Accuracy of a home-based device for giving an early estimate of pregnancy duration compared with reference methods

Abstract

Objective: To assess a home pregnancy test's accuracy to concurrently detect pregnancy and determine pregnancy duration.

Design: Multicenter, prospective study.

Setting: Study sites in the United States.

Patient(s): Women actively attempting to conceive who have menstrual bleeds (18-45 years).

Intervention(s): Volunteers collected early morning urine samples (three or fewer menstrual cycles). Pregnant volunteers underwent ultrasound dating scans. Ovulation day (LH surge +1 day) during pregnancy-resulting cycles was determined by quantitative measurement of LH. Random urine samples were tested with the hCG-measuring pregnancy test from 4 days before the expected period until 4 weeks later.

Main outcome measure(s): A home pregnancy test's accuracy in determining pregnancy duration compared with ultrasound and ovulation day.

Result(s): Agreement between pregnancy test results and time since ovulation was 93% (confidence interval [CI], 91.5-94.4). Agreement with ultrasound was dependent on the formula: there was 99% agreement when calculated with adjustment for Hadlock formula bias (Pexsters; CI, 98.2-99.4) or using a nonbias formula (Wu; CI, 98.6-99.6), when ultrasound error was accommodated. Agreement was lower when bias/measurement errors were not accounted for (Wu, 86%, CI, 83.9-88; Hadlock, 80.8, CI, 78.2-83.3).

Conclusion(s): This home pregnancy test provides an accurate estimation of pregnancy duration in weeks categories, 1-2, 2-3, 3+ weeks since ovulation, thereby showing utility in dating pregnancy.

Keywords: Pregnancy duration; human chorionic gonadotropin (hCG); ultrasound.