Home pregnancy test sensitivity by test timing and cycle length with real-world use among pregnancy planners who...

To determine the sensitivity of home pregnancy tests with real-world use by day of testing and cycle length among pregnancy planners who conceived.

Pregnancy Study Online (PRESTO, 2018-2024) is an internet-based prospective cohort of pregnancy planners in North America who are not using fertility treatments to conceive.

We analyzed data from 19,908 pregnancy tests from 5,979 participants who detected a pregnancy with a home pregnancy test.

Participants reported day-specific information about pregnancy testing and test results from 4 days before expected menses through 4 days after expected menses. Participants also provided the date of their last menstrual period, their most recent menstrual cycle length, and cycle regularity.

We calculated pregnancy test sensitivity by timing of test in relation to expected menstrual period and cycle length. Because most participants did not test daily during the observed window, we imputed missing tests using participants' known test results and binomial distributions specific to the day of testing and cycle length determined by predicted probabilities from generalized linear models on the basis of observed tests. We present the mean probability of a positive test by day of testing and participant-reported cycle length.

Median cycle length was 28 days (interquartile range: 27-31 days) and 83.3% of participants reported regular cycles. Among participants with regular cycles, we found a relatively high probability of false negative tests with testing 2, 3, and 4 days before expected menses (12.7%, 21.4%, 46.4%, respectively). The probability of having a positive test increased with cycle day in a supralinear fashion, with over 99% of participants having a positive test by 4 days after the day of expected menses. Longer cycle length was associated with incremental increases in pregnancy test sensitivity, conditional on day of testing. Differences in test sensitivity by cycle length were most pronounced among early testers.

Real-world estimates of home pregnancy test sensitivity by day of testing and cycle length can empower individuals to make informed decisions about when to start testing. Sensitivity was lower than anticipated before expected menses. For any given day of testing, test sensitivity was higher for individuals with longer menstrual cycle lengths.