Objective: Bipolar-I disorder (BPI) often co-occurred with anxiety (ANX) and substance use disorders (SUD), which poses challenges in public health and clinical treatment, and adds complexity in searching for relevant etiologic factors. The present study sought to identify subgroups of BPI patients using comorbidity patterns with ANX and SUD.
Methods: Clinical patients (N=306) diagnosed with BPI were recruited and interviewed using the Composite International Diagnostic Interview to collect data on demographics and clinical features, including episodic information, impairments, and lifetime diagnoses of ANX (panic, agoraphobia, generalized anxiety disorder, specific and social phobia) and SUD (nicotine dependence, alcohol use and drug use disorder). We applied latent class analysis to empirically derive classes of BPI. A number of exogenous variables were examined for each class.
Results: A three-class model provides excellent discriminability for subgrouping BPI patients with different comorbidity patterns. The BPI-LOW class (83.99%) had more pure mania without most lifetime comorbidity, higher numbers of last year mania episodes, and less suicidality and impairments. The BPI-ANX class (3.60%) was female predominant, tended to comorbid with multiple anxiety disorders but no SUD, and had early onset age. The BPI-SUD class (12.42%) was male predominant, had high prevalence of lifetime SUD and frequent mood episodes in the last year. Both the BPI-ANX and BPI-SUD classes had severe functional impairments and suicidal behaviors.
Limitations: Clinical information was retrospectively collected. Besides, we did not comprehensively access lifetime comorbidity for all psychiatric disorders.
Conclusion: The three empirically identified subgroups of BPI patients exhibited distinguished comorbidity patterns and clinical features, including suicidal behaviors, frequent mood episodes and functional impairments. Our findings have clinical implication in intervention and treatment as well as to explore their different underlying mechanisms.
Copyright © 2011 Elsevier B.V. All rights reserved.