Project Summary The goal of this qualitative study was to generate a better understanding of relationship processes with male partners that affect women’s prevention of mother-to-child HIV transmission (PMTCT)-related health behaviours through a critical examination of gender and power. This study was part of a larger concurrent mixed-methods parent study on the relationship between gender power dynamics within heterosexual couples and women’s PMTCT adherence. From March to August 2014, a cross-sectional survey was administered to 320 postpartum women living with HIV attending well-child paediatric healthcare visits at a large public health centre within a densely populated, low socio-economic neighbourhood of Lusaka. A convenience sub-sample of 32 participants in the parent study was invited to also participate in a semi-structured qualitative interview. The goal of the interviews was to expand on and explain the quantitative survey findings regarding the relationship between gender power dynamics and PMTCT-related health behaviours. Data Overview Participants were recruited during routine paediatric healthcare visits (e.g., child immunisations, height and weight measurements). Women were eligible for participation if they were married or cohabiting with a male partner, HIV-positive, over 18 years of age (legal age to provide consent for research in Zambia), and had a biological child between 3 to 9 months of age. Infant age criteria were meant to capture the essential PMTCT protocols, match the paediatric immunisation schedule, and limit recall bias. Because a major focus of the parent study was on intimate partner violence (IPV), as a safety measure, we excluded any women who were at the clinic with their male partners; only one woman was excluded for this reason. Nurses at the clinic determined eligibility for the parent study using the child’s “Under-Five Card” (i.e., a mother’s copy of her child’s health record that she is required to bring to all healthcare visits) or other available medical records. Eligible women were consented by research staff and received a small travel reimbursement. All survey participants were invited to stay and participate in a semi-structured interview immediately after the survey on the same day in the same location. Interviews were conducted by experienced, trained local Zambian research assistants in the most commonly spoken languages (English, Nyanja, Bemba, Tonga) using a semi-structured interview guide. The interview guide included broad, open-ended questions regarding PMTCT experiences and gender power dynamics. All research assistants had qualitative public health experience and participated in a three-day training. Data analysis and recruitment occurred concurrently and continued until the research team agreed we had achieved theoretical saturation of themes informing how gender power dynamics affect women’s PMTCT-related health behaviours. Throughout data collection, memos were kept in order to create a rich description of the data and to identify any needed changes to the interview guide, as well as establish theoretical saturation. Interviews were audio-recorded, translated and transcribed verbatim into Microsoft Word, and imported into Atlas.ti for analysis. The codebook was developed and applied to the transcripts by the primary investigator (Dr. Hampanda) using a combination of a priori codes from the interview guide and emergent codes. The author began with initial, line-by-line coding of transcripts to identify meanings and assumptions within the data, as well as comparisons between the codes and participants. In the final stages of analysis, focused coding by two of the investigators (Dr. Hampanda and Dr. Mweemba) explored the underlying meanings of the participant narratives and how they add to, form, transform, or reflect gendered social structures and processes in relation to women’s HIV care during and after pregnancy. We applied Fairclough’s method of critical discourse analysis, which emphasises how participant narratives are linked to societal and cultural processes and structures. Our critical discourse analysis interrogated the transcripts by paying attention to issues of explicit and implicit gender power dynamics and how participants navigated these in the context of PMTCT care. A table of excerpts is included with one short excerpt from the interview transcripts for every code used in the qualitative coding. Full transcripts are not shared to conform to assurances made to participants during informed consent. Publicly sharing the full transcripts would violate the agreement to which the participants consented. more...