Abstract

Context: Tobacco usage accounts for one of the most preventable causes of death. Objective: This study aimed to explore mechanisms linking cigarette pack factors, point-of-sale marketing, and individual factors (psychological reactant trait) to predict smoking intention among school-going adolescents. Design, Setting, and Participants: This was a cross-sectional study conducted among 6 urban secondary schools. A pretested and validated self-administered questionnaire was used. Data analysis for structural equation modeling was done using SMART-PLS v3.2.8. Main Outcome Measure: The main outcome measure was to determine the direct and indirect effects of cigarette pack factors, point-of-sale marketing, and individual factors (psychological reactant trait) to predict smoking intention among school-going adolescents in a theory-based model. Results: A total of 386 adolescents fulfilling the inclusion criteria participated. Pictorial warning message reactance (β = .153, P ≤ .001), pack receptivity of conventional pack (β = .297, P = .004), and psychological reactant trait (β = .174, P ≤ .001) were positively related to smoking intention. Pictorial warning negative affect (β = −.153, P = .001) was negatively related to smoking intention. The psychological reactant trait was positively related to message reactance (β = .340, P ≤ .001). However, recall exposure to point-of-sale marketing and pack appraisal of conventional pack was not positively related to smoking intention (β = .038, P = .215 and β = −.026, P = .39, respectively). Pictorial warning message reactance also positively mediates the relationship between psychological reactant trait and smoking intention (β = 0.05, p = .001). The model has strong predictive power. Conclusion: In conclusion, cigarette pack factors and psychological reactant traits are essential in predicting smoking intention. Hence, policymakers should consider these factors in developing smoking policies.