Get access

Cigarette smoking in the early course of bipolar disorder: association with ages-at-onset of alcohol and marijuana use


  • This study was presented in part at the Society for Research on Nicotine and Tobacco meeting, February 2007, Austin, TX, USA.

  • The Tri-State Tobacco and Alcohol Research Center receives research support from Ortho McNeil, Sanofi-Aventis, and Addex. MPD receives research support, provides consulting and/or speakers bureau services to AstraZeneca, Eli Lilly & Co., Johnson & Johnson, Pfizer, Shire, Janssen, Bristol-Myers Squibb, Otsuka, and Repligen. RMA provides consultancy, advisory, and/or speakers bureau services to Sanofi-Aventis, Pfizer, and Alkermes/Cephalon. SMS receives research support, provides consulting and/or speakers bureau services to Eli Lilly & Co., Janssen, Pfizer, Forrest, AstraZeneca, Bristol-Myers Squibb, Martek Biosciences, Nutrition 21, Ortho McNeil, and Solvay. JLH and DEF have no competing interests to report.

Jaimee L. Heffner, PhD, Tri-State Tobacco and Alcohol Research Center, University of Cincinnati Genome Research Institute, 2120 E. Galbraith Road, Building A (mail code: 0506), Cincinnati, OH 45237, USA. Fax: +1 513 475 6568; e-mail:


Objectives:  Despite the high prevalence of smoking among individuals with bipolar disorder, few studies have attempted to identify correlates of smoking status in this group. We examined illness characteristics of bipolar disorder as well as co-occurring alcohol and marijuana use as correlates of cigarette smoking, including the developmental timing of the onset of regular alcohol and cannabis use (i.e., three or more times per week for a month or more).

Methods:  Demographic and clinical characteristics of 134 patients with bipolar I disorder, the majority of whom were adolescents, who were hospitalized for their first manic episode were analyzed to identify correlates of smoking status.

Results:  A total of 61 (45.5%) of the patients were smokers at the time of their first hospitalization. Smokers were significantly more likely than nonsmokers to report recent use of marijuana (55.7% versus 18.1%) and alcohol (67.2% versus 25.4%). Among those who had ever used marijuana (48.5%) or alcohol (45.5%) regularly, current cigarette smokers reported a significantly earlier age-at-onset of regular use of both substances than reported by nonsmokers. Earlier age-at-onset of marijuana use was the only significant predictor of current smoking in a multivariate analysis. None of the bipolar disorder characteristics examined (i.e., symptom severity, age-at-onset of illness, rapid cycling, and psychosis) were correlated with smoking status.

Conclusions:  Smoking status in the early course of bipolar disorder is related to both current and past alcohol and marijuana use, but not to characteristics of bipolar illness. Earlier initiation of regular marijuana use is associated with an increased risk of smoking cigarettes.