We investigated lymphoma risk following hepatitis infection in a case-control study of 274 incident lymphoma cases, defined according to the WHO classification, and 336 population controls in Sardinia, Italy. Part of our study population (198 cases and 219 controls) was included in the EPILYMPH study of Hepatitis C virus (HCV) infection in relation to non-Hodgklin's lymphoma risk. Based on questionnaire information on whether and at what age a diagnosis of hepatitis was posed by a physician, systematic anti-HCV antibodies testing in cases and controls by enzyme-linked immunoassay, and HCV-RNA assessment by PCR analyses in positive samples, we investigated more in detail whether hepatitis non-C is also associated with lymphoma risk, and whether risk varies by clinical form of hepatitis (acute or chronic infection). After adjusting by age, gender, education, and area of birth whether from the study area or elsewhere in Italy, a previous generic diagnosis of hepatitis was associated with a significantly elevated lymphoma risk [odds ratio (OR) = 1.8; 95% CI 1.1, 2.8], which was equally increased for hepatitis B (OR = 1.8; 95% CI 0.9, 3.5), for HCV positive subjects overall (OR = 2.0; 95% CI 0.8, 4.8), and for hepatitis non-B non-C (OR = 1.6; 95% CI 0.7, 3.9). Once concurrent infection from other hepatitis viruses was excluded, acute or chronic hepatitis C was the only one showing a consistent risk increase in all lymphoma subtypes, but follicular lymphoma. Some indications of an excess risk of lymphoma were observed also for acute, but not chronic forms of hepatitis B and hepatitis non-B, non C. Self-limited hepatitis C did not show an association. No significant heterogeneity in the risk of major lymphoma subtype was observed. Our results confirm a role of either acute or chronic active HCV infection in lymphomagenesis. Further studies are warranted to test the hypothesis that acute infection from other hepatitis viruses might also increase lymphoma risk.