<b>Background:</b> This is the first epidemiological study of internalised stigma among adolescents living with the human immunodeficiency virus (HIV) in Sub-Saharan Africa. It aims to establish predictors of internalized HIV-stigma among people living with HIV in Sub-Saharan Africa (Paper 1), develop an HIV-stigma scale for use with adolescents (Paper 2) and build and test a model of risk pathways for internalised stigma (Paper 3). The data used for papers 2 and 3 is part of the world's largest social science study of adolescents living with HIV (n=1060). <b>Paper One</b> systematically reviews evidence on the prevalence and predictors of internalised HIV stigma amongst people living with HIV in Sub-Saharan Africa. PRISMA guidelines were followed. An adapted version of the Cambridge Quality Checklist was used to assess the quality of the findings. A total of 18 papers were included. The prevalence of internalised stigma among adults living with HIV was 27% - 66%. The longitudinal predictors for internalised HIV stigma were poor HIV-related health and psychological distress. The review identifies two critical limitations of the literature. First, no studies on adolescents were found. One of the reasons for this may be the lack of a scale for measuring internalised HIV stigma in this population. Second, only individual-level risk factors for internalised stigma were examined. Papers 2 and 3 aim to address these limitations. <b>Paper Two</b> develops an HIV stigma scale with and for adolescents living with HIV. First, a multidimensional stigma scale previously used with adolescents in the US was cross-culturally adapted using semi-structured cognitive interviews with nine South African adolescents living with HIV. These data were interpreted through thematic analysis, and items were adapted in consultation with interviewees. Second, the revised version of the scale was administered to 1060 adolescents living with HIV. Confirmatory factor analysis confirmed the predicted 3-factor structure, and associations with hypothesised correlates provided evidence of validity. <b>Paper Three</b> develops and tests a model of risk pathways to internalised HIV stigma among adolescents living with HIV. Drawing on findings from the systematic review (Paper 1) and using the scale developed in Paper 2, both inter and intrapersonal pathways of risk from HIV-related disability to internalised HIV stigma were hypothesized. Following from modified labelling theory, interpersonal mechanisms were hypothesized to occur through maltreatment within power-unequal relationships, i.e. enacted HIV stigma and violence victimization. Hypothesized intrapersonal risks were anticipated HIV stigma and depression. Structural equation modelling enabled the grouping of theoretically related constructs and assessment of multiple, simultaneous pathways of risk. Prevalence of any internalised HIV stigma among adolescents living with HIV was 26.5%. As hypothesized, significant associations between internalised stigma and anticipated stigma, as well as depression were obtained. Unexpectedly, HIV-related disability, violence victimization, and enacted stigma were not directly associated with internalised stigma. Rather, indirect pathways via intrapersonal risks were observed. <b>Conclusions:</b> More than a quarter of adolescents living with HIV in this study reported experiencing some level of internalised stigma. Findings suggest a need to expand programmatic responses to internalised HIV stigma, from individualistic, clinic-based programmes to integrative, community-based approaches. Providing mental health support and reducing the maltreatment of adolescents living with HIV might interrupt pathways from HIV-related disability to internalised stigma. This highlights the potential for interventions that do not necessarily target HIV-positive adolescents but are sensitive to their needs. Such efforts must be coupled with rigorous process and outcome evaluations, and longitudinal data is urgently needed. It is hoped that the adolescent-friendly stigma scale developed within this DPhil will enable further research with this understudied population. Prior to this thesis, there were no known epidemiological studies of internalised HIV stigma among adolescents living with HIV. Moreover, the broader, adult-focused corpus of research has overlooked interpersonal risk factors. This thesis highlights the relevance of power inequalities and domination for the study of internalised HIV stigma.