Comorbid Eating Disorders in a Sample of Youth With Bipolar Disorder: Elevated Burden of Dimensional and Categorical Psychopathology

Abstract

Objective: There is growing recognition of the importance of comorbid eating disorders (ED) among individuals with bipolar disorder (BD). However, most studies on this topic have focused on adult samples, and little is known regarding comorbid ED among youth with BD.

Methods: The sample included 197 youth with DSM-IV BD (BD-I, BD-II, or BD-NOS [not otherwise specified]), aged 13-20 years and recruited from a subspecialized clinic within a tertiary academic health sciences center from 2009 to 2017. Univariate analyses examined demographic and clinical variables among participants with versus without lifetime DSM-IV ED. Variables significant at P < .10 were entered into a backward stepwise regression.

Results: Fifty-six participants (28.4%) had lifetime DSM-IV ED (3.6% anorexia nervosa, 8.1% bulimia nervosa, 16.8% ED not otherwise specified). Significant correlates of lifetime ED were female sex (P < .001), BD-II subtype (P = .03), suicidal ideation (P = .006), suicide attempts (P = .004), non-suicidal self-injury (P < .001), sexual abuse (P = .02), cigarette smoking (P = .001), anxiety disorders (P = .004), posttraumatic stress disorder (P = .004), substance use disorders (P = .006), history of individual therapy (P = .01), and family history of anxiety (P = .01). Significant correlates of no lifetime ED were BD-I subtype (P < .001) and lifetime lithium use (P = .01). The ED group had significantly more severe lifetime depression (P < .001) and significantly more self-reported affective lability (P < .001) and borderline personality traits (P < .001). In multivariate analysis, the most robust predictors of lifetime ED were female sex (odds ratio [OR] = 4.61, P = .004), BD-I subtype (OR = 0.21, P = .03), cigarette smoking (OR = 2.78, P = .02), individual therapy (OR = 3.92, P = .03), family history of anxiety (OR = 2.86, P = .02), and borderline personality traits (OR = 1.01, P = .009).

Conclusions: ED are common among youth with BD and associated with adverse clinical characteristics, many of which converge with prior adult literature. Future studies evaluating specific ED subtypes are warranted, as are treatment studies targeting comorbid ED in youth with BD.