Cognitive performance and self-reported physical exercise in bipolar disorder: A cross-sectional analysis of pre-randomization screening data from an ongoing randomized controlled trial.

In 30 seconds

This cross-sectional analysis of pre-randomization screening data from an ongoing randomized trial examined self-reported exercise and cognition in 108 euthymic adults aged 18–55 with bipolar disorder. Participants reporting structured exercise at least twice weekly had higher SCIP total scores than those reporting no exercise (64.63 ± 9.19 vs. 56.50 ± 7.84; p < 0.001, d = 0.95). The exercise group also performed better in working memory, which remained significant after correction for multiple comparisons; the verbal-learning difference did not. The observational design cannot establish that exercise caused the cognitive differences.

Key findings

• SCIP total scores were higher in the exercise group than the no-exercise group (64.63 ± 9.19 vs. 56.50 ± 7.84; p < 0.001, d = 0.95).
• Working-memory performance favored the exercise group (p = 0.002) and remained significant after Bonferroni correction.
• Verbal learning favored the exercise group (p = 0.010), but this difference did not remain significant after correction.

Why it matters

Cognitive difficulties can persist in bipolar disorder and affect functioning. The observed association between regular exercise and better global cognition and working memory supports further investigation of exercise as a possible component of neurocognitive rehabilitation. Because exercise was self-reported and the analysis was cross-sectional, the findings do not show that exercise improves cognition.

What to keep in mind

Exercise was self-reported and recorded as a binary measure, without information on intensity, duration, or modality. Demographic and clinical covariates were unavailable for adjustment, and the cross-sectional design precludes causal inference.