Effects of Tobacco Versus Electronic Cigarette Usage on Nonsuicidal Self-Injury and Suicidality Among Chinese Youth: Cross-Sectional Self-Report Survey Study

Abstract

BackgroundThe increase in tobacco/conventional cigarette (CC) and electronic cigarette (EC) usage among Chinese youth has become a growing public health concern. This is the first large-scale study to compare the impact of CC and EC usage on risk for nonsuicidal self-injury (NSSI) and suicidality in cis-heterosexual and sexual and gender minority (SGM) youth populations in China. ObjectiveThis study examines the CC and EC risks for NSSI and suicidality among Chinese youth and compares the extent to which SGM and cis-heterosexual youth’s risks for NSSI and suicidality are influenced by their CC and EC usage and dependence. MethodsA total of 89,342 Chinese participants completed a cross-sectional self-report survey in 2021. Sociodemographic information, sexual orientations, gender identities, CC and EC usage, CC and EC dependence, and risks for suicidality and NSSI were assessed. The Mann-Whitney U test and chi-square test were performed for nonnormally distributed continuous variables and categorical variables, respectively. The multivariable linear regression model was used to examine both the influence of CC and EC usage and CC and EC dependence on NSSI and suicidality as well as the interaction effects of CC and EC usage and CC and EC dependence on NSSI and suicidality by group. ResultsThe prevalence of CC usage (P<.001) and dependence (P<.001) among SGM participants was lower than that among their cis-heterosexual counterparts. However, the prevalence of EC usage (P=.03) and EC dependence (P<.001) among SGM participants was higher than that among their cis-heterosexual counterparts. The multivariable linear regression model showed that CC dependence and EC dependence had a unique effect on NSSI and suicidality (CCs: B=0.02, P<.001; B=0.09, P<.001; ECs: B=0.05, P<.001; B=0.14, P<.001, respectively). The interaction effects of (1) CC usage and group type on NSSI and suicidality (B=0.34, P<.001; B=0.24, P=.03, respectively) and dual usage and group type on NSSI and suicidality (B=0.54, P<.001; B=0.84, P<.001, respectively) were significant, (2) CC dependence and group type on NSSI were significant (B=0.07, P<.001), and (3) EC dependence and group type on NSSI and suicidality were significant (B=0.04, P<.001; B=0.09, P<.001, respectively). No significant interaction effect was observed between EC usage and group type on NSSI and suicidality (B=0.15, P=.12; B=0.33, P=.32, respectively) and between CC dependence and group type on suicidality (B=–0.01, P=.72). ConclusionsOur study shows evidence of intergroup differences in NSSI and suicidality risks between SGM and cis-heterosexual youth related to CC and EC usage. These findings contribute to the growing literature on CC and EC in cis-heterosexual and SGM populations. Concerted efforts are necessary at a societal level to curb the aggressive marketing strategies of the EC industry and media coverage and to maximize the impact of educational campaigns on EC prevention and intervention among the youth population.