Prevalence and correlates of manic/hypomanic and depressive predominant polarity in bipolar disorder: systematic review and meta-analysis

Background Identification of the predominant polarity, i.e. hypomanic/manic (mPP) or depressive predominant polarity (dPP), might help clinicians to improve personalised management of bipolar disorder. Aims We performed a systematic review and meta-analysis to estimate prevalence and correlates of mPP and dPP in bipolar disorder. Method The protocol was registered in the Open Science Framework Registries (https://doi.org/10.17605/OSF.IO/8S2HU). We searched main electronic databases up to December 2023 and performed random-effects meta-analyses of weighted prevalence of mPP and dPP. Odds ratios and weighted mean differences (WMDs) were used for relevant correlates. Results We included 28 studies, providing information on rates and/or correlates of mPP and dPP. We estimated similar rates of mPP (weighted prevalence = 30.0%, 95% CI: 23.1 to 37.4%) and dPP (weighted prevalence = 28.5%, 95% CI: 23.7 to 33.7%) in bipolar disorder. Younger age (WMD = -3.19, 95% CI: -5.30 to -1.08 years), male gender (odds ratio = 1.39, 95% CI: 1.10 to 1.76), bipolar-I disorder (odds ratio = 4.82, 95% CI: 2.27 to 10.24), psychotic features (odds ratio = 1.56, 95% CI: 1.01 to 2.41), earlier onset (WMD = -1.57, 95% CI: -2.88 to -0.26 years) and manic onset (odds ratio = 13.54, 95% CI: 5.83 to 31.46) were associated with mPP (P < 0.05). Depressive onset (odds ratio = 12.09, 95% CI: 6.38 to 22.90), number of mood episodes (WMD = 0.99, 95% CI: 0.28 to 1.70 episodes), history of suicide attempts (odds ratio = 2.09, 95% CI: 1.49 to 2.93) and being in a relationship (odds ratio = 1.98, 95% CI: 1.22 to 3.22) were associated with dPP (P < 0.05). No differences were estimated for other variables. Conclusions Despite some limitations, our findings support the hypothesis that predominant polarity might be a useful specifier of bipolar disorder. Evidence quality was mixed, considering effects magnitude, consistency, precision and publication bias. Different predominant polarities may identify subgroups of patients with specific clinical characteristics.