Explaining why childhood abuse is a risk factor for poorer clinical course in bipolar disorder: a path analysis of 923 people with bipolar I disorder.

Steven Marwaha,Paul M Briley,Ian Jones,Arianna Diflorio,Amy Perry,Phillip Rankin,Lisa Jones,Matthew Broome,Nick Craddock,Katherine Gordon-Smith

doi:10.1017/s0033291719002411

Abstract

Childhood abuse is a risk factor for poorer illness course in bipolar disorder, but the reasons why are unclear. Trait-like features such as affective instability and impulsivity could be part of the explanation. We aimed to examine whether childhood abuse was associated with clinical features of bipolar disorder, and whether associations were mediated by affective instability or impulsivity. We analysed data from 923 people with bipolar I disorder recruited by the Bipolar Disorder Research Network. Adjusted associations between childhood abuse, affective instability and impulsivity and eight clinical variables were analysed. A path analysis examined the direct and indirect links between childhood abuse and clinical features with affective instability and impulsivity as mediators. Affective instability significantly mediated the association between childhood abuse and earlier age of onset [effect estimate (θ)/standard error (SE): 2.49], number of depressive (θ/SE: 2.08) and manic episodes/illness year (θ/SE: 1.32), anxiety disorders (θ/SE: 1.98) and rapid cycling (θ/SE: 2.25). Impulsivity significantly mediated the association between childhood abuse and manic episodes/illness year (θ/SE: 1.79), anxiety disorders (θ/SE: 1.59), rapid cycling (θ/SE: 1.809), suicidal behaviour (θ/SE: 2.12) and substance misuse (θ/SE: 3.09). Measures of path analysis fit indicated an excellent fit to the data. Affective instability and impulsivity are likely part of the mechanism of why childhood abuse increases risk of poorer clinical course in bipolar disorder, with each showing some selectivity in pathways. They are potential novel targets for intervention to improve outcome in bipolar disorder.

Highlights

Bipolar disorder is a multi-component illness (Rowland and Marwaha, 2018), and globally amongst the top 10 causes of disability (Murray and Lopez, 2007; Gore et al, 2011) with annual projected costs of £8 billion by 2020 (Knapp et al, 2011)
There were no significant differences between responders and non-responders in mean age at onset (24 years in both), frequency of suicidal behaviour, substance misuse, rapid cycling, psychosis or mean DSM-IV Global Assessment of Functioning scores in the worst episode of mania and depression in each group
History of suicidal behaviour was present in 48.8% of cases (n = 450), substance misuse in 38.1% (n = 352), rapid cycling in 20.4% (n = 188), psychotic symptoms in 68.7% (n = 634) and anxiety disorder in 53.2% (n = 491)

Summary

Introduction

Bipolar disorder is a multi-component illness (Rowland and Marwaha, 2018), and globally amongst the top 10 causes of disability (Murray and Lopez, 2007; Gore et al, 2011) with annual projected costs of £8 billion by 2020 (Knapp et al, 2011). A path analysis examined the direct and indirect links between childhood abuse and clinical features with affective instability and impulsivity as mediators. Affective instability significantly mediated the association between childhood abuse and earlier age of onset [effect estimate (θ)/standard error (SE): 2.49], number of depressive (θ/SE: 2.08) and manic episodes/illness year (θ/SE: 1.32), anxiety disorders (θ/SE: 1.98) and rapid cycling (θ/SE: 2.25). Affective instability and impulsivity are likely part of the mechanism of why childhood abuse increases risk of poorer clinical course in bipolar disorder, with each showing some selectivity in pathways. They are potential novel targets for intervention to improve outcome in bipolar disorder

Objectives

Methods

Results

Discussion

Conclusion