Digital physical activity intervention shows limited benefit for cancer survivors' anxiety and depression

In the primary intent-to-treat analysis, the intervention produced no significant improvements in depression, anxiety, cognitive complaints, or FCR across the full group. However, post-hoc stratified analyses revealed a notable exception: among survivors who reported high baseline FCR and who actively engaged with the website, FCR scores decreased significantly from 23.8 at baseline to 19.2 at 12 months—a clinically meaningful reduction. The intervention showed no meaningful benefit for low-baseline-symptom survivors, and compliance with the website protocol appeared critical to any benefit. This finding suggests the intervention may help only a specific subgroup and points to the need for prospective validation of these post-hoc results, which carry higher risk of false discovery.

The null findings for depression and anxiety in the primary analysis are striking given that observational data support PA for mental health in cancer survivors. The post-hoc reduction in FCR among compliant, high-baseline-FCR survivors (95% CI 16.8–21.6 at month 12 vs. baseline 21.3–26.2) is noteworthy because FCR is one of the most distressing long-term sequelae in survivorship, yet very few trials directly target it. The mechanism may relate to mastery and symptom control via consistent exercise engagement rather than a universal PA effect. Critically, the post-hoc nature of this finding (not pre-specified) inflates the risk of false positive discovery, and the lack of a specific high-FCR subgroup arm means the effect cannot be confirmed until a prospective trial is run. Compliance emerged as a confounding factor—participants who engaged more may differ in motivation, baseline symptom severity, or access to support, limiting causal inference. The 94% female sample also limits generalizability to male survivors, who may experience FCR differently.