When the law makes doors slightly open: ethical dilemmas among abortion service providers in Addis Ababa, Ethiopia

BACKGROUND: In 2005, Ethiopia changed its abortion law to curb its high maternal mortality. This has led to a considerable reduction in deaths from unsafe abortions. Abortion is now legal if the woman’s pregnancy is a result of rape or incest, if her health is endangered, if the fetus has a serious deformity, if she suffers from a physical or mental deficiency, or if she is under 18 years of age. The word of the woman, if in compliance with the law, is sufficient to qualify for an abortion. In this context, where the law makes the door slightly open, health workers become important in deciding who gets access to safe services and who doesn’t, thus creating considerable ethical dilemmas. METHODS: The objective of this study was to explore abortion service providers’ personal experiences and reflections, perceptions of the abortion law, and ethical and dilemmas that arise. Data collection took place from March to May 2016 in Addis Ababa, at different health clinics providing abortion services. Thirty in-depth interviews and three focus group discussions were conducted with 41 abortion service providers at governmental and non-governmental clinics. Content analysis was drawn upon in the interpretation of the findings. RESULTS: When working in a context where the law has slightly opened the door for abortion seeking women, the health workers describe conflicting concerns, burdensome responsibilities, and ambiguity concerning how to interpret and implement the law. They describe efforts to balance their religious faith and values against their professional obligations and concern for women’s health and well-being. This negotiation is particularly evident in the care of women who fall outside the law’s indications. They usually handle ethical dilemmas and decision-making alone without guidance. Moreover, many health workers face a stigma from fellow colleagues not performing abortions and therefore keep their job a secret from family and friends. CONCLUSIONS: Health workers in Ethiopia experience ethical dilemmas trying to maneuver between the abortion law, their personal values, and their genuine concern for the health of women. More research is needed to further explore this. The study was conducted in Addis Ababa, the capital city of Ethiopia, a melting pot of different religions and ethnicities. The city stands out as the richest and most developed area in the country [20, 21]. Administratively, the city is divided into 10 sub-cities. The total fertility rate of 1.5 is half the national average [21]. Further, the city has the highest contraceptive coverage in the country at 50,1% usage of modern contraceptive methods [20]. Addis Ababa also has the highest registered abortion rate in the country estimated at 92 per 1000 women aged 14–49 years [17]. Investigation was carried out at four public health centres, two hospitals, and five non-governmental organizations (NGOs) clinics providing abortion as well as other health services such as family planning, post-abortion care and other gynaecological services. The abortion service providers met the women seeking induced abortion either through an elective appointment or at the emergency room. The various health facilities provided abortion services in different ways. At the public health centers, abortion was provided free of charge and was open Monday to Friday. At the NGO clinics the price varied with some providing abortion to a reduced price to poor women. They were also open Monday – Friday. The hospitals also provided free abortion services, but there women could seek an abortion by going to the gynecological emergency room which was open 24 h all days. There were commonly one or two nurses working as abortion service providers, although at the hospitals and NGO clinics a doctor often worked alongside the nurses with providing abortion services or would be called for when encountering challenging cases, such as second trimester abortions which only the doctors could perform. Commonly, the head nurse or doctor assessed the woman’s eligibility through a consultation where they would either accept or reject her request for an abortion. Before the abortion was conducted the doctor or nurse and the woman wishing to obtain safe abortion signed a consent form. The size of the study sites varied, at the public hospitals induced abortions were performed at the minor gynecological emergency room which consisted of three to four beds, while the health centers only had one small private room with one bed used for the procedure. The NGOs either had a whole department only for abortion services or it was mixed with general gynecological services. They always had several private rooms to perform the procedure. No official statistics on the abortion caseload per clinic was available but through the interviews we got to understand that the health centers performed the least abortions with three to ten per month, the hospitals seemed to perform more especially since second trimester abortions all had to be performed here. Though most abortions seemed to be performed at the NGO clinics who stated that they performed several hundred abortions per month. All clinics at times experienced a lack of staff and medicines. At all the clinics visited during the study, surgical abortion was reported to be more common than medical abortion, though this picture was reported to be changing. Data collection took place from March to May 2016. Participants were included if they worked with any aspect of induced abortion services provision at the time of the study including either provision of pre-abortion information and counselling, provision of abortion pills or manual vacuum aspiration (MVA) and post-abortion care. They were recruited by the first author (EM), with assistance from the co-author (DD), who works as a gynecologist in Addis Ababa. At the initial stage of the recruitment purposive sampling was used to ensure the inclusion of participants from different abortion service providing institutions and from different cadres of health workers. Later snowballing was employed to identify new participants. A total of 31 in-depth interviews (IDs) were conducted, of which three were follow-up interviews made to clarify important emerging topics. One interview took place with two people as the attendance of them both was requested by the participants. In addition three focus group discussions were conducted (FGDs) with two groups of five and one group of three participants. The participants were between 23 and 42 years old and had between a few months and 15 years of experience working with all aspects of abortion service provision from actually inducing the abortions to taking care of women undergoing an induced abortion. Of the abortion service providers, 19 were male and 23 were female. The majority, 24 participants, worked as nurses with nine having additional training in midwifery, five worked as doctors, three as health officers, three as medical students that had training in induced abortion and one pharmacist. The majority, 16 participants, considered themselves Orthodox Ethiopian Christian, three as Protestant, five as Christian, two as Muslims and six were religious without further specification. It must be noted that information about age, profession, years of working experience and religion was not obtained from all the participants as abortion is a sensitive topic and it was not always appropriate to ask about this. An additional file with more details about the characteristics of the study participants is provided (see Additional file 1). A semi-structured interview guide was employed. To ensure that the questions were as relevant as possible to the context, adaptations to the guide were made during the course of the fieldwork to incorporate emerging issues. The majority of the interviews were conducted in English by the first author (EM). The focus group discussions and seven individual interviews were conducted in Amharic by a research assistant trained in qualitative methodology (with EM present). The interviews took place in separate rooms at the health institution during breaks or after working hour. Two interviews were conducted at a restaurant following the participants request. Interview participants were recruited until a sense of saturation was reached, that is, when no major new themes emerged [22]. All interviews and focus group discussions were audio recorded. Preliminary analysis started during the fieldwork through a process of continuous reflection on the information emerging during the interviews. Moreover, field notes with reflections were written on a weekly basis. The interviews in English were transcribed by the first author (EM), while the Amharic interviews were transcribed and translated by a professional Amharic-English translator. Based on the patterns identified while in the field, content analysis was drawn upon for systematic analysis post fieldwork [23]. EM read through the full data set several times, getting a sense of the overarching themes, and divided the material into several meaning units, which were then discussed with the co-authors. The content within each unit was then condensed, coded, and sorted into categories. Finally, the categories were turned into generalized descriptions of the most common dilemmas, judgments, and reflections emerging from the material. These were supported by specific verbatim statements from the interviews. NVivo11 software was used in the process of coding and organizing the data [24]. Ethical approval was provided by the Initial Review Board and by the Institutional Review Board at Addis Ababa University, College of Health Sciences. Ethical approval was applied for from Regional Ethical Committee of Norway (REK), but the study was considered to be outside their mandate. General research ethics principles of anonymity, confidentiality, and rights of withdrawal without consequences were followed. The interviews took place in rooms where privacy was ensured and all the recordings were kept safe on a closed file in the authors computer. In cases where the participants had to travel to be part of the study, compensation for the travel was provided. All participants signed a consent form.