The role of alcohol intake in the risk of Hodgkin lymphoma (HL) is still largely unclear. To summarize the evidence on the issue, we carried out a meta-analysis of the available studies. We identified eight case-control and two cohort studies, including a total of 1488 cases of HL. We derived meta-analytic estimates using random-effects models, taking into account the correlation between estimates, and carried out a dose-risk analysis using nonlinear random-effects metaregression models. Compared with nondrinkers, the relative risk for alcohol consumers was 0.70 [95% confidence interval (CI), 0.60-0.81] overall, 0.66 (95% CI, 0.56-0.78) among case-control, and 0.92 (95% CI, 0.63-1.33) among cohort studies. Compared with nondrinkers, the pooled relative risks were 0.71 (95% CI, 0.57-0.89) for light (i.e. ≤1 drink/day) and 0.73 (95% CI, 0.60-0.87) for moderate-to-heavy (i.e. >1 drink/day) alcohol drinking. This meta-analysis suggests a favourable effect of alcohol on HL, in the absence, however, of a dose-risk relationship. The inverse association was restricted to--or greater in--case-control as compared with cohort studies. This indicates caution in the interpretation of results.