Accepted for/Published in: JMIR Formative Research
Date Submitted: Aug 10, 2025
Open Peer Review Period: Aug 10, 2025 - Aug 19, 2025
Date Accepted: May 28, 2026
(closed for review but you can still tweet)
Preliminary Diagnostic Accuracy of BOKE STARS, a Bimodal Continuous Performance Test, for ADHD Assessment in Chinese Children: A Single-Center Case-Control Study
ABSTRACT
Background:
ADHD diagnosis in China demands culturally adapted, objective screening tools
Objective:
In this study, we examined the application of continuous performance tests as a diagnostic tool for children with ADHD.
Methods:
A total of 100 children (aged 6–12 years; 50 with ADHD and 50 controls) were assessed with BOKE STARS from January to May 2024. The BOKE STARS app was used as the continuous performance test tool for all participants. For group comparisons, Mann-Whitney U tests or independent samples t-tests were performed. ADHD subtype differences were determined by conducting the Kruskal-Wallis test or one-way ANOVA. Relationships between SNAP-IV scores and BOKE STARS indices were analyzed using Spearman’s rank correlation. Diagnostic accuracy was evaluated via receiver operating characteristic (ROC) curve analysis, with optimal cutoff values determined by maximizing Youden’s index.
Results:
Compared to the control group, the ADHD group presented significantly impaired performance across all BOKE STARS indices (omission, commission, reaction time, RTV, etc.) (RTV: ADHD: 80.08 ±13.45, control: 103.32 ±11.50, p < 0.01). The ROC curves yielded an AUC of 0.962 for the total score and 0.919 for the RTV (sensitivity 84%, specificity 88%). No significant differences were found among the ADHD subtypes. Most BOKE STARS indices correlated negatively with the SNAP-IV inattention and hyperactivity dimensions (ρ = –0.230 to –0.390, p < 0.05).
Conclusions:
The dual-modality design of BOKE STARS, optimized for Chinese children, achieved high diagnostic accuracy (AUC ≥ 0.92). It is recommended as a sensitive, specific screening tool along with clinical evaluation, with machine learning integration being assessed in larger cohorts.
Citation
The author of this paper has made a PDF available, but requires the user to login, or create an account.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.