Prevalence of unipolar mania in bipolar I disorder: a systematic review and meta-analysis of observational studies

Unipolar mania (UM), defined by the occurrence of manic episodes without a history of depression, is a topic of debate within the classification of affective disorders. However, its epidemiological burden remains unclear. This systematic review and meta-analysis aimed to estimate the prevalence of UM among individuals with bipolar type I disorder (BD-I) while exploring potential sources of heterogeneity.

The study protocol was registered in Open Science Framework on 27 March 2025. Embase, MEDLINE and APA PsycInfo were searched. We included observational studies reporting data on UM prevalence rates in adults with BD-I. Pooled prevalence was estimated using the Freeman-Tukey double arcsine transformation, employing a restricted maximum likelihood random-effects model. Subgroup and meta-regression analyses were implemented to explore sources of heterogeneity.

We included 26 studies, encompassing 35 independent samples and 17,716 individuals with BD-I. The pooled prevalence of UM was 21.1% (95% confidence interval: 15.5-27.4%). Although potential publication bias was detected (Egger's p = 0.010), the trim-and-fill method did not impute any missing studies. No differences were found between clinical and community-based studies (p = 0.966). However, prevalence estimates were influenced by both geographical area (p = 0.020) and study quality (p = 0.014). Rate differences across studies may also be attributable to variations in UM diagnostic definition.

People with UM represent a significant subset of BD-I cases worldwide, warranting greater clinical awareness. The observed rate variability emphasizes the impact of sociocultural and methodological factors on UM diagnosis. Further research is necessary to refine diagnostic criteria and evaluate optimal treatment approaches for individuals with UM.