Background: This study aimed to compare the treatment responses between smokers and non-smokers in bipolar mania clinical trials.
Methods: Post-hoc analysis was conducted on data collected from three double-blind, randomised controlled trials in bipolar mania that had similar inclusion criteria. Patients were randomised to olanzapine (N=70) or placebo (N=69) for 3 weeks in Trial 1, olanzapine (N=234) or haloperidol (N=216) for 12 weeks in Trial 2, and olanzapine (N=125) or divalproex (N=126) for 47 weeks in Trial 3. This study analysed the Young Mania Rating Scale (YMRS) total scores and Clinical Global Impressions scale for bipolar disorder (CGI-BP) mania severity scores between smokers and non-smokers for each trial and for the pooled data from all three trials, using a mixed-effects model repeated measures approach.
Results: For the pooled data, non-smokers showed superior treatment outcomes on both the YMRS (P=0.002) and CGI-BP (P<0.001), as well as longer time to discontinuation for any cause utilising Kaplan-Meier survival curves. For the individual trials, non-smokers showed greater improvement than smokers on both CGI-BP and YMRS in both treatment arms of Trial 2 (CGI-BP: haloperidol P=0.011, olanzapine P=0.042; YMRS: haloperidol P=0.010, olanzapine P=0.019), and in the olanzapine arm of Trial 3 (CGI-BP: P=0.002; YMRS: P=0.006). No significant difference in outcomes was found between smokers and non-smokers in Trial 1.
Limitations: Post-hoc design, categorical definition of smoking status, unavailable antipsychotic drug levels, confounding effects of trial medications and substance abuse.
Conclusions: Smoking appears to be associated with worse treatment outcomes in mania.