Human immunodeficiency virus (HIV) can result in posttraumatic stress disorder (PTSD) as events related to illness act as traumatic stressors. This study tested some basic hypotheses of Sewell and Cromwell's personal construct model of PTSD in HIV+ women both with and without diagnoses of PTSD. Trauma-related constructs of HIV+ women with PTSD with HIV+ non-PTSD controls at varying stages of illness were compared. The elaboration, rankings, and valence of trauma-related constructs were examined using the Life Events Repertory Grid (LERG) procedure. Findings provided evidence that a clinical diagnosis of PTSD in women was not associated with the degree of construct elaboration. These findings may imply a qualitative difference in cognitive processing of social stressors and violent stressors.

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Master of Science. Johannesburg, 2014. / Determining e ective treatment strategies for life-threatening illnesses such as HIV is a signi cant problem in clinical research. Currently, HIV treatment involves using combinations of anti-HIV drugs to inhibit the formation of drug-resistant strains. From a clinician's perspective, this usually requires careful selection of drugs on the basis of an individual's immune responses at a particular time. As the number of drugs available for treatment increases, this task becomes di cult. In a clinical trial setting, the task is even more challenging since experience using new drugs is limited. For these reasons, this research examines whether machine learning techniques, and more speci cally batch reinforcement learning, can be used for the purposes of determining the appropriate treatment for an HIV-infected patient at a particular time. To do so, we consider using tted Q-iteration with extremely randomized trees, neural tted Q-iteration and least squares policy iteration. The use of batch reinforcement learning means that samples of patient data are captured prior to learning to avoid imposing risks on a patient. Because samples are re-used, these methods are data-e cient and particularly suited to situations where large amounts of data are unavailable. We apply each of these learning methods to both numerically generated and real data sets. Results from this research highlight the advantages and disadvantages associated with each learning technique. Real data testing has revealed that these batch reinforcement learning techniques have the ability to suggest treatments that are reasonably consistent with those prescribed by clinicians. The inclusion of additional state variables describing more about an individual's health could further improve this learning process. Ultimately, the use of such reinforcement learning methods could be coupled with a clinician's knowledge for enhanced treatment design.

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree in Master of Science in the School of Molecular and Cell Biology August 2017 / Antiretroviral therapy (ART) drugs such as stavudine (d4T) are known to have off-target side-effects, including the inhibition of DNA polymerase gamma which replicates mitochondrial DNA (mtDNA). ART-induced depletion of mtDNA copy number may cause mitochondrial toxicities such as sensory neuropathy (SN). Genetic variation in DNA polymerase gamma or in other nuclear genes influencing mtDNA replication and mtDNA copy number may therefore contribute to susceptibility to d4T-induced SN. DNA samples from 263 HIV-positive South African adults exposed to d4T were classified as cases with SN (n = 143) and controls without SN (n = 120). A total of 28 single nucleotide polymorphism (SNPs) were chosen in nuclear genes from the mtDNA replication pathway and from a GWAS paper examining SNP association with ART-induced SN (Leger et al. 2014). Genotyping was performed using Sequenom Mass Spectrometry. MtDNA copy number was determined using a qPCR assay. Associations between SN and genetic variants, between genetic variants and mtDNA copy number, and between mtDNA copy number and SN were evaluated in univariate and multivariate models using Plink v1.07 and GraphPad v7. Age and height were significantly different in the cases with SN vs controls without SN. In univariate analyses, three SNPs and two haplotypes were significantly associated with SN, three SNPs were associated with pain intensity and three haplotypes were significantly associated with mtDNA copy number. However, there were no significant associations with SN, pain intensity or mtDNA copy number after correction for multiple SNP testing. No significant difference in mtDNA copy number in cases vs. controls was observed. In conclusion variation in nuclear-encoded mitochondrial genes examined in the current study do not play a role in ART-related mitochondrial complications such as changes in mtDNA copy number, or occurrence of SN. / MT2018

Executive Function is conceptualized in this study as the ability to form (the planning functionality obtained through initiation and working memory), maintain (response selection and the ability to self-regulate and inhibit) and switch (cognitive flexibility, mental tracking, organization and sequencing) mental processes in order to effect a positive outcome. The present research is a quasi-experimental study embedded in the Positivist tradition that sets out to empirically evaluate the Executive Function profile of seropositive adolescents (n = 29) emerging from a low socio-economic background and currently on a managed ART programme when compared to a healthy contrast group (based on age, socio-demographic and educational system). As a quantitative study, Executive Function was operationalized through the use of multiple tests of Executive Function such as the Delis-Kaplan Executive Function Colour Word Interference Test (D-KEFS CWIT), the Wisconsin Card Sorting Test (WCST) and the Trail Making Test Part B (TMT-B). As the study formed part of a larger study that included additional neurocognitive tests, including the WISC-R, selected subtests from the WISC-R were used to validate specific arguments relating to the study. The results showed that HIV positive adolescents were inclined to have poorer Executive Function performance especially under situations of higher cognitive load when compared to the unaffected group. The implications of these results are discussed in this research.

A thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy in the Political Studies Department, Faculty of Humanities, University of Witwatersrand, Johannesburg, South Africa 8 March 2015 / HIV is the most pressing public health and development challenge facing Botswana. Reducing gender-related vulnerability to HIV is one of the top priorities of the government and its development partners. Civil society organisations (CSOs) have been identified as crucial in these efforts. As a result, civil society has grown in Botswana, in both numbers and size, to deliver services such as home-based care, counselling, and testing. Yet to reduce gendered vulnerability to HIV, social and human development goals must be met in several sectors of society. The focus on HIV-related services has implications in practise, policy, and theory that may compromise long-term development aims and co-opt civil society. This research draws on critical theory and uses action research methods to investigate the role of civil society in Botswana for reducing gendered vulnerability to HIV, now and in the future. The case of Botswana is a crucial one, as it has one of the highest HIV prevalence rates, as well as the resources, both domestic and from partners, to mobilise a comprehensive response. The combination of these factors has afforded the opportunity to gain insights to inform civil society theory and development approaches in both policy and practise to improve the HIV response and civil society’s role in it. Through a literature review, interviews with key informants, a survey, and a workshop, this research found that the HIV response in Botswana is addressing many of the issues suggested by global development partners, such as UNAIDS, at the policy level, though implementation is lacking, especially concerning male involvement in gender programming. It found that efforts to meet the immediate needs are in place, but the long-term strategic interests are only incrementally addressed. This suggests that HIV is causing a development deficit. Additionally, the roles that CSOs serve in the response are focussed on serving these immediate needs, making it increasingly difficult for the response to effect broader social change to achieve gender equality and development. Civil society is taking on more responsibility in the public sector, which puts it in a vulnerable position. Its role needs to be reconceptualised in the HIV response and in development more broadly. This research proposes theoretical and policy implications to inform civil society-state relations; approaches to address complicated social development issues, such as genderbased violence; and offers an 18-point analytical framework to address operational and programmatic capacities in civil society. The framework offers a new category for the dynamic analysis of civil society organisations while working with the state called ‘civil agents’. It also describes the bridge function that CSOs serve when working with key populations, such as sexual minorities, in criminalised settings. Together these theoretical and policy implications can contribute to the understanding of civil society in the HIV response, and gender equity in the context of the post-2015 global development agenda. Key words: Civil Society Organisations, Non-governmental Organisations, HIV, Gender, Botswana, Development, Critical Theory, Action Research / MT2016

With 10% of the population being infected with Human Immunodeficiency Virus (HIV), South Africa has the highest number of infections in the world (StatsSA, 2013). HIV results in cognitive and motor deficits in children as the severe compromise of the immune system leads to neurodevelopmental dysfunction peri-natally (Ruel, Boivin, Boal, Bangirana, Charlebois, & Havlir, 2011). Neurocognitive deficits affect overall general intellectual abilities and include difficulties with attention and speed of information processing, verbal language, executive –abstraction, complex-perceptual motor function, memory and motor and sensory function (Dawes & Grant, 2007). Developmentally, it is evident that working memory provides a crucial interface between perception, attention, memory and action (Baddeley, 1996; Baddeley 2003). Therefore the purpose of the study was to investigate the working memory profiles of both an HIV positive children and a control sample, on cognitive tasks (Automated Working Memory Assessment), general intellect tasks (Raven’s Colored Progressive Matrices) and language competence tasks (Sentence Repetition Test). The current study compared 26 HIV positive children (mean age = 6.58 years) to 26 matched controls (mean age = 6.73 years). It was found that both non-verbal IQ and language proficiency were correlated to HIV status and thus were used as covariates in the study. MANCOVA’s were conducted on the data and produced findings that showed that there were only significant differences in visuo-spatial short-term memory between the two groups. Furthermore, it was also found that there were significant differences between the groups on nonverbal IQ and language proficiency. Therefore, the results showed that HIV may have an overall effect on non-verbal ability and language proficiency and a few aspects of working memory such as visuo-spatial short-term memory. Together with future studies focused on larger sample sizes and children who are not currently on HAART, early developmental interventions can be formulated to assist South African HIV-infected children so that the neurocognitive effects are lessened and their overall lifestyle is improved.

In 2009 an estimated 33 million people were living with the Human Immunodeficiency Virus (HIV). Of this global population, 35% live in South Africa. Furthermore, sub-Saharan Africa is home to 80% of the world’s population of HIV-1 positive children and adolescents. The most prominent form of transmission of HIV in children in South Africa is from mother to child. Until 2004, South Africans had limited access to ARV treatment at and after birth due to the government legislation. As a consequence, treatment of HIV in children may only have been initiated after clinical presentation of immune deficiency. Therefore, currently, HIV-1 positive adolescents born during the period of restricted ARV-access may have experienced physical and developmental symptoms associated with the virus including neurological deficits, prior to initiating treatment. This study investigated the current psychomotor functioning, such as psychomotor speed, manual dexterity, graphomotor and visual-motor coordination of a group of low socio-economic HIV-1 positive adolescents in Johannesburg, South Africa, who are now on a managed antiretroviral programme and how this compared to a HIV negative contrast group. A Mann-Whitney U Test indicated a significant difference in mean non-dominant hand performance in the Grooved Pegboard Test between the two groups (U = 738, p < .05), with the HIV positive group performing slower than the HIV negative group. An independent samples t-test indicated a significant difference between groups in the Block Design subtest of the WISC-R [t(88) = -2.93, p < .01] where the HIV positive group performed significantly worse than the HIV negative group. Additionally, a Mann-Whitney U Test revealed a significant difference in number of errors made in the WISC-R Mazes subtest between groups (U = 736.50, p < .05), where the HIV negative group made more errors. Another Mann-Whitney U Test revealed a significant difference between groups in the ROCFT Copy score (U = 534.50, p < .01) where the HIV positive group achieved a significantly lower score than the HIV negative group. Lastly, a Mann-Whitney U Test demonstrated significant differences between the groups in the Trail Making Test A time (U = 445.00, p < .01), Trail Making Test B time (U = 509.00, p < .01), the number of errors made on the Trail Making Test B (U = 729.00, p < .05) and the difference between Trail Making Test B – A time (U = 769.50, p < .05) with the HIV positive group performing slower and making more errors in Part B than the contrast group. The findings of the current study imply that HIV-1 vertically-infected adolescents in Johannesburg, South Africa, on a delayed HAART programme appear to have persisting difficulties in complex psychomotor skills where an integration of functions is required. Furthermore, these results indicate an overall poor psychomotor performance in comparison to international normative data, supporting previous findings. Developmental, remedial and therapeutic recommendations were made.

South Africa has the most prominent percentage of individuals living with the Human Immunodeficiency Virus (HIV) in the world, with the most prominent form of transmission of HIV in South Africa being vertical mother-to-child transmission. From 1997 until 2004, South Africa had limited access to ARV treatment at and after birth due to the government legislation. As a consequence, treatment of HIV may only have been initiated after clinical presentation of immune deficiency. A paucity of information therefore exists regarding this population in addition to the specific age demographic of adolescents. Adolescents may be negatively influenced by the cortical thinning associated with HIV, and this study therefore aims to investigate the verbal fluency and vocabulary (in English) of 30 bi- or multilingual seropositive adolescents that are currently on a managed anti-retroviral programme in comparison to an HIV-negative contrast group of 70 bi- or multilingual adolescents in South Africa (matched for age, education, and socioeconomic status). The study found that there were no significant results between the HIV-positive and HIV-negative groups on the measures of vocabulary, semantic naming, or phonemic naming in ‘F’ as determined by their performance on the neuropsychological assessments. Significant results were noted between the HIV-positive and HIV-negative groups on the phonemic naming categories of ‘A’ and ‘S’ however, and negative correlations between performance in these categories and current viral load, and viral load at Highly Active Antiretroviral Therapy (HAART) initiation were also noted. This research formed part of a broader study examining the overall neurocognitive effects of HIV-1 infection in adolescents in South Africa.

Approximately 11.5 million Human Immunodeficiency Virus (HIV)-positive individuals were living in South Africa in 2007, many of whom were infected via mother-to-child transmission. The current study aimed to compare the attentional and concentration functioning of 30 seropositive adolescents on managed anti-retroviral (ARV) programmes, with a comparable group of 71 seronegative adolescents. The results showed that the uncorrected errors on trial 1; self-corrected errors on trial 2; time taken, uncorrected and self-corrected errors on trial 3 of the Stroop Colour-Word Interference Test; and the errors on the Trail Making Test Part B were significantly poorer in the seropositive sample. The results also indicated that the clinical variations in the HIV-positive sample, including the age at which ARVs were commenced; duration of ARV treatment; World Health Organisation (WHO) stage at diagnosis; starting and current CD4+ counts; and starting viral load, but with the exception of the current viral load, impacted significantly on test performance.

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of Science in Genetics. Johannesburg, South Africa 2017 / Background: Stavudine (d4T) use is associated with the development of sensory neuropathy (SN), several mechanisms may underlie d4T-induced toxicity, including: (1) Inter-patient genetic variability in the genes modulating the deoxynucleotide triphosphate (dNTP) pool sizes. (2) Variation in intracellular ARV drug concentrations due to genetic variation in drug transporters. In our study we examined the genetic variation in four stavudine transporter genes and seven genes regulating the deoxythymidine triphosphate (dTTP) synthesis and their associations with d4T-induced SN or CD4+ T cell count or mtDNA copy number. Methods: We examined a cohort of HIV-positive South African (SA) adults exposed to d4T, including 143 cases with SN and 120 controls without SN. 26 single nucleotide polymorphisms (SNPs) from the literature were chosen, prioritised on being tagSNPs with minor allele frequency >5% in Kenyan Luhya (a proxy population for the SA Black population); SNP functional effects and suitability for multiplex analysis on the genotyping platform. Genotyping was performed using Sequenom mass spectrometry. A qPCR assay was used to measure the mtDNA copy number. Association of sensory neuropathy, CD4+ T cell count and mtDNA copy number with genetic variants was evaluated using PLINK. Results: All 26 SNPs were in Hardy-Weinberg equilibrium (HWE) in both the cases and controls. SNP rs8187758 of the SLC28A1 transporter gene and a 3-SNP haplotype ABCG2 were significantly associated with CD4+ T cell count after correction for multiple testing (p = 0.043 and p=0.042 respectively), but were not significant in multivariate testing. No SNP remained significantly associated with SN or mtDNA copy number, after correction for multiple testing. Conclusion: Variation in genes encoding molecular transporters of d4T may influence CD4+ T cell counts after ART. This study presents a positive step towards achieving personalized medicine in SA. / MT 2018