Women who assess their status at home after having a medical abortion are as likely as those who rely on a clinic follow-up to have a safe and complete medical abortion. In a randomized controlled noninferiority trial conducted among women living in rural and urban areas of Rajasthan, India, 93% and 95%, respectively, of participants in the clinic and home follow-up groups reported complete abortions without a need for further treatment.1 Overall, only 1% of women experienced a continuing pregnancy at follow-up, and 4–5% had incomplete abortions.

Medical abortion initiated by nine weeks’ gestation is widely recognized as safe and effective, yet many women in low-resource settings have limited access to the procedure. The common requirement that women attend a follow-up visit at a clinic diminishes both the access and the acceptability of medical abortion, because a return visit may be difficult for those with limited financial resources or low autonomy. The main purpose of a follow-up visit is to detect a continuing pregnancy, which occurs in about 1% of procedures using mifepristone and misoprostol. World Health Organization recommendations state that routine clinic follow-up is not medically necessary, and a number of studies—mostly in high-resource settings—have assessed the efficacy of having women use low-sensitivity urine pregnancy tests at home instead of requiring clinic-based testing to confirm that a complete abortion has occurred. This simple test has a high negative predictive value for detecting a continuing pregnancy, and hence carries an extremely low risk of missing one.

This noninferiority trial compared medical abortion outcomes among women who relied on either clinic follow-up or self-assessment at home. The study was conducted at three rural and three urban health centers in two districts of Rajasthan between April 2013 and May 2014. All clinics offered a range of reproductive health services and were staffed with specialists in obstetrics and gynecology. Women who chose to have a medical abortion were eligible for the study if their gestational age was nine weeks or less (as determined by bimanual pelvic examination) and they were 18 or older, agreed to a telephone or home follow-up after two weeks, had an adequate hemoglobin level and did not have a contraindication to the procedure.

Participants were randomly assigned to one of the two follow-up groups. After receiving 200 milligrams of mifepristone orally, women were given instructions on how to take 800 micrograms of misoprostol two days later (whether a woman took this drug at a clinic or at home depended on her provider’s assessment and her own preference); administration of misoprostol differed by clinic and could be sublingual, vaginal or oral. Participants assigned to clinic follow-up were asked to return 10–14 days after receiving mifepristone, whereupon a doctor or nurse assessed the abortion outcome and performed a low-sensitivity urine pregnancy test. Women in the home assessment group were given the same pregnancy test to be done 10–14 days later; they also received pictorial instructions on how to take and interpret the test, as well as information on symptoms of complications and how to contact the clinic if they experienced problems or had a positive or unclear result. Follow-up interviews with these women were conducted via telephone or home visit 12–15 days after mifepristone administration.

The primary outcome was complete abortion without need for surgical or further drug treatment, and secondary outcomes were the safety and feasibility of the procedure. The margin of noninferiority was a difference of 5% in the rate of unsuccessful abortions in each follow-up group; the reported rate of complete abortion with this drug protocol is 95%. Chi-square and t tests were used to identify significant differences in outcomes between the groups.

The 731 women who enrolled in the study had a mean age of 27, three-fourths lived in rural areas, more than half belonged to scheduled castes or tribes and had no formal education, nearly half were literate and almost all had had a previous birth. About a third had had a previous abortion, three-fourths of which had been medical abortions. Women in the two follow-up groups were similar across all of these characteristics. The analytic sample—women who used misoprostol and whose outcome was determined—consisted of 336 and 364 individuals, respectively, in the clinic and home assessment groups (95–96% of the full samples).

Complete abortion was reported for 93% of women in the clinic follow-up arm and for 95% of those in the home assessment arm. Pregnancy testing indicated that 1% of women in each group had a continuing pregnancy at follow-up, and 4–5% had incomplete abortions (defined as retained products of conception requiring surgical intervention or additional misoprostol administration); most women with either of these outcomes subsequently underwent surgical evacuation procedures.

Overall, the rate of adverse events was 0.3%. One woman in the clinic group required a blood transfusion and hospital admission for hemorrhage, and one in the home assessment group needed intravenous fluids because of hemorrhage. Fifteen percent of women who had a scheduled follow-up reported various side effects (e.g., heavy bleeding, severe abdominal pain, fever). Among the 289 home assessment participants whose outcome was not identified in an interim clinic visit, 81% took their pregnancy test before being contacted by a researcher, and 19% did so following contact. Of women who performed the test without a reminder, 98% said it was easy to use. Women were less likely to take the test if they had experienced a side effect, lived in a rural area, belonged to a scheduled caste or tribe, or were illiterate. Ninety-six percent of participants in both study arms were satisfied with their method of abortion follow-up.

The authors note that this study provides evidence that women in low-resource, low-literacy settings can safely and feasibly assess the outcome of their early medical abortion by using a low-sensitivity urine pregnancy test in their home, and that alteration of service delivery guidelines to allow such home assessment may be worth considering. Among the strengths of the study are that participants were representative of the underserved population in India, and that the health centers included were typical of the area’s abortion-providing facilities. The use of home-based pregnancy testing, according to the researchers, appears to be an effective method for assessing medical abortion outcomes, particularly “for women living in remote areas of developing countries [and] where an additional visit to the clinic could deter women from seeking services from safe legal providers.”—J. Thomas

REFERENCE

1. Iyengar K et al., Self-assessment of the outcome of early medical abortion versus clinic follow-up in India: a randomised, controlled, non-inferiority trial, Lancet Global Health, 2015, 3(9):e537–e545, doi:10.1016/S2214-109X(15)00150-3.