Predictors of first-onset substance use disorders during the prospective course of bipolar spectrum disorders in adolescents

J Am Acad Child Adolesc Psychiatry. 2013 Oct;52(10):1026-37. doi: 10.1016/j.jaac.2013.07.009. Epub 2013 Jul 31.

Abstract

Objective: Substance use disorders (SUD) are common and problematic in bipolar disorder (BP). We prospectively examined predictors of first-onset SUD among adolescents with BP.

Method: Adolescents (12-17 years old; N = 167) in the Course and Outcome of Bipolar Youth (COBY) study fulfilling criteria for BP-I, BP-II, or operationalized BP not otherwise specified, without SUD at intake, were included. Baseline demographic, clinical, and family history variables, and clinical variables assessed during follow-up, were examined in relation to first-onset SUD. Participants were prospectively interviewed every 38.5 ± 22.2 weeks for an average of 4.25 ± 2.11 years.

Results: First-onset SUD developed among 32% of subjects, after a mean of 2.7 ± 2.0 years from intake. Lifetime alcohol experimentation at intake most robustly predicted first-onset SUD. Lifetime oppositional defiant disorder and panic disorder, family history of SUD, low family cohesiveness, and absence of antidepressant treatment at intake were also associated with increased risk of SUD, whereas BP subtype was not. Risk of SUD increased with increasing number of these 6 predictors: 54.7% of subjects with 3 or more predictors developed SUD vs. 14.1% of those with fewer than 3 predictors (hazard ratio = 5.41 95% confidence interval = 2.7-11.0 p < .0001). Greater hypo/manic symptom severity in the preceding 12 weeks predicted greater likelihood of SUD onset. Lithium exposure in the preceding 12 weeks predicted lower likelihood of SUD.

Conclusions: This study identifies several predictors of first-onset SUD in the COBY sample that, if replicated, may suggest targets for preventive interventions for SUD among youth with BP. Treatment-related findings are inconclusive and must be interpreted tentatively, given the limitations of observational naturalistic treatment data. There is a substantial window of opportunity between BP and SUD onset during which preventive strategies may be used.

Keywords: bipolar; predictors; prevention; prospective; substance use disorder.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Age of Onset
  • Attention Deficit and Disruptive Behavior Disorders / epidemiology
  • Bipolar Disorder / classification
  • Bipolar Disorder / drug therapy
  • Bipolar Disorder / epidemiology*
  • Child
  • Comorbidity
  • Family Relations
  • Female
  • Humans
  • Lithium / therapeutic use
  • Male
  • Panic Disorder / epidemiology
  • Predictive Value of Tests
  • Prospective Studies
  • Risk Factors
  • Severity of Illness Index
  • Substance-Related Disorders / epidemiology*

Substances

  • Lithium