Incidence and safety of abortion in two humanitarian settings in Uganda and Kenya: a respondent-driven sampling study

Abstract

Background: Access to abortion is a fundamental human right. The need for abortion services is amplified in complex humanitarian emergencies. However, most humanitarian agencies do not provide abortion services. There is a lack of data on the direct experiences of abortion of those living in displacement. This study aimed to describe abortion practices, safety, and incidence in two refugee settings.

Methods: Between March and October 2022, we surveyed 1201 women and girls with recent abortion experiences (past 5 years) from Bidibidi Refugee Settlement, Uganda, and Kakuma Refugee Camp, Kenya, using respondent-driven sampling (RDS). Participants completed an interviewer-administered survey. Population-based estimates of abortion experiences were weighted using the RDS-II estimator to account for the sampling design. We used the sequential sampling population size estimation method to estimate annual abortion incidence. We also conducted a health facility assessment of 27 facilities (16 in Bidibidi, 11 in Kakuma) to describe the availability of facility-based abortion services in these communities.

Findings: Among those with an abortion in the past 5 years, the most common methods of abortion were traditional herbs (81% in Bidibidi, 45% in Kakuma) and non-medication abortion pharmaceuticals such as painkillers and antimalarials. Few participants reported using WHO-recommended methods of abortion (mifepristone in combination with misoprostol, misoprostol alone, or manual vacuum aspiration). Self-reported morbidity was high. Nearly a quarter reported avoiding seeking post-abortion care. The estimated annual abortion rate was 52 per 1000 in Bidibidi (95% simulation interval 20-106) and 55 per 1000 in Kakuma (95% simulation interval 19-119). Only 5 of 27 health facilities (1 of 16 in Bidibidi, 4 of 11 in Kakuma) reported providing safe abortion services. 15 of 16 in Bidibidi and 9 of 11 in Kakuma reported providing post-abortion care.

Interpretation: Refugees in these two contexts have little access to WHO-recommended methods of abortion, and the need for safe abortion services is high.

Funding: This study was funded by Elrha's Research for Health in Humanitarian Crises (R2HC) Programme. Elrha aims to improve health outcomes by strengthening the evidence base for public health interventions in humanitarian crises. R2HC is funded by the Foreign, Commonwealth & Development Office of the United Kingdom, Wellcome Trust, and the UK National Institute for Health Research.

Keywords: Abortion; Abortion incidence; Humanitarian settings; Refugee; Respondent-driven sampling.

Fig. 1

Legend: ∗ Kenya and Uganda country estimates of abortion incidence are from Bearak and colleagues. Error bars for country estimates correspond to 95% uncertainty intervals (UI). Kenya estimated abortion incidence is 43 abortions per 1000 women of reproductive age (95% UI 29–61). Uganda estimated abortion incidence is 43 abortions per 1000 women of reproductive age (95% UI 29–60). +Kakuma and Bidibidi estimates of abortion incidence were computed using the successive sampling population size estimation approach using data from the present study. Error bars for Kakuma and Bidibidi estimates correspond to 95% simulation intervals (SI) over 1000 Markov chain Monte Carlo simulations. Kakuma estimated abortion incidence is 55 abortions per 1000 women of reproductive age (95% SI 19–119). Bidibidi estimated abortion incidence is 52 abortions per 1000 women of reproductive age (95% SI 20–106).