Smoking as a correlate of suicidal behavior and self-harm in adolescents with depressive disorders.

Suicidal behavior and self-harm are major public health concerns among adolescents, particularly those with depressive disorders. While smoking has been linked to suicidality in general populations, its independent role in both suicidal behavior and self-harm within well-characterized clinical samples of youth with depressive disorders remains understudied. This study consecutively enrolled 2,343 adolescents (aged 12-18 years) diagnosed with unipolar depression, bipolar disorder, or depressive episode according to DSM-5 criteria at Nanhai Public Health Hospital of Foshan City and The Third People's Hospital of Foshan between January 2025 and December 2025. Suicidal behavior (≥1 suicide attempt) and self-harm (intentional self-injury or poisoning regardless of intent) were assessed via clinical interviews. Multivariable binary logistic regression models were constructed to identify factors independently associated with each outcome, adjusting for age, sex, education, income, parental education, psychiatric diagnosis, family history of mental illness, physical disease, and smoking status. Among 2,343 participants (mean age 14.99 ± 1.65 years; 77.8% female), the prevalence of self-harm was 76.0% and suicidal behavior was 44.2%. In fully adjusted models, smoking status was the only variable significantly associated with suicidal behavior after adjustment for measured covariates: current smokers (OR = 2.74, 95% CI: 1.81-4.22, P<0.001) and past quitters (OR = 2.32, 95% CI: 1.66-3.28, P<0.001) had higher odds compared to never smokers. For self-harm, current smoking was significantly associated with increased risk (OR = 2.31, 95% CI: 1.33-4.37, P = 0.006). Education level showed a borderline association with self-harm, with each additional year of schooling corresponding to a 10% lower odds (OR = 0.90, 95% CI: 0.81-1.00, P = 0.050); however, this finding should be interpreted cautiously given the exploratory nature of the analysis. No other demographic or clinical variables, including age, sex, or psychiatric diagnosis, were independently associated with either outcome. Smoking was strongly associated with both suicidal behavior and self-harm after adjustment for measured demographic and clinical covariates. These findings underscore the importance of assessing tobacco use as a potential clinical marker of vulnerability in youth mental health settings.