A questionnaire on pregnancy outcome was distributed to 782 women engaged in laboratory work at some time between 1968 and 1979. The response rate was 95%. Reported data on miscarriages, malformations and birth weights were compared to data in hospital records and central registers. Twelve per cent of the reported miscarriages could not be verified. Work with solvents was more frequently reported for these miscarriages as compared to those which could be verified or those which were anamnestically recorded. Fifty per cent of the malformations reported in the questionnaires were also found in registers or hospital records. Regarding birth weights, 28% of the infants had different birth weights reported in questionnaires than were found recorded in the Swedish Medical Birth Register. It is suggested that questionnaire data should be completed with information from sources having a higher validity when the influence of environmental factors on pregnancy outcome is studied.