Previous studies on ambulatory blood pressure monitoring as a potential screening test for hypertension in pregnancy have not carefully considered sampling requirements. We have examined the impact of duration and frequency of blood pressure sampling in the reproducibility of mean values in pregnancy. We analyzed 2430 blood pressure series sampled every 20 minutes during the day and every 30 minutes at night for 48 hours every 4 weeks from the first obstetric visit until delivery in 235 normotensive and 168 hypertensive pregnant women. Blood pressure series were decimated to generate shorter series with data sampled every 1, 2, 3, or 4 hours for 48 hours, as well as at the original rate for the first day. Reproducibility of mean blood pressure as well as sensitivity and specificity in the diagnosis of hypertension were compared between the original and the decimated series. Sensitivity and specificity of the 24-hour blood pressure mean are similar for the values calculated from the original series and for those obtained from shorter profiles up to data sampled every 3 hours but reduced by 5% to 12% when diagnosis is based on data sampled at 20- to 30-minute intervals for the first 24 hours. Results also indicate that the 24-hour blood pressure mean is better reproduced with data sampled at 3-hour intervals for 48 hours than by data sampled at 20- to 30-minute intervals for 1 day only. This study demonstrates that reproducibility of mean blood pressure values is more dependent on duration of sampling than on sampling rate.