The study was conducted over a four month period in 2006 at a public sector tertiary hospital and two non-governmental organization (NGO) health care clinics located in the greater Cape Town area, Western Cape, South Africa. All three study facilities provide both first and second trimester abortion services and are broadly representative of the types of services available for women seeking a second trimester abortion in the Western Cape Province of South Africa. Services at the public hospital are provided free of charge whereas payments for services are required at the NGO facilities. These three facilities were selected as they all provide second trimester services using the dilation and evacuation (D&E) method. In the Western Cape Province, most second trimester abortions performed in the public and NGO sector use the D & E method. A few tertiary hospitals provide the medication method of abortion using misoprostol-alone regime which requires several days of hospital admission. These hospitals were not considered suitable for several reasons: the small numbers of second trimester abortions performed at these facilities made recruitment difficult, and, in addition, interviewing women while they were aborting, in discomfort and with limited privacy, was not deemed appropriate.
A total of 27 in-depth interviews were conducted with women seeking a second trimester abortion. Participants were selected through purposive sampling. Interviews were conducted either on the day of the abortion prior to the procedure or on the booking day according to the participant's preference. Misoprostol 400 – 800 micrograms was administered orally to all women prior to the D& E procedure to facilitate cervical ripening. Women at all three sites initially made a booking and then returned on a pre-scheduled day for the abortion. Gestational age was confirmed by ultrasound at the booking visit.
The study interviewers visited the three facilities on the days that women either made an appointment or on the day of the abortion. Women presenting for a second trimester abortion were identified by the facility staff according to the gestational age determined by ultrasound. These women were then approached by the interviewers and invited to participate in the study. Out of the 39 women approached, 12 refused to participate citing the need for privacy and/or physical and emotional discomfort as reasons for not participating.
In-depth interviews were undertaken with women seeking a second trimester abortion. Individual in-depth interviews were deemed more appropriate for women seeking an abortion due to the sensitivity of the subject matter and the need for privacy.
Interview guides were semi-structured, open-ended and probing. Key issues explored included: barriers to accessing abortion services including both personal and structural barriers; knowledge of the CTOP Act and the circumstances under which a woman can obtain an abortion; recognition and response to an unplanned pregnancy and contraceptive practices.
Interview guides and consent forms were piloted to check for language appropriateness and understanding. Experienced fieldworkers trained in qualitative research methods conducted the in-depth interviews in the participant's first language Xhosa, or English. Interviews were tape-recorded and transcribed verbatim. All participants provided written informed consent. Confidentiality and anonymity were ensured. Ethical approval was obtained from the Research Ethics Committee, Faculty of Health Sciences at the University of Cape Town. Approval to conduct the study was obtained from the Western Cape Provincial Department of Health and from the two NGO clinics.
Data were analysed using a grounded theory approach, based on a process that helps researchers to "discover" categories, themes and patterns that emerge from the data. Initial categories for analysing data were drawn from the interview guide and themes and patterns emerged after reviewing the data within and across respondent groups. The data was coded manually by two senior members of the research team. Trends and crosscutting themes were identified and issues for further exploration were prioritized for final analysis. Any coding discrepancies were resolved through discussion and consensus between research group members.