OBJECTIVES: The objective of this pooled analysis was to examine whether exposure to DDT was associated with the risk of non-Hodgkin's lymphoma among male farmers. METHODS: Data from three case-control studies from four midwestern states in the United States (Nebraska, Iowa, Minnesota, Kansas) were pooled to carry out analyses of 993 cases and 2918 controls. Information on use of agricultural pesticides and other risk factors was based on interviews. Non-farmers (people who had never lived or worked on a farm) were used as a reference category. RESULTS: There were 161 cases and 340 controls who reported use of DDT on animals or crops, or on both, yielding an odds ratio (OR) of 1.2 (95% confidence intervals (95% CI) 1.0 to 1.6). Farmers who had used DDT for > or = 15 years had an OR of 1.5 (95% CI 1.0 to 2.3). Adjustment for respondent status and use of other pesticides resulted in slightly reduced ORs. Analyses by the number of days of use a year was limited to the Nebraska data. The most notable increase was found among farmers who used DDT for > or = 5 days a year (OR 2.6, 95% CI 1.1 to 5.9); however, additional adjustment for use of organophosphates, phenoxyacetic acids, and the individual pesticides lindane, malathion, and atrazine reduced the ORs to 1.0, 0.9, 1.1, 1.6, and 1.9 respectively. CONCLUSIONS: No strong consistent evidence was found for an association between exposure to DDT and risk of non-Hodgkin's lymphoma. It seems that the excess risk initially found may be explained by use of other pesticides.