A thesis submitted to the School of Public Health, Faculty of Health Sciences, University of Witwatersrand in fulfilment of the requirements for the degree of Doctor of Philosophy Johannesburg, South Africa 2017. / Introduction: As Prevention of Mother-to-Child Transmission and HIV treatment programmes have scaled-up, more women are being kept alive and fewer children are infected with HIV. One of the challenges that HIV-infected women face is how to disclose their own HIV status to their children. The disclosure literature suggests that the main reason for women’s HIV-disclosure is to obtain social support, including financial, practical and emotional assistance, to help women to manage their HIV infection. HIV-related stigma, lack of knowledge of how to disclose, uncertainty about a child’s reaction to disclosure, and a perception that a child lacks the developmental capacity to handle HIV-disclosure are factors that affect disclosure decisions. HIV-disclosure is considered a critical element in strengthening the capacity of families in the continuum of HIV care. In 2011, the World Health Organization published guidelines for parental HIV-disclosure to children. These guidelines recommended full disclosure to primary school-aged children (from 6 years up to 12 years) and partial disclosure to younger children. Globally, but particularly in sub-Saharan Africa, HIV-disclosure interventions are lacking. The Amagugu maternal HIV-disclosure intervention was developed, piloted and evaluated between 2010 and 2012 to assist mothers to disclose their HIV status to their HIV-uninfected children. The intervention enrolled 281 mothers and their primary school-aged HIV-uninfected children, and aimed to increase maternal capacity to disclose their HIV status. This was implemented through six lay counsellor- led, home-based, sessions, including a health intervention at a primary health care clinic. The Amagugu intervention was acceptable and feasible in a high HIV prevalence, resource-poor, rural setting, and increased maternal HIV-disclosure to primary school-aged HIV-uninfected children. Methods: This PhD study was nested within the Amagugu study and was conducted at the Africa Centre for Population Health, now the Africa Health Research Institute (AHRI), in the Hlabisa sub-district of Umkhanyakude, northern KwaZulu-Natal, South Africa. The aim of this PhD was to explore, in more depth, the experiences of women enrolled in the Amagugu intervention, including their experiences of the health intervention. Both quantitative and qualitative methods were used. The sample for the PhD study comprised three groups: 1) all mothers from the Amagugu study (N=281); 2) a sub-sample of mothers from the Amagugu study (N=20) and 3) health care staff employed in the clinics where the Amagugu study took place (N=87). The quantitative data used in the PhD study xvi were collected at different time points from the 281 women, using questionnaires specifically designed for the Amagugu study, including baseline and post-disclosure questionnaires. Data on the clinic experiences were collected from the 281 mothers using semi-structured questionnaires administered after the health intervention. Semi-structured questionnaires, specifically designed for this PhD study, were administered to the clinic staff during the health intervention. Qualitative data were collected using semi-structured questionnaires with the clinic staff (N=87), nine focus groups with clinic staff after the health intervention, and in-depth and semi-structured interviews with the sub-sample of women (N=20) who were enrolled after the Amagugu intervention had been completed. The University of KwaZulu-Natal Biomedical Research Ethics Committee (BREC Ref: BF 144/010) and the University of Witwatersrand Human Research Ethics Committee (Ref: R14/49) granted ethical approval for the PhD study. Results: The results reported in this PhD are drawn from four papers written during the course of this PhD, (three published papers and one paper accepted for publication). The results have been integrated from the data collected from the women and health care staff that were used for the PhD, and a literature review that resulted in a publication. The three main themes that emerged from the PhD are: 1. HIV-related stigma and HIV-disclosure: The literature review revealed that fear of HIV-related stigma was the most common reason for non-HIV-disclosure of HIV to both adults and children. In the sub-sample of 20 women, only two women had not disclosed to other adults due to fear of HIV-related stigma prior to the intervention. Those who had disclosed to only some, but not all other adults in their close social networks, reported HIV-related stigma at household (6/18), community (2/18) and clinic levels (1/18). Although HIV-related stigma was reported, there was also a normalisation of HIV and antiretroviral therapy (ART) in some communities due to the high social exposure to HIV in the study area. Qualitative data from the focus groups with clinic staff, and interviews with the sub-sample of 20 women, recognised that children’s exposure to HIV education has played a role in the normalisation of HIV. 2. HIV-disclosure and family strengthening: The majority of women in the sub-sample had disclosed their HIV status to other adults including their partners, friends and xvii other relatives, prior to the Amagugu intervention. Most women reported living positively with HIV and receiving necessary support from those to whom they had disclosed. Of those women who had disclosed to partners, about half had disclosed to their partners first before disclosing to any other adults. Their partners had mixed reactions to disclosure but were overall supportive. Whilst women’s original fear of disclosing their HIV status to their children had been that they would be stigmatised by their children, on the contrary, most children were supportive. The majority of women in the sub-sample expressed that they would advise other women in similar circumstances to disclose their HIV status to their children for social support, because disclosure increased family cohesion and improved antiretroviral therapy adherence. The health staff echoed the same sentiments regarding social support and family cohesion in the focus groups. 3. HIV-disclosure and access and adherence to HIV treatment: The interviews with the sub-sample of 20 women, and the semi-structured questionnaires and focus groups with 87 clinic staff, all revealed that participants agreed that whilst maternal HIVdisclosure was challenging, it was necessary for women to obtain social support from their children. Health care staff also agreed that maternal HIV-disclosure to their children was instrumental in supporting HIV-infected women to access and adhere to their HIV treatment. The clinic staff recognised the role they could play in health promotion and increasing opportunities for children to participate in activities at health facilities, but acknowledged that they needed support to address logistical constraints that hinder child-friendliness in health facilities, including heavy workloads, poor clinic infrastructure and staff shortages. The health intervention provided clinic staff with child-friendliness training and materials that were found to be acceptable and feasible, and yielded encouraging results. Discussion: This study contributes to the literature about the experiences of African, HIVinfected, rural women with HIV-uninfected children living in an ART-era. In particular the experiences of women who have participated in a maternal HIV disclosure intervention in Africa have not been explored previously. The findings of this work indicate that a decade after being diagnosed with HIV, women in this resource-poor setting are generally living positively with HIV. However, HIV-disclosure to other adults does not necessarily translate xviii to disclosure to children, and parents require specific interventions to assist them with this, and to understand the development and level of understanding of their children. Conclusion: Despite concerns raised by women prior to the intervention, including fear of HIV-related stigma and a perception that children lacked the developmental capacity to grasp knowledge about maternal HIV status, the women in this study reported no regrets in disclosing their status to children. They also reported receiving support from their children, which in turn, assisted them with adherence to their own HIV treatment. Future studies could test the same Amagugu intervention materials in a group of HIV-infected women with uninfected children not previously involved in research to explore whether similar results are found. / MT2017