Changes in Depressive Symptoms Associated With Use of SPARX, a Game-Based Cognitive Behavioral Therapy Application: A Real-World Repeated Measures Observational Study in Japan (Preprint)

BACKGROUND Depressive symptoms are common across populations worldwide. In Japan, a nationwide survey of more than 8,000 adults reported that approximately one-fifth of respondents experienced clinically significant anxiety or depressive symptoms. Depressive symptoms are associated with impaired work functioning, reduced quality of life, and increased suicide risk, yet access to evidence-based psychological treatments remains limited. Mobile health (mHealth) interventions have the potential to increase access to psychological support by enabling self-guided use independent of time and location. Computerized cognitive behavioral therapy (cCBT) has been shown to reduce depressive symptoms in adolescents and adults; however, engagement and attrition remain key challenges in unguided digital interventions. SPARX is a game-based CBT program originally developed for adolescents, consisting of seven modules delivered through an interactive fantasy game. While previous randomized controlled trials have demonstrated benefits of SPARX under controlled conditions, less is known about its use and outcomes when delivered in real-world settings, particularly among adult users in Japan. OBJECTIVE This study aimed to examine changes in depressive symptoms associated with use of SPARX, a game-based cognitive behavioral therapy application, among Japanese adults in a real-world setting, and to explore whether symptom trajectories differed by age or baseline depressive symptom severity. METHODS This was a real-world repeated measures observational study of individuals who used SPARX in Japan between May 2016 and July 2021. In total, 5,374 individuals registered to use SPARX, and 3,367 users met the following eligibility criteria: (1) aged 20–69 years at the time of registration, and (2) having completed at least Module 1 of SPARX. Each user purchased the Japanese version of SPARX from the Apple App Store or Android Google Play Store. The primary outcome measure was the Japanese version of the Patient Health Questionnaire-9. Linear mixed model (LMM) analysis was used to examine changes in depressive symptom scores across program modules, comparing Module 4 and Module 7 with Module 1. RESULTS Overall, 3,367 individuals aged 20–69 years (2,925 men and 410 women; mean age = 34.28 years, SD = 9.28) completed Module 1. Linear mixed model analyses showed significant reductions in depressive symptom scores over program use among those who progressed through later modules. Reductions were observed at Module 4 (n = 1,356; estimate = −2.87, t = −15.40, p < .05, d = 0.69, 95% CI [0.53, 0.84]) and at Module 7 (n = 844; estimate = −4.57, t = −19.23, p < .05, d = 1.06, 95% CI [0.86, 1.27]). CONCLUSIONS In this large real-world repeated measures observational study without a control group, use of SPARX among Japanese adults was associated with improvements in self-reported depressive symptom scores among users who progressed through later modules. Although causal inferences regarding effectiveness cannot be made, these findings provide evidence on usage patterns and symptom trajectories of a game-based CBT program originally developed for adolescents when delivered to an adult population in a naturalistic setting.