Background: The inability of individuals with Alcohol Use Disorders (AUD) to recognize and describe their feelings and cravings may be due to alexithymia. Previous researches have shown evidence for a negative influence of alexithymia on treatment outcomes in patients with AUD. Therefore, it was hypothesized that high alexithymic patients with AUD would benefit less from cognitive behavioral therapy (CBT) compared with low alexithymic patients. Methods: One hundred alcohol-dependent inpatients (DSM IV) were assessed with the Mini International Neuropsychiatric Interview for psychiatric disorders, the Toronto Alexithymia Scale (TAS-20), and the European Addiction Severity Index (EuropASI). Baseline alexithymia, as a categorical and continuous variable, was used to compare or relate baseline demographic and addiction characteristics, time in treatment, abstinence, and differences in addiction severity at 1-year follow-up. Analyses were performed using χ2 test, analysis of variance or Kruskal–Wallis, paired t-tests or Wilcoxon’s signed rank tests, multivariate logistic, and linear regression models, as appropriate. Results: The prevalence of high alexithymia (TAS-20 > 61) was 45%. The total TAS-20 score correlated negatively with years of education (r = −.21; p = .04) and positively with the psychiatry domain of the EuropASI (r = .23; p = .04). Alexithymia showed no relation to abstinence, time in treatment, or change in severity of alcohol-related problems on the EuropASI. Conclusion: High alexithymic patients with AUD do benefit equally from inpatient CBT-like treatment as low alexithymic patients with AUD. Scientific significance: Multimethod alexithymia assessments with an observer scale have been advised to judge the relationship with resulting outcome in CBT.