Randomized Trial Assessing Home Use of Two Pregnancy Tests for Determining Early Medical Abortion Outcomes at 3, 7 and 14 Days after Mifepristone

Contraception; August 01, 2016; Vol. 94(2):115-121; doi: https://doi.org/10.1016/j.contraception.2016.04.001

Objective: To evaluate the accuracy, feasibility and acceptability of two urine pregnancy tests in assessing abortion outcomes at three time points after mifepristone administration.

Study design: This randomized trial enrolled women seeking early medical abortion at two hospitals in Vietnam. Investigators randomly allocated participants to at-home administration of a multilevel urine pregnancy test (MLPT) or a high sensitivity urine pregnancy test (HSPT) to assess their abortion outcomes. A baseline test was administered on the same day as mifepristone. Participants performed and interpreted results of pregnancy tests taken 3, 7 and 14 days after mifepristone. Ultrasound exam determined continuing pregnancy.

Results: Six hundred women enrolled, and 300 received each test. A percentage of 97.4 (584) had follow-up, of whom 13 women had continuing pregnancies. The specificity of MLPT at detecting absence of continuing pregnancy was 63.9%, 90.4% and 97.1% at study day 3, 7 and 14. The specificity of HSPT was 6.0%, 19.8% and 62.2%, respectively. The positive predictive value (PPV) of MLPT at detecting continuing pregnancy was 6.4% at day 3 and rose to 46.7% at day 14. In contrast, the PPV for HSPT was 2.2% at day 3 and rose to 6.5% at day 14. At all three time points, the sensitivity and negative predictive values for both tests were 100.0%. Most women found their assigned tests easy to use and would prefer future home follow-up with a pregnancy test.

Conclusions: The MLPT enables women to assess their abortion outcomes more reliably than with HSPT. With MLPT, women can know their outcomes as early as 3 days after mifepristone.

Implications: Medical abortion service delivery with an MLPT to obtain a baseline (preabortion) human chorionic gonadotropin (hCG) estimate and a second follow-up MLPT 1 to 2 weeks later can establish whether there has been a drop in hCG, signifying absence of a continuing pregnancy. Used this way, MLPTs can enable women to assess their abortion status outside of a clinic setting and without serum hCG testing and/or ultrasound.