Skip to content

School exercise trial found no clear reduction in suicidal thoughts or behavior

No overall self-injury benefit was detected; a subgroup with low starting activity had a lower rate.

By 100HP editorialAbstract-based explanation checked

Based on the published abstract. The full paper may contain additional methods, results and limitations.

The 30-second takeaway

In this school-based trial, aerobic exercise did not show an overall reduction in suicidal thoughts, suicidal behavior or self-injury without suicidal intent. The low-activity subgroup had a lower self-injury rate with exercise. That finding does not establish that effects differed between activity subgroups, or that all adolescents would benefit.

Keep in mind

These were secondary outcomes. The abstract does not report event counts, whether subgroup analyses were prespecified, or a test comparing effects across activity subgroups.

THE NUMBERS, WITH CONTEXT

What researchers found

Incidence rate ratio 0.49 (95% CI 0.25–0.97)

Self-injury without suicidal intent

Among adolescents with low starting activity, exercise versus psychoeducation over 12 months: the estimated self-injury rate was 0.49 times the control rate; the confidence interval spans 0.25–0.97 times that rate. This is a relative rate comparison, not an absolute reduction.

J Psychiatr Res, 2026 · Original source ↓

How groups were compared

Groups of students were assigned to exercise or psychoeducation. The exercise program began with supervised school sessions and continued without supervision; analyses kept participants in their assigned groups.

Reading the overall result

Adjusted analyses found no clear self-injury benefit across all participants or by initial subclinical depressive symptoms. A null result does not prove the intervention has no effect or establish equivalence.

CHECK THE ORIGINAL

The original publication

Can exercise reduce the risk of suicidal ideation and behaviors in youth at school? A 12-month cluster randomized control trial.

Werneck AO, Tian X, Lu W et al.
J Psychiatr Res · 2026

PubMed ID
41448028
Record checked

AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.