Expectancies for the effectiveness of different tobacco interventions account for racial and gender differences in motivation to quit and abstinence self-efficacy

Abstract

Introduction: Racial and gender disparities for smoking cessation might be accounted for by differences in expectancies for tobacco interventions, but few studies have investigated such differences or their relationships with motivation to quit and abstinence self-efficacy.

Methods: In this cross-sectional study, 673 smokers (African American: n = 443, 65.8%; women: n = 222, 33.0%) under criminal justice supervision who enrolled in a clinical smoking cessation trial in which all received bupropion and half received counseling. All participants completed pretreatment measures of expectancies for different tobacco interventions, motivation to quit, and abstinence self-efficacy. The indirect effects of race and gender on motivation to quit and abstinence self-efficacy through expectancies for different tobacco interventions were evaluated.

Results: African Americans' stronger expectancies that behavioral interventions would be effective accounted for their greater motivation to quit and abstinence self-efficacy. Women's stronger expectancies for the effectiveness of pharmacotherapy accounted for their greater motivation to quit, whereas their stronger expectancies for the effectiveness of behavioral treatments accounted for their greater abstinence self-efficacy.

Conclusions: Findings point to the mediating role of expectancies for treatment effectiveness and suggest the importance of exploring expectancies among African Americans and women as a way to augment motivation and self-efficacy.

Figure 1.

Mediation path models of the association between race (African American and White race; White race was the reference group), treatment effectiveness expectancies (behavioral therapy and pharmacotherapy), and (a) motivation to quit and (b) abstinence self-efficacy. Gender, age, current substance use, and cigarette dependence (Fagerström Test of Cigarette Dependence) are included as covariates. Reported path coefficients are unstandardized. The values in parentheses are the direct effects of race, and the values outside of the parentheses are the total effects.*p < .07. **p < .05. ***p < .01. ****p < .001.

Figure 2.

Mediation path models of the association between gender (women and men; males were the reference group), treatment effectiveness expectancies (behavioral therapy and pharmacotherapy), and (a) motivation to quit and (b) abstinence self-efficacy. Race, age, current substance use, and cigarette dependence (Fagerström Test of Cigarette Dependence) are included as covariates. Reported path coefficients are unstandardized. The values in parentheses are the direct effects of gender, and the values outside of the parentheses are the total effects.*p < .07. **p < .05. ***p < .01. ****p < .001.