This study calculated sample size estimates from a previous study [3] using G*Power (version 3.1.9.2; Universität Düsseldorf) with a Cohen f effect size of 0.7, an α of .05, and a power of 0.8. The calculated sample size was 54 plus 6 for 10% dropout, so a final sample size of 60 participants (30 participants for each group) was needed.

This study was a single-center, assessor-blinded, randomized controlled trial. The participants were matched by age and gender and randomly assigned to either the AVG group (n=30) or the therapeutic exercise program for the control group (CG, n=30) using a computer-generated program. The CONSORT (Consolidated Standards of Reporting Trials) flow diagram of this study is shown in Figure 1. Sixty nonathletic, typically developing children with CAI aged between 7 and 12 years and with BMI between the 5th and 85th percentiles on the US Centers for Disease Control and Prevention growth chart [31] volunteered for the study. Inclusion criteria were (1) a score equal to or less than 25 on the Cumberland Ankle Instability Tool–Youth (CAITY), Thai version [32], (2) recurrent ankle sprain more than 3 months before enrollment, (3) a feeling of giving way at least twice a year, and (4) unilateral ankle sprain at least 1 year before enrollment. Participants were excluded if they had a history of either (1) ankle fracture, ankle surgery, or neurological disorder; (2) health problems such as uncontrolled seizure, asthma, severe heart disease, hearing problems, or visual problems that could not be corrected by using a lens; and (3) current participation in another rehabilitation program for the ankle joint.

The study conformed to the Declaration of Helsinki and was approved by the local Institutional Review Board Ethics Committee (MU-CIRB 2020/387.2311). After an explanation of the exercise protocol, testing procedures, and benefits and possible risks of the study, the informed consent form was signed by the parents or guardians and the assent form was signed by the participants.

The experimental group performed the AVG exercise program, whereas the CG performed the therapeutic exercise for the CAI program [33]. Both programs were scheduled for 3 sessions per week and 30 minutes per session for 4 weeks at home (a total of 12 exercise sessions). The 30-minute CG and AVG exercise programs consisted of a 5-minute active warm-up, a 20-minute main exercise, and a 5-minute cooldown. To assess compliance with the study protocol, the researchers followed up with participants once a week with a phone call. Balance ability tests and ankle instability questionnaires were evaluated at week 0 and week 4. The Intrinsic Motivation Level (IML) questionnaire was assessed each week over the course of the 4-week program [34].

The design of the exercise program used in this study was adapted from the Star Excursion Balance Test (SEBT) [28]. Briefly, participants were asked to stand on one leg and then slowly reach with the other leg in different directions. This involved reaching in the anterior posteromedial and posterolateral directions as far as possible, touching down along the guideline, and then returning to the starting position (Figure 2). Each direction was completed as a separate trial. Participants were asked to perform 10 trials in each direction for 2 rounds with a 5-minute rest between rounds (for a total main exercise duration of around 20 minutes). Participants used their affected limbs first and then changed to the other limb. The program was designed to gradually increase the reaching distance of the leg (in centimeters) every week (by 50%, 60%, 70%, and 80% of the baseline value in weeks 1, 2, 3, and 4, respectively).

The AVG program (Figure 3) comprised 2 selected video games (ie, Catching Fish and Russian Block) from Stepmania Coilmix, a cross-platform rhythm video game and engine with a wireless dance mat [30]. In Catching Fish, participants stood in the center of the mat before starting the game, then moved their feet either in the left or right direction in response to the game (ie, they pressed the corresponding arrows to catch the yellow or red fish). Participants had to continue playing the game as long as possible, with 4 chances to obtain a higher score. Similarly, for the Russian Block game, participants stood in the center of the mat before starting, then moved their feet in the left or right direction to change and manipulate the shape of the falling Tetromino piece shown on screen. To challenge the players, the piece will be moved faster, and it will become more difficult to correct the position. By the end of the game, the total score is shown.

The total exercise duration for the 2 games was approximately 20 minutes. If one of the games ended before 10 minutes, participants were required to start playing again from the beginning. For convenience and to promote adherence, the AVG devices were set up in the participants’ homes. The researchers provided instructions and demonstrations until the parents and children could operate the device correctly with no assistance.

Shapiro-Wilk tests were carried out to examine the normality of the data, and all data were found to be suitable for parametric testing. In addition, the sphericity of data was considered using the Mauchly test of sphericity; where the sphericity assumption was violated, Greenhouse-Geisser corrections were applied. Independent 2-tailed t tests were performed to compare the means and differences of each dependent variable at the baseline assessment. A 2-factor mixed model ANOVA was used to evaluate the effect of treatment group and time for the SLST, functional balance and strength test, and FAAM questionnaire. A repeated measure ANOVA was used for the IMI questionnaire. Any interactions between group and time were revealed using post hoc 2-tailed paired t tests with a Bonferroni correction to determine any differences within each group between the time points. The observed effect size was expressed as partial η squared (ηp²), with values of 0.1-0.29, 0.3-0.49, and >0.50 representing a small, medium, and large effect size, respectively [42]. The change score was calculated from the difference between data from week 4 minus data from week 0 using independent 2-tailed t tests. A P<.05 was considered statistically significant.

The major finding of this study provides evidence that a 4-week AVG home-based exercise training program can have a positive and significant therapeutic effect on children with CAI as determined by improving static balance, foot and ankle ability, and key measures of intrinsic motivation (ie, interest/enjoyment, pressure/tension, and value/usefulness).

For balance control, after 4 weeks of the training program, the AVG group could perform the SLST with EO on a stable surface as well as on a foam surface better than those who performed the traditional CAI program. When playing video games, children learn how to act and respond to the game scenario promptly, which is considered to enhance sensorimotor learning after training [43]. Moreover, the improvement of balance following the AVG training could be related to the movement strategies inherent in the games. The games provide visual feedback via a variety of visual-perceptual processing challenges and prompt lower-extremity muscle responses to control posture and balance. The results of this study support previous findings that have demonstrated the effectiveness of using visual feedback tools as a therapeutic training modality [28,44-46]. For example, Kim and Heo [44] reported the benefits of 4 weeks of balance exercise using the Wii Fit Plus program (30 minutes, 3 times a week) for improving static balance in people with functional ankle instability. Moreover, Fitzgerald et al [28] found that a therapeutic exergaming system showed an improvement in postural stability when compared to a group doing similar balance training without the game system. Therefore, this study offers further support for using an AVG visual feedback training program as an effective modality for improving the balance of children with CAI.

The results of this study showed no significant difference in the subtest of the BOT-2 for functional balance, measured by walking forward on a line, and for leg muscle strength, measured by a standing long jump, in the AVG group and the CG after 4 weeks of training. This might be caused by either the movement patterns of the AVGs (moving feet to the left or right) and the traditional CAI program not being potent enough or the period of the exercise program (4 weeks) being too short to improve functional balance and lower-extremity strength.

This study revealed that foot and ankle ability measured by the FAAM-ADL in the AVG group improved after the 4-week training, and interestingly, it was greater than the CG. This result is congruent with the results of Punt et al [47], who found an improvement in foot and ankle ability during a 6-week follow-up of users of the Wii Fit game (twice per week, for 30 minutes per session) in lateral ankle sprain patients. Perceived instability of the foot and ankle has been reported to be associated with decreased health-related quality of life in individuals with CAI [43]. Thus, the exercise program with AVG could be used as exercise therapy to increase foot and ankle ability and might provide a better quality of life in children with CAI.

The participants’ subjective perception of the program was assessed every week. Notably, in week 2, the perception of interest/enjoyment was increased in the AVG group, whereas the perception of value/usefulness was decreased in the CG. Interestingly, the perception of pressure/tension was increased in the CG and greater than the AVG group at week 4. These results are consistent with previous findings [28,45,48] that reported that a video game–based rehabilitation approach could provide a greater intrinsic motivation level to engage in the activity for a longer period of time when compared to a group doing similar balance training. These results might explain that increased interest and enjoyment may have occurred due to the achievable challenges provided by the games (ie, Catching Fish and Russian Block) and the subsequent feelings of achievement when each game level was completed. By contrast, in the traditional CAI program, there was no feedback mechanism during the exercises. The participants simply had to count the number of repetitions and sets performed, and this might have contributed to the reduction in value/usefulness levels, as well as the pressure/tension level increasing during the program. It is noteworthy that the results of this study illustrated that effort/importance slightly declined after 2 weeks and was still declining in week 4. It may be possible that the children became accustomed to the exercise program, as they had to repeat the same exercise again and again to achieve the aim of the intervention. This result is consistent with the findings of previous research that reported nonadherence and lack of motivation in long-duration training in home-based therapeutic practices [15,16].

Furthermore, the protocol and training intervention using the AVG program has been shown to be feasible, as indicated by the retention and adherence of participants. The safety of the program is evidenced by the lack of adverse events reported. This suggests that the design and progression of the training had an appropriate intensity and the interactive video game did not present any harm. Therefore, the above results suggest that AVG programs provide an engaging and interactive environment that results in less fatigue, more relaxation, and enhanced adherence throughout the rehabilitation period.

There are some limitations to this study. First, observations were not obtained in a controlled laboratory environment. However, the researchers gave instructions, set up the exercise environment at the participants’ homes, and made telephone calls to give reminders to the participants. Second, the relatively small sample size may not be representative of the characteristics of the broader population of children with CAI. Third, follow-up measurements are required so that the longitudinal effects of the AVG program can be better understood. Lastly, a 4-week program is a relatively short intervention period. Thus, further studies would appear warranted to determine the effects of a longer-duration AVG training program.

This study found that children with CAI showed significant improvements in single-leg standing, foot, and ankle ability measured by the FAAM-ADL and intrinsic motivation in the interest/enjoyment, pressure/tension, and value/usefulness dimensions after following a home-based exercise program with AVGs. Furthermore, it was discovered that playing video games while exercising comprised a visuomotor feedback task that was lacking in the therapeutic exercise program. Thus, visual feedback while exercising with AVGs may be an effective way to improve balance ability, especially in the EO condition. This highlights that AVGs focused on balance exercises can be beneficial and may be an effective approach for improving balance and enhancing positive motivation in children with CAI who require long-term rehabilitation sessions.