Determination of medical abortion success by women and community health volunteers in Nepal using a symptom checklist.

BACKGROUND

We sought to determine if female community health volunteers (FCHVs) and literate women in Nepal can accurately determine success of medical abortion (MA) using a symptom checklist, compared to experienced abortion providers.

METHODS

Women undergoing MA, and FCHVs, independently assessed the success of each woman's abortion using an 8-question symptom checklist. Any answers in a red-shaded box indicated that the abortion may not have been successful. Women's/FCHVs' assessments were compared to experienced abortion providers using standard of care.

RESULTS

Women's (n = 1153) self-assessment of MA success agreed with abortion providers' determinations 85% of the time (positive predictive value = 90, 95% CI 88, 92); agreement between FCHVs and providers was 82% (positive predictive value = 90, 95% CI 88, 92). Of the 92 women (8%) requiring uterine evacuation with manual vacuum aspiration (n = 84, 7%) or medications (n = 8, 0.7%), 64% self-identified as needing additional care; FCHVs identified 61%. However, both women and FCHVs had difficulty recognizing that an answer in a red-shaded box indicated that the abortion may not have been successful. Of the 453 women with a red-shaded box marked, only 35% of women and 41% of FCHVs identified the need for additional care.

CONCLUSION

Use of a checklist to determine MA success is a promising strategy, however further refinement of such a tool, particularly for low-literacy settings, is needed before widespread use.