Results of the three logistic regression analyses predicting relapse in men and the three parallel analyses in women are presented in Table 2. For the men, after controlling for age, none of the three trauma measures were significant predictors of relapse. For the women, however, all three of the measures were significant relapse predictors after controlling for days of alcohol abstinence. Odds ratio estimates obtained from these analyses suggested that each 1-point increase in scores on the CTQ-SF total scores and the TSC-40 total scores was associated with a 7–9% increase in the odds of relapsing. On the TLEQ, each 1-event increase in the number of traumatic events evoking fear, helplessness, or horror was associated with a 46% increase in the odds of relapsing.
