In patients with psychosis, rates of tobacco smoking and childhood trauma are significantly higher compared to the general population. Childhood trauma has been proposed as a risk factor for tobacco smoking. However, little is known about the relationship between childhood trauma and smoking in psychosis. In a subsample of the Genetic Risk and Outcome of Psychosis study (760 patients with psychosis, 991 unaffected siblings, and 491 healthy controls), tobacco smoking was assessed using the Composite International Diagnostic Interview and childhood trauma was measured with the Childhood Trauma Questionnaire. Logistic regression models were used to assess associations between trauma and smoking, while correcting for confounders. Positive associations were found between total trauma, abuse, and neglect, and an increased risk for smoking in patients, while correcting for age and gender (ORtrauma 1.77, 95% CI 1.30–2.42, p < 0.001; ORabuse 1.69, 95% CI 1.23–2.31, p = 0.001; ORneglect 1.48, 95% CI 1.08–2.02, p = 0.014). In controls, total trauma and abuse were positively associated with smoking, while correcting for age and gender (ORtrauma 2.40, 95% CI 1.49–3.88, p < 0.001; ORabuse 2.02, 96% CI 1.23–3.32, p = 0.006). All associations lost their significance after controlling for additional covariates and multiple testing. Findings suggest that the association between childhood trauma and tobacco smoking can be mainly explained by confounders (gender, cannabis use, and education) in patients with psychosis. These identified aspects should be acknowledged in tobacco cessation programs.