• bipolar disorder;
  • major depression;
  • nicotine;
  • schizophrenia;
  • smoking;
  • tobacco

Objectives:  This study compared the prevalence of tobacco smoking behaviors in patients with bipolar disorder with normal and psychiatric (schizophrenia and major depression) controls. The main goal was to establish that bipolar patients smoke more than normal controls. Differences with psychiatric controls were explored.

Methods:  Samples of 424 patients (99 bipolar, 258 schizophrenia and 67 major depression) and 402 volunteer controls were collected in Central Kentucky. Smoking data for Kentucky’s general population were available. Odds ratios (ORs) and their 95% confidence intervals (CIs) were used to establish the strength of associations. Logistic regression was used to adjust ORs for confounding variables.

Results:  Using epidemiological definitions of smoking behaviors and the general population as controls provided bipolar disorder unadjusted ORs of 5.0 (95% CI: 3.3–7.8) for current cigarette smoking, 2.6 (95% CI: 1.7–4.4) for ever cigarette smoking, and 0.13 (95% CI: 0.03–0.24) for smoking cessation. Using a clinical definition and volunteers as controls provided respective bipolar disorder adjusted ORs of 7.3 (95% CI: 4.3–12.4), 4.0 (95% CI: 2.4–6.7), and 0.15 (95% CI: 0.06–0.36). Prevalences of current daily smoking for patients with major depression, bipolar disorder, and schizophrenia were 57%, 66%, and 74%, respectively.

Conclusions:  Bipolar disorder was associated with significantly higher prevalences of tobacco smoking behaviors compared with the general population or volunteer controls, independently of the definition used. It is possible that smoking behaviors in bipolar disorder may have intermediate prevalences between major depression and schizophrenia, but larger samples or a combination of multiple studies (meta-analysis) will be needed to establish whether this hypothesis is correct.