AIMS: To estimate the association between educational status and alcohol-related somatic and non-somatic morbidity and mortality among full siblings in comparison with non-related individuals. DESIGN: Cohort study. SETTING: Denmark. PARTICIPANTS: Approximately 1.4 million full siblings born in Denmark between 1950 and 1979 were followed from age 28-58 years or censoring due to alcohol-related hospitalization and mortality. MEASUREMENTS: Cox regression analyses were used to estimate associations of educational status with alcohol-related outcomes. Results from cohort analyses based on non-related individuals and inter-sibling analyses were compared. FINDINGS: A lower educational status was associated with a higher rate of alcohol-related outcomes, especially among the youngest (aged 28-37 years) and individuals born 1970-79. Compared with the cohort analyses, the associations attenuated slightly in the inter-sibling analysis. For example, in the cohort analysis, females with a basic school education born 1970-79 had an increased rate of alcohol-related non-somatic morbidity and mortality [hazard rate ratio (HR) = 4.05, 95% confidence interval (CI) = 3.27-5.02] compared to those with a vocational education. In the inter-sibling analysis, the HR attenuated (HR = 2.66, 95% CI = 1.95-3.63). For alcohol-related somatic outcomes the corresponding figures were HR = 3.47 (95% CI = 2.63-4.58) and HR = 3.36 (95% CI = 2.10-5.38), respectively. In general, the associations were stronger among females than males (aged 28-37) in the analyses of alcohol-related non-somatic outcomes. Health conditions earlier in life explained only a minor part of the associations. CONCLUSIONS: The association between educational status and alcohol-related somatic and non-somatic morbidity and mortality is only driven by familial factors to a small degree.