Mabetha, Khuthala2022-12-142022-12-142021https://hdl.handle.net/10539/33769A thesis submitted in fulfilment of the requirements for the Doctor of Philosophy in Demography and Population Studies to the Faculty of Humanities, School of Social Sciences, University of the Witwatersrand, 2021Background Non-orphaned kinship care is a family arrangement in which children are raised by relatives in the absence of their biological parents who are alive but are unable to provide parental care. In South Africa, many non-orphaned children live in households where aunts, uncles, or grandparents play the primary caregiver role. Literature has shown that 65% of children (both orphaned and non-orphaned) who do not live in the same household as their biological parents, reside with their grandmothers, close to 20% reside with aunts, 6% with additional extended kin and only 1% reside with non-related adults. Although kinship care is associated with a number of benefits, the spatial diffusion of children to grandparents and other extended kin has been found to be coupled with negative child health outcomes. The issue of under-five mortality is a major public health concern that has been largely debated in the fields of Demography, Public Health and Social Sciences. Under-five mortality is defined as a death that has occurred between birth and the fifth birthday. Several scholars have made considerable efforts to establish individual, household and community-level factors that explain under-five mortality. Scholars have identified maternal individual, household and community-level characteristics such as maternal education, maternal age, breastfeeding status, wealth status and place of residence as the main contributors of under-five mortality. Under-five mortality has received considerable attention in existing demographic literature in relation to family structure although the mortality risks of children raised in alternative care contexts, particularly non-orphaned children raised by extended family members, remain largely unknown.Therefore, this study aimed to investigate under-five mortality in kinship care to examine the role of extended kin on child survival, especially given how the living arrangements of kin caregivers and the kinship system in itself, plays a pivotal role in influencing child health outcomes. This was done by examining how sociodemographic factors, health-seeking behaviours and familial characteristics of kin caregivers influence under-five mortality among children raised in non-orphaned kinship care in South Africa. Methodology This study employed an explanatory sequential mixed-methods design. This research design involves using both quantitative and qualitative research approaches in two distinct phases. The first phase is the analysis of quantitative data, followed by the collection and analysis of qualitative data. The outcome variable in this study was under-five mortality and children who 2 had died before the observation period were right censored. For the quantitative phase, secondary data obtained from the 2014/15 to 2017 (Wave 4 to 5) South African National Income Dynamics Survey (NIDS) was analysed. The weighted sample size of under-five children (using appropriate NIDS sampling weights) who were living with a kin-caregiver in 2014/15 (Wave 4) was 126,859. These children were then followed to investigate their probability of dying in the year 2017 (Wave 5). The weighted sample size of under-five deaths obtained among children raised by kin in Wave 5 was 7,761. Event History Analysis was employed in the study using a series of Kaplan-Meier limit product estimators and Cox Proportional Hazard Regression models. In the follow up qualitative phase, in-depth interviews that explored the influence of kin caregiver health-seeking behaviours and familial characteristics on child survival were conducted. This was conducted with 24 kin caregivers, from the Eastern Cape and KwaZulu-Natal provinces between the period February – April 2019, to help explain and provide an in-depth understanding of the quantitative results. Various methodological strategies were employed in order to ensure validity and reliability of the qualitative findings in order to complement the quantitative findings. Results and Findings The Kaplan-Meier estimates in the quantitative phase showed that there are differences in the hazards of under-five mortality by type of kin caregiver. The hazard of death was highest among children raised by grandparents (almost 80%), followed by children raised by aunts or uncles (>75%). Lower mortality hazards were observed among children who were raised by other extended family members. Key results that emerged from the adjusted Cox Proportional Hazard Regression models showed that grandparent and aunt or uncle perceived health status and age had a significant influence on under-five mortality among children in their care. Results showed that there were 22% increased hazards of death among children whose grandparents perceived their health to be excellent or very good. Further, there were 10% increased hazards of death among children whose aunts or uncles also perceived their health to be excellent or very good. Children whose grandparents were 60 years and above had 22% increased hazards of dying before age 5. Amongst children living with aunts or uncles between the age groups 30-34 there was a 6.87 times higher hazard of death. Perceived health status of non-orphaned under-five children was a child characteristic that had a significant influence on under-five mortality among children raised by grandparents and aunts or uncles. Mortality risks were highest among children who were perceived to have excellent or very good health by their grandparents and aunts or 3 uncles. Perceived health status had a significant influence on under-five mortality. Among children raised by grandparents it was 22.9 times higher and 48% higher among children raised by aunts or uncles who perceived the children’s health to be excellent or very good. Qualitative findings emerging from the thematic analysis adequately explained and corroborated the quantitative results. Various health-seeking practices and beliefs in particular notions and perceptions held by kin caregivers pertaining to health status and illness such as increased reliance in traditional medicine and traditional healers, notion of witchcraft and faith healing were significant impediments that contributed to kin caregivers not accessing healthcare services for the children under their care. In addition, the family environment has helped provide an understanding of the child’s health outcomes as the characteristics of families emerging from the findings have emerged as important determinants of health. Also, the way the family functions and the support and care, or lack thereof, that family members provide to each other influences the families’ subsequent life course outcomes. Specifically, among children raised in such a family institution. Conclusions The findings of this study show that the survival of a child is influenced greatly by complex interactions that occur at the individual and family level. Importantly, the findings suggest that the individual characteristics of kin caregivers, healthcare preferences and behaviours and their family environments are risk factors. These risk factors play a crucial role in compromising the ability of kin caregivers to provide adequate care. These factors also affect caregivers’ ability to provide a conducive caregiving environment to the children under their care, rather than the kinship system itself. These circumstances expose children placed under such care to various risks that may be detrimental to their health and development. Policy and Research Recommendations Kin caregivers need to be provided with adequate counselling and support services to enable them to feel equipped and ready enough to assume the caregiving role. Child welfare authorities need to achieve this by equipping kin caregivers with the necessary knowledge and skills needed to provide adequate care to children. This will ultimately foster positive developmental outcomes. In addition, kin caregivers can be provided with informative training videos that educate them on how to handle situations pertaining to childcare. Further research, in particular longitudinal research, should be conducted to examine the multiple transitions or placement instability in kinship care. Such studies must also investigate 4 the subsequent effect this may have on child developmental outcomes and wellbeing. Such research will provide an important groundwork in understanding this complex relationship and the life trajectories of non-orphaned children who experience these familyenThesis