The investigators searched seven databases through 1 June 2026 and updated the search on 20 August 2026, finding no additional eligible review. They included reviews addressing exercise and emotional-behavioral outcomes among participants younger than 18 years with ADHD. Screening and data extraction were independently duplicated, with disagreements adjudicated by a third reviewer.

Methodological confidence was evaluated with AMSTAR 2, while a planned ROBIS assessment of bias in systematic reviews was not undertaken. The researchers retained only verifiable, outcome-specific certainty ratings and did not re-pool estimates. Overlap among the reviews was quantified with corrected covered area, or CCA.

It was 7.46% across all included reviews and 8.79% for the ADHD-specific core evidence base. Outcome-specific overlap was higher: 23.44% for anxiety, 26.67% for depression and 12.64% for social functioning. Across overlapping reviews, reported standardized estimates ranged from −0.66 to −0.22 for anxiety, −1.67 to −0.42 for depression and −0.59 to −0.27 for social functioning.

These were estimates reported by the source reviews, not new pooled effects calculated by the umbrella review. To address highly overlapping constructs, the researchers established a post hoc hierarchy that selected one review for each relevant area. Sensitivity analyses examined the review by Zhou and colleagues for anxiety and depression and the review by Seiffer and colleagues for social functioning.

The selected reviews reported standardized effects of −0.43 for anxiety, with a 95% confidence interval of −0.84 to −0.01; −0.42 for depression, with a 95% confidence interval of −0.71 to −0.13; and −0.46 for social functioning, with a 95% confidence interval of −0.85 to −0.08. The Zhou review had low methodological confidence, while the Seiffer review had moderate confidence. Certainty could not be assessed for the anxiety or depression evidence and was low for the social-functioning evidence.

A bias-adjusted analysis of social functioning yielded a posterior mean of 0.146 and a 95% credible interval of −0.509 to 0.820. The estimate slightly favored the comparator but did not establish a direction of effect. No dedicated synthesis of adverse events was reported, leaving the review unable to provide a systematic account of harms.

Overall, the review-level estimates generally favored exercise, but methodological limitations, overlap and uncertainty prevented firm conclusions. The findings do not support outcome-, exercise-modality- or dose-specific clinical recommendations, and general physical-activity guidance is distinct from the outcome-specific evidence evaluated in this synthesis.

Subject of Research: Umbrella review of exercise interventions and emotional-behavioral outcomes in youth with ADHD

Article Title: Exercise-based interventions for anxiety and emotional-behavioral outcomes in youth with ADHD: an umbrella review and evidence gap map

Article References: Hu, Y., Ma, X., Sun, M., Chen, S., & Liang, W. (2026). Exercise-based interventions for anxiety and emotional-behavioral outcomes in youth with ADHD: an umbrella review and evidence gap map. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08776-6

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08776-6

Keywords: ADHD, youth mental health, exercise, anxiety, depression, social functioning, umbrella review, evidence certainty, research overlap, systematic reviews, Exercise-based, interventions