Adolescents estimate gestational duration as accurately as older counterparts

Key takeaways :

• Comparable accuracy was seen between all age groups using last menstrual period date and number of weeks pregnant.
• False-negative rate was low using the date of last menstrual period alone for adolescents.

Adolescents seeking an abortion demonstrated comparable accuracy in estimating gestational duration as older age groups, according to a research letter published in JAMA Pediatrics.

“We examined whether adolescents’ self-assessment of gestational duration using an expanded set of pregnancy dating questions was of comparable accuracy to that of young adults and older adults,” Lauren Ralph, PhD, MPH, an epidemiologist and assistant professor in the department of obstetrics, gynecology and reproductive sciences at the University of California, San Francisco, and colleagues wrote.

Researchers recruited 1,039 women (mean age, 26.5 years) aged at least 15 years who sought medication or procedural abortion, spoke English or Spanish, with no prior ultrasonography at one of seven geographically diverse abortion facilities in the U.S. Researchers collected demographic information, reproductive history and current pregnancy duration, including the first day of their last menstrual period, number of weeks since their last menstrual period, date they became pregnant and the number of weeks pregnant. All participants completed self-assessment of gestational duration, and these responses were compared with gestational duration measured on transvaginal or abdominal ultrasonography.

Of all participants, 111 were adolescents aged 15 to 19 years, 331 were young adults aged 20 to 24 years and 597 were older adults aged at least 25 years.

Adolescents were more likely to not report the date of their last menstrual period (24.6% vs. 17.6%; P = .11) and the number of weeks since the first day of their last menstrual period (20% vs. 14.2%; P = .15) compared with young adults. Both adolescents and young adults had similar rates of nonresponse to the number of weeks pregnant (4.5% vs. 3.9%; P = .68).

The area under the receiving operating characteristic (AUROC) values were good and comparable between adolescents, young adults and older adults using the last menstrual period date (0.8 vs. 0.82 vs. 0.8, respectively; P = .7) and the number of weeks pregnant (0.87 for all; P = .99). The AUROC values were also good and comparable between adolescents, young adults and older adults when relying on the last menstrual period response first and, when not available, relying on the number of weeks pregnant (0.86 vs. 0.86 vs. 0.82, respectively; P = .45). Using the date of their last menstrual period alone, the proportion of false-negative findings — when estimated gestational duration was within the window for legal abortion but measured duration was not — for adolescents was low at 1.8%. However, false-positive findings were more likely (24.3%) among adolescents when other screening questions were used.

When relying on the number of weeks pregnant first or, if not available, relying on the date of last menstrual period, researchers noted more balanced values on the proportions of false-negative and false-positive results for adolescents (4.5% and 5.4%, respectively), young adults (5.7% and 4.5%, respectively) and older adults (4.1% and 4.4%, respectively).

“Despite study limitations, including a small sample of adolescents aged 15 to 17 years and a focus on only one contraindication to medication abortion use, these findings support access to medication abortion to adolescents using history-based screening models,” the researchers wrote.