Interaction and effect modification in the association between socioeconomic status and adolescent smoking: a systematic review.

Abstract

This systematic review was conducted to identify effect modification and interaction factors that moderate the association between socioeconomic status (SES) and smoking behavior among adolescents. We searched the PubMed, Embase, PsycINFO, and Web of Science databases using keywords including "adolescents," "smoking," "inequality," "effect modification," and "interaction." Peer-reviewed articles published in English or French between January 1, 2011, and December 31, 2021, were included, as were relevant studies identified from reference lists. Of 3485 articles, 23 met the eligibility criteria. All reported quantitative observational study designs to identify factors that modify the SES-smoking behavior association. Two independent reviewers extracted data from each article, using a standardized form. Reporting quality was assessed using the Strengthening the Reporting of Observational Articles in Epidemiology guidelines. Of 23 studies, 13 reported statistically significant moderation associations. The most frequently studied moderators were race/ethnicity (n = 5; significant in 4) and sex (n = 5; no significant associations). Contextual factors, including school-level SES (n = 2; significant in 1), neighborhood SES (n = 2; significant in both), and peer influence (n = 2; significant in both), were also examined. Time trends (n = 2), country-level factors (n = 2), and social capital (n = 1) were significant moderators in the few studies that investigated them. Methodologically, adherence to best practices was limited. This review highlights the need for use of a wider range of SES measures, exploration of understudied potential moderators, and consistent adherence to standardized methodologies to better inform public health interventions addressing adolescent smoking inequalities.