Predictors of Smoking Cessation Among African-Americans Enrolled in a Randomized Controlled Trial of Bupropion

Objectives: Identification of individual characteristics that predict successful smoking cessation treatment has been limited to studies with mostly white participants. This study identifies factors that predict successful quitting among African-Americans participating in a smoking cessation trial.

Methods: Twenty-one baseline variables were analyzed as potential predictors from a double-blind placebo-controlled, randomized trial that used bupropion SR for smoking cessation among 600 African-American smokers. Chi-square tests, two sample t tests, and multiple logistic regression procedures were employed to identify predictors of 7-day abstinence among the 535 participants who completed the 7-week medication phase.

Results: Univariate predictors of cessation were receiving bupropion (P < 0.0001), not smoking menthol cigarettes (P = 0.0062), smoking after 30 min of waking (P < 0.0001), older age (P = 0.0085), smoking fewer cigarettes per day (P = 0.0038), and lower cotinine levels (P = 0.0002). Logistic regression identified three significant independent predictors. Participants who received bupropion treatment were more than twice as likely to quit smoking at the end of treatment compared to participants who received placebo (OR = 2.62; 95% CI = 1.77-3.88, P < 0.0001), while smoking within 30 min of waking (OR = 0.40; 95% CI = 0.25-0.62, P < 0.0001) and higher salivary cotinine levels at baseline (OR = 0.799; 95% CI = 0.629-0.922, P < 0.0001) reduced the likelihood of quitting.

Conclusions: This is the first report identifying predictors of smoking cessation among African-Americans participating in a clinical trial. Results indicate that, aside from bupropion treatment, various indicators of addiction were the strongest predictors. While this is similar to findings among white smokers, thresholds of addiction may need to be adjusted for African-American smoking patterns. Additional studies focused on diverse populations are needed to improve treatment approaches and to identify population-specific factors that are important for treatment-matching approaches.