P007Accuracy of self-assessment of pregnancy duration eligibility for medication abortion among young people: The advantages of expanding screening questions beyond date of last menstrual period

Abstract

<h3>Objectives</h3> We examine young people's accuracy of self-assessment of gestational duration using an expanded set of screening questions beyond date of last menstrual period (LMP). <h3>Methods</h3> From October 2019 to March 2020, we surveyed patients seeking abortion at seven facilities and compared their self-assessments of gestational duration with measurement on ultrasound. Screening questions included estimated date of LMP, date they got pregnant and number of weeks pregnant. Using the 70-day threshold for medication abortion, we defined accuracy as area under the receiving operating curve (AUC) and compared AUC values between adolescents (aged 15–19, n=111) and young adults (aged 20–24, n=331) using Sidak-adjusted p-values. <h3>Results</h3> Non-response to the LMP date question was somewhat higher among adolescents (24.6%) vs. young adults (17.6, p=0.11). Accuracy (AUC) was similar for adolescents (0.80, 95% CI, 0.73–0.87) and young adults (0.82, 95% CI, 0.78–0.87) (p=0.55) using LMP date alone and assuming ineligibility for medication abortion (>70 days' gestation) if no LMP date was provided. Accuracy increased significantly, particularly among adolescents, when additional screening questions were utilized when LMP was missing. Adding the date they got pregnant marginally improved AUC to 0.83 (95% CI, 0.75–0.81) for adolescents (p=0.35). Adding number of weeks pregnant improved AUC for adolescents (0.86, 95% CI, 0.78–0.93) and young adults (0.86, 95% CI, 0.81–0.90) at p<0.10. <h3>Conclusions</h3> Young people are as accurate as young adults in dating pregnancy duration, particularly with screening questions beyond date of LMP. Young people should be offered expanded models of medication abortion care including no-test and telehealth options.