Background. Few population-based studies have estimated alcohol consumption and binge drinking before and during pregnancy. Methods. The study is a longitudinal questionnaire study. In Norway, virtually 100% attend an ultrasound screening at 17–18 weeks of pregnancy. In Oslo, a representative sample attending this between June 2000 and May 2001 were invited to join the study, accepted by 92%. Non-Norwegian-speaking and/or immigrants from non-Western countries were not invited; 1,749 (93%) completed the first questionnaire. Measures: T-ACE (screening for pregnancy risk drinking), frequency of alcohol use, Standard Units (SU) per occasion, SU/week, and binge drinking (≥5 SU per occasion). Binary logistic regression analyses, Student's t-test and Pearson's chi square were used. Results. Alcohol use was reported by 89% pre-pregnancy and by 23% after pregnancy week 12. Binge drinking was reported by 59% pre-pregnancy and by 25% during weeks 0–6. Change of drinking pattern occurred at pregnancy recognition for 85%, although 78% had planned the pregnancy, fetal welfare being the main reason. Participants reporting alcohol use during pregnancy were older, smokers, with more SU/week pre-pregnancy, had some elevated anxiety score, higher income, and a partner with higher education. However, mental health, high income, or education were not predictive of usually drinking 1 SU per occasion or more, and not of binge drinking. The latter was predicted mainly by smoking during pregnancy and being T-ACE positive. Conclusions. The study shows a drinking pattern before pregnancy recognition that, according to other studies, in 25% or more may cause suboptimal fetal development.