Background <p>Major depressive disorder (MDD) is associated with an elevated risk of self-harm. However, the independent and combined associations of smoking and alcohol consumption with self-harm ideation and attempts remain unclear.</p> Methods <p>We conducted a cross-sectional analysis using data from the UK Biobank. Smoking and alcohol consumption were assessed using baseline touchscreen questionnaires. Self-harm outcomes were ascertained from the Mental Health Questionnaire. Logistic regression models were employed to evaluate the associations between smoking, alcohol consumption, and self-harm outcomes. Restricted cubic spline (RCS) transformations were applied to continuous variables to account for potential non-linear relationships. Subgroup and sensitivity analyses were performed.</p> Results <p>Among 31,359 participants with MDD (excluding those with psychosis, bipolar disorder, or substance use disorders), 17,137 (54.6%) reported self-harm ideation and 2,530 (8.1%) reported self-harm attempts. Smoking and alcohol consumption, as well as their combined exposure, were significantly associated with both self-harm ideation and attempts. Compared with current non-smoker and non-drinker, dual users had elevated odds of self-harm ideation (OR = 1.20, 95% CI: 1.05–1.37) and attempts (OR = 1.30, 95% CI: 1.06–1.61). Each additional cigarette per day was associated with higher odds of self-harm ideation (OR = 1.01, 95% CI: 1.01–1.02) and attempts (OR = 1.02, 95% CI: 1.01–1.04). Alcohol consumption showed J-shaped associations with both self-harm ideation and attempts, with inflection points at 14.00 units/week for ideation and 3.34 units/week for attempts. For self-harm ideation, females exhibited higher odds than males at equivalent consumption levels.</p> Conclusions <p>This cross-sectional analysis identified significant associations between smoking, alcohol use, and self-harm behaviors in MDD patients, with higher odds among dual users.</p> Clinical trial number <p>Not applicable.</p>