Genetic and Environmental Influences on the Association Between Depressive Symptom Dimensions and Smoking Initiation Among Chinese Adolescent Twins

摘要

Extant twin research on the depression-smoking association in adolescents has been conducted in U.S. and European samples and considered depression as a unitary phenotype. This study explored genetic and environmental influences on covariation between smoking initiation and 4 depressive symptom dimensions (positive affect [PA], negative affect [NA], somatic features [SF], and interpersonal problems [IP]) in adolescent Chinese twins. Questionnaires measuring current depressive symptoms and lifetime smoking initiation were administered to 602 twin pairs (M [SD] age = 12.2 (1.93) years, range 9-16 years). Cholesky bivariate decomposition models examined influences on each depressive symptom dimension, smoking initiation, and their covariation using age- and sex-adjusted threshold variables. Within-twin correlations between smoking initiation and each depressive symptom dimension were significant (|r|s = .29-.61). Bivariate twin modeling showed significant genetic effects on overall depressive symptoms (55% variance), shared environment effects on NA (36%) and PA (53%), and shared environment effects on smoking initiation (46%) unique from PA. No other familial influences on the individual phenotypes (apart from those accounting for smoking-depression covariance) were significant. Relations of smoking initiation to overall depressive symptoms and IP were influenced by familial (shared environment and/or genetic) factors and nonshared environmental factors. The SF-smoking initiation relation was influenced mostly by familial factors. Only shared environment significantly influenced the association of lower PA and higher NA to smoking initiation. Relations between each symptom dimension and smoking initiation are of sizeable magnitude in Chinese adolescents. Genetic and environmental factors underlying depression-smoking comorbidity may vary across different depressive symptom dimensions.