Design: Systematic review and meta-analysis of randomised trials
Population: 15 trials, 1,041 adults with lung cancer
Primary outcome: Depressive symptom scores
Effect: Hedges' g −0.75 (95% CI −1.18 to −0.32); I² 87%; very low certainty

This meta-analysis, published on 19 September, pooled 15 randomised trials with 1,041 adults with lung cancer comparing structured exercise with usual care or other rehabilitation.

Exercise lowered depressive symptom scores on average (Hedges' g −0.75), but heterogeneity was high (I² 87%) and the prediction interval crossed zero. Removing two influential trials reduced the effect to −0.49 and removed the heterogeneity. The authors rated certainty very low, and the trials measured symptom scores, not major depressive disorder.

For liaison psychiatry and oncology teams, it supports offering exercise as part of supportive care for low mood, not as a replacement for treating a depressive episode.

Screen for depression in lung cancer; symptom scores are not a diagnosis.
Consider structured exercise as an addition to supportive care for low mood.
Treat a major depressive episode with established psychological or drug treatment.
Tailor exercise to breathlessness and performance status.

Why it matters

Exercise is cheap and low-risk, but this evidence does not make it a treatment for depressive disorder.

The statistics, in plain English

A prediction interval crossing zero means that in some future settings exercise might have no effect at all, even though the average effect favours it. The drop to −0.49 without two outlying trials suggests the realistic effect is moderate.