Includes bibliographical references. / Short-term treatment outcomes of children starting ART in the ICU, general medical wards and outpatient HIV clinics at Red Cross War Memorial Children’s Hospital (RCWMCH): A Retrospective Cohort Study. Background: Antiretroviral therapy (ART) has proven to decrease morbidity and mortality in HIV-infected children and improve immunologic, virologic and clinical outcomes. As clinical management policies evolved, an emphasis on early infant testing was adopted resulting in an increasing number of children being diagnosed and commenced on therapy before the onset of severe disease progression. However, a fair proportion still remain untested and subsequently present to hospital with advanced immunosuppression and severe disease. Since the advent of the 2013 national Standard Treatment Guidelines which encourage expedited initiation of ART within 7 days of HIV diagnosis in all children under the age of 12 months and in those with advanced immunosuppression, it is likely that many HIV-infected children are being initiated on ART during hospitalisation in South Africa. No local published data on these outcomes exist. We assessed the short-term outcomes of children initiated on ART in the intensive care unit (ICU), general medical wards (GMWs) and outpatient HIV clinics (OHCs) at RCWMCH. Methods: Structured Literature Review A Pubmed search looking at outcomes of treatment naïve HIV-infected children and adolescents up to 19 years of age living in South Africa commenced on 1st line ART regimens in accordance to the national guidelines presiding at the time, over a 10 year period was performed. This served to identify gaps in knowledge around paediatric ART in a South African context warranting further research. Retrospective Cohort Study We conducted a retrospective cohort study of HIV-infected children <13 years of age, commenced on first line ART between January 2008 and December 2011 at RCWMCH. Outcome measures included death, virologic suppression and changes in CD4 count and percentage. Kaplan-Meier estimates, multivariate Cox proportional hazard ratios and logistic regression were used to estimate outcomes 6 months after ART initiation. Results: Structured Literature Review This review identified several knowledge gaps. One of these gaps, the treatment outcomes of children started on ART at different service levels within tertiary health care settings was addressed in our retrospective cohort study and described in section C of this dissertation Retrospective Cohort Study Seven hundred and forty-nine children were included: 106 were commenced on ART in the ICU, 509 in the GMWs and 127 in the OHCs. Four hundred and ninety-two (65.7%) children were <12 months old. Children in the ICU and GMW cohorts were significantly younger than the OHC cohort (median ages: 3 and 5 months respectively vs. 22 months) and had lower WAZ scores (-2.48 and -2.33 respectively vs -1.14). Three hundred and eighty-five (51.4%) children qualified for rapid ART initiation within 7 days of HIV diagnosis or hospitalisation, based on CD4 criteria in the 2013 national Standard Treatment Guidelines. Overall mortality was 6.4% (CI: 4.9 - 8.4). Mortality was significantly higher in the ICU cohort i.e. 14 (13.2%) deaths compared to 28 (5.5%) and 5 (3.9%) deaths in the GMWs and OHCs cohorts, logrank p=0.004. Predictors of mortality included being moderately underweight HR 2.4 (CI: 1.1 – 5.2; p=0.02), severely underweight HR 3.2 (CI: 1.6 – 6.5; p=0.001), absence of caregiver counselling sessions HR 2.9 (CI: 1.4 – 6.0; p=0.005) and ART initiation in ICU HR 2.6 (CI: 1.4 – 4.9; p=0.003). Conclusion: The findings of our retrospective cohort study serve as a basis for understanding the implications of ART initiation in children during hospitalisation.

Includes bibliographical references. / This study is a retrospective cohort analysis of patients undergoing two different treatment modalities (brachytherapy and endoresection) for medium sized choroidal melanoma. Study methods involve the collection of baseline and follow-up data from three sources: 1) A database collected by the department of Radiation Oncology at Groote Schuur Hospital; 2) Private physicians responsible for patient follow-up following brachytherapy; 3) Private physicians responsible for endoresection surgery and patient follow-up. To date there has been limited publication of the outcomes of patients treated for choroidal melanoma in South Africa. The study aims to compare the outcomes of these procedures to help identify the possible benefits of each form of treatment.

Includes abstract. Includes bibliographical references.

During the mid to late 1990's, nearly all HIV infected children admitted to South African paediatric intensive care units died. This was in the context of an increasing HIV epidemic in Sub-Saharan Africa, a limited number of intensive care beds in public hospitals and the South African government refusing to supply antiretroviral medication to public sector patients. HIV infected children all die without ARV medication, and it resulted in an increase in the South African under-5 mortality rate. In this context critically ill HIV infected children were often denied PICU admission. Developed countries introduced ARV medication in the early 1990's and the South African government only started supplying ARV medication in late 2003. When ARV medication became available in South Africa, it was started on the basis of the individual child's clinical and immunological status and there was not much published data on initiation of ARV therapy in critical ill children in intensive care units. Many HIV infected children had recurrent hospital admissions and many children died before initiating ARV medication. HIV infected children are not only susceptible to the normal bacteria and viruses, but at increased risk of opportunistic and mycobacterial infections. CMV has increasingly been recognized as a common co-infection with PCP, but has been difficult to diagnose and treat effectively. We retrospectively reviewed all HIV exposed and infected children admitted to our PICU in 2009. In addition to our standard treatment, we initiated ARV medication as soon as logistically possible and children with suspected CMV infections were empirically treated with gancyclovir.

Includes bibliographical references. / Sub-Saharan Africa accounts for 69% of the global HIV burden. Due to a variety of social, economic and behavioural factors, mentally ill patients are more likely to engage in high risk sexual behaviours. In turn, co-morbid substance use which is present in up to 75% of mentally ill patients is a leading risk factor for sexual risk behaviours. Worldwide methamphetamines are the most commonly used illicit stimulant. Both injectable and noninjectable methamphetamines have evidenced associations with high risk sexual behaviours. Smoking and inhalation of crystal methamphetamine is the predominant mode of use in South Africa. The use of crystal methamphetamine amongst mentally ill persons in Cape Town has escalated over the last decade. We aimed to determine the occurrence of methamphetamine use and risky sexual practices amongst mentally ill patients. In addition we aimed to explore the associations between methamphetamine use and HIV sexual risk behaviours in a sample of mentally ill inpatients in Cape Town, South Africa.

Includes abstract. / Includes bibliographical references. / Pulmonary tuberculosis (TB) as a cause of COPD is debated, with some, but not all evidence suggesting an association between the two conditions. Aim: To systematically review evidence for the association between pulmonary tuberculosis and the development of chronic obstructive pulmonary disease. We performed a systematic review of original English language, peer-reviewed literature using the PUBMED/MEDLINE database. Chronic Airflow Obstruction was defined on spirometric data (FEV1: FVC Ratio < 0.70; or FEV1: FVC Ratio < lower limit of normal for age, with or without bronchodilator use). Conclusions: Evidence was found for an association between a past history of tuberculosis and the presence of COPD. This association is independent of cigarette smoking. Causality is likely but cannot be assumed.

The reported research was an investigation of attitudes and beliefs associated with participation in screening programmes and clinical trials, carried out by general practitioners. Particular focus was given to cardiovascular risk-reduction. The work comprised two main studies. The preliminary study was entirely exploratory, designed to guage public attitudes towards GP involvement with preventive screening programmes and clinical research; and to identify the range of variables associated with participation in such projects. The subsequent study utilised a more formal approach in which the Behavioural Intention Model was utilised to evaluate the power of influencing factors. Both studies employed self-completion questionnaires, developed from preliminary in-depth interview data. For the first study instrument distribution was effected by personal approach, for the second study postal distribution was employed. In all, 1,037 respondents contributed to the surveys - 442 to the preliminary exploration and 695 to the follow-on study. These represented response rates of approximately 65% and 36% respectively. The main findings were that attitudes towards screening were generally favourable, though there was less conformity in attitudes expressed towards clinical trials. These findings were reflected in reported participatory intentions. No evidence was found of any factors which might pose widespread barr i ers to screen ing part ic i pa t ion, though some potent ia 1 deterren ts were identified for older women. It was also noted that other potential deterrents may have been masked by the 'middle class' bias of the sample. Major deterrents to trial entry were identified as worries about: sideeffects, acquired resistance, discontinuation of current effective medications and lack of adequate information. These all interacted with the 'guinea pig' factor. Response rates and responses associated with medical and non-medical sampling sources were also discussed; and consideration was made of the general utility of the Behavioural Intention Model for research of this kind.

Optimal antiretroviral treatment (ART) adherence is critical in achieving virologic suppression. Most people living with HIV (PLWH) experience HIV-related challenges which may be compounded by the experience of stressful life events (SLE) with adverse effects on their ART adherence and therefore affecting viral suppression. The highest concentration of people living with HIV (PLWH) is in low-and-middle income countries (LMIC) which are challenged with high rates of poverty and unemployment. Limited published data is available on the impact of the stressful life events experienced by PLWH in LMIC on ART adherence and viral load. A secondary analysis was conducted using data collected in a randomised control trial (RCT) which investigated the impact of a real-time electronic adherence monitoring device (EAMD) on ART adherence in ART-naïve individuals to assess the impact of SLE on ART adherence and viral load. Part A of this dissertation includes the study proposal/ protocol as approved by the Departmental Research Committee and the Human Research Committee at the University of Cape Town. Part B details the literature review which examined all published studies which report on stressful life events in PLWH, with ART adherence or a viral load as an outcome. The review included published literature from 2008 to 2019. Part C includes the publish-ready manuscript which details the statistical analysis, results and interpretation of the secondary analysis of impact of SLEs on ART adherence and viral load among 200 individuals living with HIV. Part D, appendices were included as supporting documentation necessary for the conduct of this research and as required for the completion of this dissertation. The American Psychological Association (APA) 6th Edition referencing style was used for Part A and B. The Vancouver referencing style was used for Part C as per the instructions for authors by the Aids and Behaviour journal guidelines.

The Clinical Infectious Diseases Research Group [CIDRI] has conducted high impact research over the last decade in Cape Town specifically in the townships of Khayelitsha and Mannenberg. None of this research would have been possible without robust strategies to recruit and retain study participants. Four different completed studies with different study designs have been selected, which will show the different approaches to participant recruitment into clinical research. This review will evaluate this process in relation to the approved protocol recruitment strategy, the amendments, which were required for modifications, the ability to retain participants to the end and the composition of staff used to achieve study outcomes. This entire process has been recognised as a necessary research skill and the term recruitmentology has become a practice pivotal to the research process. Recruitmentology has been unpacked to illustrate how minorities have been recruited, overlooked and over researched in the United States (US), and that experience has given a new perspective to the processes involved. Although in the South African context we do not have the identical issues to the US, these ideas can be translated in our circumstances, as both research populations can be considered as marginalised. We are challenged in the township of Khayelitsha with service disparities, which are generally impacted by the presence of clinical research groups. Although Khayelitsha has three large Day Hospital facilities, a newly built 150 bedded secondary level hospital and 11 local clinics, offering a consistently high standard of care; it remains a challenge. The CIDRI group partnered with the health services, supporting them with extra staff in the way of nurses, doctors and clinical research workers, while in return benefiting from the health system by being able to conduct effective studies. This has been and continues to be a mutually beneficial relationship, as CIDRI has been supportive to health services and the service has been a research partner of many research protocols including one of the studies being reviewed. Through the process of reviewing the databases of these four different CIRDI studies, we can examine the successes, challenges and a possible model of recruitment in the township of Khayelitsha. These studies have been chosen as they have been successfully completed by CIDRI and the databases have been locked. Each study has a different study design, from a pragmatic randomised control study, a cross sectional study, a seasonal follow-up and longitudinal study. Close attention will be paid to proposed recruitment strategy as per approved protocols, amendments (which impacted the recruitment process), staff structure, time frames of recruitment, retention and impact on study outcomes. This review will attempt to answer the following: 1. Was the proposed recruitment strategy followed as per study design and approved study protocol? 2. Was the overall recruitment impacted by staffing structure and allocated recruitment time frames? 3. How were study outcomes impacted by recruitment and retention? 4. Tuberculosis/Human Immuno-deficiency Virus TB/HIV were the diseases of study in all four studies, do these two diseases have specific challenges which impact recruitment and retention?

Includes synopsis. / Includes bibliographical references. / Cardiac auscultation is inherently qualitative, highly subjective and requires considerable skill and experience. Computer- assisted auscultation (CAA) is an objective referral-decision support tool that aims to minimise inappropriate referrals. This study evaluated the sensitivity and specificity of 2 CAA systems, Cardioscan® and Sensi®, in detecting echo-confirmed cardiac abnormalities in 79 consecutive patients referred for assessment to a tertiary cardiac clinic. CAA demonstrated suboptimal sensitivity and specificity in detecting cardiac abnormalities in children and adults. As both systems demonstrate 100% sensitivity in detecting acyanotic heart disease, and theoretically carry significant potential in resource-limited settings, further development of current technologies to improve sensitivity and specificity for clinical applications is still warranted.