Cardiorespiratory fitness in people with bipolar disorder: A systematic review and meta-analysis

Cardiorespiratory fitness (CRF) predicts cardiovascular and all-cause mortality and is reduced in severe mental illness. This systematic review with meta-analysis estimated the mean CRF of people with bipolar disorder (BD), the difference from controls without BD, and the effect of physical activity interventions on CRF in BD.

Five databases (PubMed, Embase, Web of Science, PsycINFO, SPORTDiscus) were searched from inception to November 24, 2025, and updated on August 25, 2026. Random-effects models pooled the mean CRF and the difference versus controls; subgroup and sensitivity analyses were post hoc. The review was not prospectively registered.

Twenty studies were included; ten reported oxygen uptake, nine of which were pooled. In adults, mean CRF was 27.0 mL·kg^-1·min^-1 (k = 8, n = 190; 95%CI 21.5 to 32.5; I² = 97.0%); adding the adolescent sample gave 28.7 mL·kg^-1·min^-1 (k = 9, n = 210; 95%CI 22.3 to 35.1; I² = 98.1%). CRF was 7.4 mL·kg^-1·min^-1 lower in BD than in controls (k = 7, n = 333; 95%CI -12.7 to -2.1; p = 0.006; I² = 89.9%; standardized mean difference - 0.89, 95%CI -1.41 to -0.38). Only two uncontrolled pilot studies reported pre-post CRF, precluding meta-analysis.

The evidence base is small, cross-sectional and heterogeneous; the between-group estimate was not robust to excluding one study, and two indirect protocols applied to the same participants differed by 19.3 mL·kg^-1·min^-1.

CRF was consistently lower in BD than in healthy controls, although heterogeneity means the pooled mean is a descriptive summary rather than a representative value. No conclusion about exercise efficacy can be drawn. Adequately powered controlled trials with standardized CRF assessment are required.