Objectives The Seveso accident caused the contamination of a large inhabited area by 2,3,7,8-tetrachlorodibenzodioxin. Three zones with decreasing soil TCDD levels were delimited: A, B and R. Persons ever living in the contaminated zones and in a surrounding non-contaminated area were followed-up to evaluate long-term effects. The main finding of the mortality (1976–2001) and cancer incidence studies (1977–1991) was an increase in lymphohemopoietic cancers in zones A and B. We report preliminary results of the extension of the follow-up to 30 years after the accident for lymphoemopietic cancers.

Methods Incident cases were ascertained through the hospital discharge registration system of the Lombardy region and medical records were reviewed. RR and 95% CI were estimated with Poisson regression techniques controlling for age, gender and calendar period using the non-contaminated area as reference.

Results The incidence of all lymphohemopietic cancers was increased in zone A (6 cases, RR =1.2; 95% CI 0.5 to 2.7) and B (47 cases; RR=1.5; 95% CI 1.1 to 2.0). Non-Hodgkin's lymphomas were modestly increased in zone B, whereas the RR for all leukaemias was 2.0 (95% CI 1.2 to 3.4). 5 cases were lymphatic leukaemia (RR =1.6 95% CI 0.6 to 4.1); 10 myeloid leukaemias showed a RR of 2.1 (95% CI 1.04 to 4.15). All lymphatic leukaemias occurred after 20 years since the accident (RR=4.0; 95% CI 1.4 to 11.6). Myeloid leukaemia did not show a consistent pattern across different categories of time since the accident.

Conclusions The extension of the follow-up confirms an increased risk for lymphatic and haemopoietic cancers, particularly leukemia, in the Seveso population.