Personality traits differentiate patients with bipolar disorder and healthy controls - A meta-analytic approach

Expression of specific personality traits has been associated with the presence and disease course of bipolar disorder (BD) in multiple studies. However, until today findings are inconsistent and potentially confounding factors such as age and gender as well as the limited sample size of previous studies make it difficult to generalize these findings. To overcome these limitations and to specify the role of personality traits in the context of BD, we performed a meta-analysis in patients with BD and healthy controls (HC), focusing on the traits of the big three and the big five: Neuroticism (N), Extraversion (E), Openness (O), Conscientiousness (C), Agreeableness (A) and Psychoticism (P).

Two online databases (Pubmed and Web of Science) were searched systematically to identify relevant articles, including publications up to December 31, 2019. From studies that met our inclusion criteria (n = 18), we extracted relevant data of patients with BD (n = 1694) and HC (n = 2153) and calculated effect sizes for each personality trait. Further, we performed moderator analysis on gender, age, quality score and years of publication.

Our results indicate that patients with BD exhibit higher scores on N (large positive effect size; n = 18, g = 1.44, 95%-CI : 1.11 to 1.77) and lower scores on C (medium negative effect size; n = 6, g = -0.78, 95%-CI: -1.13 to -0.43) and E (small negative effect size; n = 13, g = -0.38, 95%-CI: -0.52 to -0.23) compared to HC. We found a moderating effect of mean age on the effect size of N with smaller differences in N levels between patients with BD and HC in older samples (-0.0437, z = - 3.96, p <0.0001). Our results were robust with respect to potential publication biases and the inclusion of potentially confounding factors such as gender, age, quality score and years of publication.

Due to the lack of available data no subgroup analysis on the effect of mood states of patients and subtypes of BD could be performed. Moreover, our analyses are based on cross-sectional data so that findings should be interpreted with care, especially concerning causal conclusions.

Patients with BD showed differences in several personality traits compared to HC. Our results provide the basis for future research with focus on personality and psychopathology in patients with BD. Identifying the interaction between expressions of personality traits and BD might provide novel approaches in prevention and therapy.