The purpose of the study was to evaluate the association between smoking during pregnancy and preterm birth. The overall rate of preterm delivery was 4.3%. Smokers had a 40% higher risk of preterm birth compared to non-smokers. A dose response relationship was found between smoking and risk of preterm birth. Adjustment for women's height, pre-pregnant weight, age of the mother, marital status, education, occupational status, and alcohol intake did not change the results. Among women with an intake of less than 400 mg of caffeine per day no difference in the risk of preterm birth between smokers and non-smokers was found. However, among women with an intake of more than 400 mg of caffeine per day, the risk of preterm birth was increased almost threefold among smokers compared to non-smokers. Furthermore, among women with a high intake of caffeine a dose response relationship was found between smoking and risk of preterm delivery.