Medical abortion in Ghana: A non-randomized, non-inferiority study of access through pharmacies compared with clinics

Objectives

To compare self-reported clinical outcomes following medical abortion with mifepristone and misoprostol sourced from either a pharmacy or health clinic.

Study design

We conducted a prospective, non-randomized, non-inferiority cohort study across four regions in Ghana, from high-volume pharmacies and health clinics. Participants seeking medical abortion (less than nine weeks’ gestation) who met usual medical abortion eligibility criteria were recruited. Data collection included baseline surveys, follow-up phone interviews, and self-reported assessments of medical abortion outcomes. The study aimed to enroll 2000 medical abortion users (1000 from each source).

Results

Complete outcome data was available and analyzed from 1958 participants (of 2208 enrolled), with the adjusted risk difference of need for additional treatment to complete the abortion indicating non-inferiority of the pharmacy group compared to the clinic group [−2.3% (95% CI −5.3% to 0.7%)]. Both groups reported low rates of additional treatment (4.9%) and adhered similarly to the abortion regimen. Secondary outcomes showed no significant differences, with moderate acceptability in both groups (65.4% pharmacy, 52.3% facility). Adverse outcomes were rare: one ectopic pregnancy, one blood transfusion and no deaths or other major complications were reported.

Conclusions

Accessing medical abortion pills directly from pharmacies without prior consultation from a provider demonstrated non-inferior self-reported clinical outcomes compared to seeking care from health clinics. The findings align with the growing global evidence supporting the safety and effectiveness of medical abortion self-care.

Implications

This study contributes data which support future registration of over-the-counter use of medical abortion drugs up to nine weeks’ gestation. Such measures could expand options for safe abortion care, especially in regions where unsafe abortion poses a substantial maternal health risk.

Trial registration

ClinicalTrials.gov (NCT03727308).