Menu
Menu
Menu
Menu
Ghana has made progress in expanding providers in abortion care but access to the service is still a challenge. We explored stakeholder perspectives on task-sharing in abortion care and the opportunities that exist to optimize this strategy in Ghana. We purposively sampled 12 representatives of agencies that played a key role in expanding abortion care to include midwives for key informant interviews. All interviews were audio recorded, transcribed verbatim, and then coded for thematic analysis. Stakeholders indicated that Ghana was motivated to practice task-sharing in abortion care because unsafe abortion was contributing significantly to maternal mortality. They noted that the Ghana Health Service utilized the high maternal mortality in the country at the time, advancements in medicine, and the lack of clarity in the definition of the term “health practitioner” to work with partner nongovernmental organizations to successfully task-share abortion care to include midwives. Access, however, is still poor and provider stigma continues to contribute significantly to conscientious objection. This calls for further task-sharing in abortion care to include medical or physician assistants, community health officers, and pharmacists to ensure that more women have access to abortion care.
The study was implemented in Ghana, which has a population of about 25 million.20 About 20% of women of reproductive age (15–49 years) have ever had an abortion.21 In 2017, 53 114 abortions occurred and, of these, 13 918 were characterized as unsafe. Nonmedical methods (e.g. drinking milk/coffee/alcohol/other liquid with sugar, drinking a herbal concoction, drinking other home remedies, using a herbal enema, inserting a substance into the vagina, heavy massage, excessive physical activity, and use of all kinds of unknown tablets) used to induce abortion make up more than 27% of abortions carried out.21 More than one in 10 pregnancy‐related deaths occur as a result of an unsafe abortion and for every woman who dies from an unsafe abortion it is estimated that 15 suffer short‐ and long‐term morbidities.21 Countrywide estimates may mask regional differences; for example, 14% of pregnancies among women in urban areas end up in an induced abortion compared with 7% among women in rural areas. Women in poor and rural communities in northern Ghana have less access to comprehensive abortion care; for instance, only 3% of women in the northern part of the country have ever had an induced abortion compared with 22% of women in the more urban middle and coastal areas.21 This was an exploratory, descriptive study designed to gain insights into the policy decision to include all cadres of health workers with midwifery skills in the provision of abortion care, and to learn about stakeholder opinions regarding opportunities for further expansion. We purposively sampled individuals in the public and private sector who had themselves contributed (or whose agencies had contributed) to the policy on the expansion of abortion care to include midwives. Selection was based on an individual’s or agency’s role in advocating for the policy or contributing to the policy framework, implementation, monitoring, or evaluation. Respondents were typically heads of the agencies but, where necessary, members who were more familiar with the agency’s role in task‐sharing in abortion care were invited for the interview. Identification of the agencies was primarily via snowball sampling. The lead researcher (RAA) conducted most of the interviews. A research assistant (EK) with more than 2 years’ experience in conducting qualitative interviews assisted him. In all, 12 key informant interviews were conducted. On two occasions, the study team had to talk to two individuals from an agency to better understand the agency’s contribution to the policy shift. Consent was sought from the respondents to audio record the interviews. The interviews lasted 1–1.5 hours. All interviews were conducted in June 2018. All interviews were audio recorded and transcribed verbatim. The transcripts were imported into NVivo version 11 software (QSR International; Melbourne, Vic., Australia) for coding and thematic analysis. We predetermined codes using the interview guide and additional codes were developed for concepts that were not initially captured by the guide but emerged inductively from the data. We segmented the data into similar groups to form preliminary categories of information or themes on the expansion of health worker roles in abortion care. We examined the segments of data related to each theme and where necessary refinements were made. RAA coded all transcripts and ES coded six of those transcripts separately. The two coding sets were compared to ensure validity. Discrepancies were discussed and coding was adjusted where necessary. A coding comparison query to determine the inter‐rater reliability returned a Kappa coefficient of 0.84.
N/A
