating fatigue and reducing their body weight, waist circumference, anxiety level, and overall pain intensity [41]. Among traditional training programs, high-intensity interval training and similar methods use short-term high-intensity exercises to increase cardiorespiratory fitness and maximum oxygen consumption [42]. Moreover, a relatively short exercise time can effectively improve aerobic metabolism [43,44].

However, it remains unknown whether individuals can sustain such exercises over an extended period. In contrast, by combining exercise with digital games, emerging exergames can integrate physical activities with cognitive tasks and facilitate players to engage in physical activity according to visual, audio, and sensational feedback [45]. In addition, its user stickiness is strong owing to the introduction of multiple levels. Although this study did not include elderly individuals as part of the study population, similar research indicated that their experience with exergames was positive [46]. Exergames could also be used for rehabilitation purposes, as a study suggested that rehabilitation through Wii Fit achieved satisfactory therapeutic effects [47]. A recent study that adopted RFA as an intervention found that after 8 weeks, the chronic low back pain of participants was effectively alleviated [8]. The high level of movement they provide and the reduced cost of interactive modern technology have made exergames increasingly popular [48]; therefore, their application in health promotion is considered a future trend [49].

The existing studies on the effects of exergames on social and psychological factors, such as sleep quality and mood disorders, are somewhat limited. Our research indicated no significant changes in sleep quality after the intervention. The CPSQI adopted in this study is a psychometrically reliable measure of sleep quality and sleep disorders in patients with primary insomnia [25]. Although a study found that 6 weeks of Xbox Kinect training could improve the sleep quality of elderly individuals and alleviate their anxiety symptoms [50], this study did not reach a similar conclusion.

Alternatively, the BSRS-5 scale used in this study to measure mood disorders can be used to assess psychological symptoms, including anxiety, depression, hostility, interpersonal sensitivity, and difficulty falling asleep [26,28,51]. The modified BSRS-5 was found to rapidly detect mood disorders in a previous Taiwanese study and showed high validity and reliability among the general public as well as patients with mental or medical diseases [26]. In this study, although the BSRS-5 scores of males (1.81 to 3.31) and females (3.17 to 4.54) in the RFA group increased considerably after the intervention, they did not reach the category of mild mood disorder (BSRS-5 score>6). At present, no studies have used the BSRS-5 to explore the effectiveness of exergames. Although a study suggested that exergames could alleviate anxiety symptoms [41], a different questionnaire was used.

Therefore, more relevant research may be required to provide further evidence. The increased mood disorders in this study were likely because the pretest was performed immediately before the Lunar New Year Festival, whereas the posttest was performed after the holiday and before the midterm examinations, during which the participants were possibly more stressed. Some researchers also pointed out that during the COVID-19 pandemic, even low-threshold and mild psychological reactions could provoke intense mood disorder reactions [52].

This study proposed a new index, CMPB, based on the original CFI. CMPB refers to the distance (in cm) the body can move during each heartbeat, which makes it easier for the public to understand. Our study showed that sex, exercise habits, alcohol consumption status, CMPB_run, age, and BMI were important factors that affected an individual’s performance in the 1600-meter run. A previous study found that the older the person, the longer the run completion time [16], which was consistent with our findings. In addition, our study included exercise habits as a variable and found that compared to those who did not have an exercise habit, participants who regularly exercised 4 to 6 times a week required substantially less time to complete the 1600-meter run, which was consistent with the result of another study [53].

Moreover, we found that compared to nondrinkers, participants who consumed alcohol 2 to 3 times a week took significantly longer to complete the run, which was consistent with the findings of a previous study that focused on the correlation between alcohol consumption and exercise [54,55]. Furthermore, our study showed that each additional unit increase in CMPB_run reduced the time required to complete the run, and the difference was statistically significant. This was in line with the results of a CFI study [16], which indicated that CMPB could provide good predictions similar to those of CFI. However, further research is required to support this conclusion.

Finally, our study used the novel Zephyr BioPatch device. As a result, the participants were no longer required to wear a strap on the lower edge of the chest for the Zephyr BioHarness 3.0 module to take measurements, which previously prevented women from participating [16]. Therefore, the CMPB indicator allowed measurements to be taken among female participants and thus identify sex-specific differences.

This study had several strengths. First, the study found that RFA provided an opportunity for indoor exercise during the COVID-19 pandemic [56]. In addition, users’ constant engagement in training and activities could be encouraged by different elements in the game. Second, aside from running performance, this study also investigated the effect of RFA on sleep quality and mood disorders. Third, a new index, CMPB, was proposed based on the original CFI. Hence, the CMPB allows for easier comprehension and shares the same important factors in predicting the time required to complete the run.

However, this study had some limitations. First, the participants were limited to military students (college, master, and doctoral students) in northern Taiwan. Second, the study did not explore potential interfering elements, such as lifestyle habits and environmental factors. In addition, healthy participants may have contributed to a plateau effect, as their performance was already close to an optimal level [14].

This study found that training using RFA could maintain or improve users’ physical fitness. Therefore, RFA provides a good solution for people who engage in distance learning for a prolonged period or those who do not have sufficient time for exercise. Exergames not only serve as an alternative for exercising at home but also show the potential to evolve with societal changes.