This cross-sectional study investigated the relation between intake of caffeine-containing beverages and time to conception in a population of 1,909 married women in New Haven, Connecticut, between May 12, 1980 and March 12, 1982. Women were interviewed shortly after the first prenatal visit regarding the length of time taken to conceive the index pregnancy, consumption of caffeine during pregnancy, and other exposures occurring prior to and during pregnancy. In logistic regression analyses, intake of caffeine from coffee, tea, and caffeinated soft drinks was associated with an increased risk of a delay of conception of 1 year or more. Compared with no caffeine use, consumption of 1-150 mg/day of caffeine resulted in an odds ratio for delayed conception of 1.39 (95% confidence interval (CI) 0.90-2.13), consumption of 151-300 mg/day of caffeine was associated with an odds ratio of 1.88 (95% CI 1.13-3.11), and that of over 300 mg/day (the equivalent of approximately three cups of coffee) resulted in an odds ratio of 2.24 (95% CI 1.06-4.73), after controlling for last method of birth control used, parity, and number of cigarettes per day. When the risk of conception for each cycle was examined using a discrete analogue of the Cox proportional hazards model, women who reported drinking over 300 mg/day of caffeine had a 27% lower chance of conceiving for each cycle, and those who reported drinking less than 300 mg/day had a 10% reduction in per cycle conception rates compared with women who consumed no caffeine. Risks for coffee, tea, and colas were examined simultaneously in logistic models and were found not to improve the fit of a model that contained a variable for total caffeine intake from all sources.