Objectives: Despite warnings to abstain from alcohol, American women who are or could become pregnant still drink. This study evaluates whether women who consume alcohol are at an increased risk of recognizing pregnancy later than women who do not, adjusting for confounding factors that have been associated with alcohol consumption during pregnancy. Methods: The sample included 863 control women from a multisite case-control study conducted from 1996 to 2002 in the United States and Canada. Telephone interviews were conducted with mothers by trained nurse interviewers who administered standardized questionnaires on demographic and reproductive factors, and pregnancy exposures. Results: Alcohol consumption was classified as none (42.0%), occasional (31.9%), regular (15.6%), and heavy (10.5%). Time to recognition of pregnancy was calculated as the date pregnancy was suspected minus the last menstrual period date (median: 31 days; range: 7–227 days). Unadjusted Cox proportional hazard models found that regular drinkers, but not heavy drinkers, had a significantly higher risk of recognizing pregnancy later than non-drinkers. However, this association went away after adjustment for demographic factors. Among women with unplanned pregnancies, heavy alcohol intake was associated with a 45% increased hazard ratio, compared to 0.80 for women with planned pregnancies; however, this finding was not statistically significant. Conclusions: While time to pregnancy recognition did not vary among drinkers and non-drinkers, results from this study reiterate previous findings that pregnant women consume alcohol, and that drinkers share social and demographic characteristics that could be used to target public health interventions.