OBJECTIVE Exposure to disinfection by-products (DBPs) during pregnancy has been associated with adverse birth outcomes. We evaluated exposure to DBPs through ingestion, inhalation and dermal absorption among pregnant women in Crete, in relation to birth weight and gestational age. METHODS The mother-child birth cohort in Crete ('Rhea' study) enrolled 1359 pregnant women at the third month of pregnancy (2007-2008), residents in the prefecture of Heraklion. Exposures were assessed through three questionnaires administered during pregnancy requesting extensive information on personal water-related habits. Tap water samples were collected in representative mother homes on the basis of detailed water distribution patterns, and were analysed for major DBPs including trihalomethanes (THMs). Logistic and linear regression models were applied. RESULTS Pregnant women reported a high consumption of bottled water at home (76%) and work (96%). More than half the women (59%) washed dishes by hand, nearly all women (94%) took showers rather than baths (1%), and only 2% attended a swimming pool. THM levels were low (<20 μg/l) with a high proportion of brominated compounds. When using quantitative estimates of residential exposure, we found no association with low birth weight (LBW, OR 0.7, 95% CI 0.4 to 1.4), small for gestational age for weight (SGAweight, OR 1.1, 95% CI 0.6 to 2.2) and preterm delivery (OR 0.8, 95% CI 0.5 to 1.3). Similar results were observed when taking into account uptake of THMs through all exposure routes. CONCLUSIONS We found no evidence for an increased risk of LBW, SGA and preterm delivery at the relatively low level exposure to THMs and particularly brominated THMs in Cretan drinking water.