Effects of cigarette smoking on serum biomarkers and its moderation by biological sex in relation to body mass index

Introduction. Smoking is a risk factor for chronic diseases and has a contradictory effect on body mass index (BMI). Metabolic biomarkers, commonly requested in primary care, may mediate this relationship, with sex acting as a moderator. Objective. To evaluate the moderating effect of biological sex in the relationship between smoking and BMI, considering the mediation of serum biomarkers. Materials and methods. An analytical study using secondary data (n=801). The variables included: smoking (dependent), BMI (independent), ALT, AST, creatinine, calcium, phosphorus, cholesterol (mediators), and biological sex (moderator). Hayes' PROCESS Model 59 was used to estimate direct, indirect, and moderated effects, as well as a directed acyclic graph (DAG) to support the causal structure. Results. In the model with BMI as the dependent variable, no significant association was found with smoking (β = -14.78; p = 0.27), sex (β = -34.59; p = 0.28), or their interaction (p = 0.37). However, several biochemical variables showed relevant associations; ALT was positively associated with BMI (β = 0.0742; p = 0.0112; 95% CI [0.0169; 0.1316]), while AST showed a significant negative association (β = -0.1869; p < 0.0001; 95% CI [-0.2629; -0.1109]). Additionally, triglycerides were negatively associated with BMI (β = -0.6008; p = 0.0386). No significant associations were found with creatinine, calcium, phosphorus, or total cholesterol. Regarding interactions with sex, there were significant moderating effects for AST (AST × sex: β = 0.0949; p = 0.0010) and TG (TG × sex: β = 0.9307; p = 0.0004), suggesting that the effect of these variables on BMI differs by sex. Conclusions. Biological sex played a key moderating role in the relationship between BMI and AST and triglycerides, indicating that the impact of these biomarkers on BMI varies by sex.