Expanded Screening Questions Resulted in Higher Accuracy in Self-Assessment Pregnancy Duration

Published January 13, 2022

To be eligible for medication abortion, health care providers must determine how long someone has been pregnant. In a clinic, this can be done through an ultrasound or an exam but can also be done by asking a pregnant person the start date of their last menstrual period (LMP). While prior research shows that people seeking abortion early in pregnancy can accurately self-assess gestational duration using the date of their last LMP, uncertainty about the date causes nearly 1 in 5 people to be automatically disqualified from having a medication abortion without an ultrasound. We set out to find ways to reduce uncertainty through the cross-sectional Gestational Duration Dating study, conducted as part of our research on Over-the-Counter Medication Abortion Study.

For a quick overview of this study, watch this three-minute video.

Key Findings

Just under one in five (17%) people in the study were not sure of the date of their LMP. By comparison, fewer (3%) were not sure how many weeks pregnant they were, or whether they had missed more than two periods (6%).
Using responses to multiple pregnancy dating questions, versus relying on a single question, works better at correctly identifying people who are > 70 days (or 10 weeks) pregnant.
A composite of people’s responses to three questions – the date of their LMP, the date they think they got pregnant, and their estimated number of weeks pregnant – was more sensitive and accurate than LMP alone. With these three questions, just 2.7% of people would incorrectly screen themselves as ≤ 70 days when they are actually > 70 days.
Questions that do not rely on date of LMP also work. Responses to a simpler set of statements about whether people thought that they were more than 10 weeks pregnant, more than two months pregnant, or had missed more than two periods also performed very well, particularly when the goal was reducing the number of people who incorrectly screen as ≤ 70 days when they are actually > 70 days. Using responses to these three statements, just 2.3% of people would falsely screen as eligible when they were not.

Study Design

In this study, we surveyed 1,000 people seeking abortion. Survey questions included different ways of asking people how far along they were in their pregnancy, or their gestational duration. Some questions were short statements, designed to be similar to how questions might look on the Drug Facts Label in an Over-the-Counter medication abortion product. Other questions were structured as they might appear in a questionnaire that a person could complete on an app on their phone or in conversation with a health care provider.

Using people’s responses to these questions, we calculated how far along they estimated they were in pregnancy – their self-assessment of gestational duration. Then, we compared their self-assessment to the gestational duration measured on ultrasound at the clinic. We calculated how well the questions worked at identifying whether someone was above or below 70 days (or 10 weeks) of pregnancy, the current threshold for eligibility for medication abortion.

Implications 

The Gestational Duration Dating study demonstrates that expanding the number and types of questions asked during eligibility screening allows for more comprehensive, accurate, and sensitive self-assessment of gestational duration among people seeking abortion. Although in-person ultrasound is typically used to establish gestational duration and eligibility for medication abortion, there is growing interest in expanding the use of telemedicine screening and removing ultrasound requirements. This data supports those efforts. 

"Our findings suggest that policies requiring in-person ultrasound for dispensing of medications such as the mifepristone are not universally necessary to establish gestation-based eligibility for medication abortion."

Based on this study's findings, we developed an interactive tool that demonstrates how different combinations of questions work in identifying people above or below 70 days gestation, available here.

The article, Accuracy of self-assessment of gestational duration among people seeking abortion, is available in the American Journal of Obstetrics and Gynecology.

Areas of Research