Background: The HIV/AIDS epidemic is continuing to impact negatively on the health of populations globally. The region most affected by this epidemic is sub-Saharan Africa, with two-thirds of people living with HIV/AIDS located in this region. In order to combat this epidemic, countries are putting up efforts to scale-up the provision of antiretrovirals to affected individuals. South Africa faces challenges similar to other sub-Saharan countries in the scaling-up of antiretrovirals, such as cost of drugs, inadequate laboratory infrastructure and the low and declining number of health professionals. Therefore, the contribution of private general practitioners as part of the overall strategy to combat the HIV/AIDS epidemic needs to be considered. However, the performance of antiretroviral programmes in the private sector are unknown and there is therefore a need to explore patient outcomes in antiretroviral programmes being run by general practitioners in the private sector. The findings of this study will add to understanding what the outcomes of ARV programmes by GPs in private practice could be in a public-private partnership. Aims: The main aim of the study was to evaluate patient outcomes in an antiretroviral programme in the setting of a solo private general practitioner. Methods: This was a cross-sectional study using retrospective data from March 2005 to February 2008. A total of 170 patient records from a private general practitioner’s rooms were examined. The data were analysed using the using the median, interquartile and proportions. The Chi square was used to test for association between baseline characteristics and patient outcomes in the univariate analysis. Results: The increase in median CD4 count was 242 cells/μl and 265 cells/μl at 6 months and 12 months respectively. The proportion of patients who achieved viral suppression at 6 months and 12 months was 75% and 73% respectively. Rate of loss to follow-up was 29% and 38% at 6 months and 12 months respectively, which was much higher than in other settings. Although there were no statistically significant associations established by the study, the patterns emerging from the study showed that the baseline characteristics that were a risk for poor virological outcomes during the period of the study were female gender, younger age group, higher WHO clinical staging, lower CD4 count and higher viral load. Baseline characteristics that were a risk for loss to follow-up during the period of the study were male gender, younger age group, higher WHO clinical staging, lower CD4 count and higher viral load. Conclusion: Patient outcomes in an antiretroviral programme in the setting of a solo private general practitioner are comparable with patient outcomes in antiretroviral programmes in public sector settings in terms of immunological and virological outcomes. However, the lost to follow-up rate was much higher than in public sector settings.

Background: The government of South Africa rolled out free anti-retroviral treatment in 2004 but many people living with HIV still present late for treatment while others choose to die rather than accessing this free treatment. This qualitative study was done at Potchefstroom Provincial Hospital Wellness Clinic to establish why many people living with HIV in Potchefstroom present late for treatment. Aim: To establish why people living with HIV/AIDS in Potchefstroom delay in accessing antiretroviral therapy. Objectives: 1.To conduct interviews with selected patients, in order to understand why they delayed accessing antiretroviral treatment. 2. To assess the demographics of patients who delayed in accessing antiretroviral treatment. Methods: This is a qualitative study carried out at the Wellness clinic of the Potchefstroom provincial hospital in the North West province of South Africa. Eight adult participants (3 females and 5 males) were interviewed individually in English, in one-on-one free attitude interviews and the conversations were audio-taped by the researcher. All the respondents were from the nearby black township of Ikageng. The respondents were selected after meeting the inclusion criteria of the study. The researcher asked the respondents to freely describe the reasons why they started antiretroviral treatment late. In addition, a semi-structured open ended questionnaire was also used by the researcher to prompt responses. Interviews went on until saturation point was reached. The audio-taped interviews were transcribed verbatim and the responses were analyzed. Responses from the different participants which bore similarity were assigned a similar color code. By using the cut and paste method, all responses bearing the same color code were pasted on one page resulting into the themes. Results: Four broad themes emerged as the reasons responsible for the late accessing of ART by the respondents. These themes were: 1. Stigma and discrimination, 2. Ignorance and lack of perceived risk of infection, 3. Denial, and 4. Health care system constraints. Conclusions: This small study though not exhaustive by any means, has highlighted some of the reasons why people living with HIV present late for help. The study was done using participants from only one peri-urban population. It is therefore not easy to generalize the results to the whole of South Africa, a country with a lot of social and economical diversities among its people. A bigger study over a wider geographical area might reveal different findings.

Research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, in partial fulfilment of the requirements for the degree MASTER OF MEDICINE (MMED.) (PAEDIATRICS AND CHILD HEALTH). Department of Paediatrics and Child Health Johannesburg, South Africa, 2012 / Cardiovascular disease is a common complication of advanced HIV disease in both paediatric and adult patient groups and may present with left ventricular (LV) dysfunction. Introduction of Highly Active Antiretroviral Therapy (HAART) has improved outcomes in patients presenting with LV dysfunction. However, mitochondrial toxicity, a complication of HAART, may present with myocardial dysfunction.

M.P.H., Faculty of Health Sciences, University of the Witwatersrand, 2011 / Background Few results are available concerning long-term clinical outcomes in ART treatment programs. The objective of this study was to describe clinical and laboratory outcomes for adult patients commenced on ART in Selebi Phikwe, Botswana from June 2004 to June 2005 within one year of commencement of ART. Methods Cross-sectional descriptive study of clinical and laboratory outcomes for 904 adult patients initiated on ART in Selebi Phikwe, Botswana, from June 2004 to June 2005. Data from ART services statistics was analyzed using descriptive statistical methods. Results Most patients had low a basal CD4 cellular count with a median count of 25 cells, which rose to 147 after 12 months of treatment. Of the 84 (9%) deaths, 75 (89%) had a basal CD4 count of less than 10 cells and 48 (57%) died within three months of commencing of ART Conclusion Good clinical and laboratory outcomes for patients on ART in resource limited are achievable. Mortality commonly occurs among patients with low CD4 counts and within three months of commencement of therapy.

M.Sc.(Med.), Epidemiology and Biostatistics, Faculty of Health Sciences, University of the Witwatersrand, 2011 / Background The therapeutic goal of HAART is sustained immune recovery and viral suppression. However some patients still have poor CD4 count responses despite achieving viral suppression. Such discordance immune response has been associated with poor clinical outcomes. We describe the prevalence of discordant immune response during the first 6-months of HAART and determine risk factors associated with this discordance at two large public sector clinics in South Africa. Methods We analysed data from 6 460 HIV-infected adults initiated onto first-line HAART at Goba and Phola Park clinics, in Johannesburg, South Africa between November 2008 and December 2009. Multivariable logistic regression models were used to estimate adjusted odds ratios (AOR) for associations between discordant immune response and clinical and demographic factors. Models were adjusted for WHO clinical stage, baseline CD4 count, education level and HAART regimen. Results At initiation of HAART, most patients were female 592(64.6%) and 803(87.6%) were initiated on 3TC-d4T-EFV/NVP. The mean CD4 count was 155 cells/mm3 (±118.4 sd), mean age was 38.5 years (±8.7 sd) and most patients had haemoglobin >11g/dL (n=645, 71.2%). By 6-months after initiation of HAART, 24% (n=220) of patients had a discordant immune response, 7% (n=67) discordant virologic 5 response and 21% (n=1359) had been lost to follow-up. In multivariable analysis, higher baseline CD4 cell count (CD4≥200cells/mm3): AOR=3.02; 95%CI 2.08-4.38; p<0.001) and moderate anemia (8-9.4 g/dL) at baseline (AOR=2.30; 95%CI 1.25-4.59; p=0.007) were the strongest predictive factors for development of discordant immune response. Conclusions We found a significant proportion of patients with discordant immune response 6-months after initiating HAART. Simple algorithms utilizing baseline characteristics can be developed for use in clinics in order to identify those patients at risk of development of discordant immune responses. Intensive monitoring of individuals at risk may improve clinical outcomes.

Background Potential causes of poor antiretroviral therapy (ART) treatment outcomes can be patient or health system related. Data on the effect of health system on ART outcomes is scarce. Objective: To compare treatment outcomes of HIV positive adults (≥18 years) initiating ART in either private or public HIV clinics in Johannesburg, South Africa. Methods: A retrospective cohort analysis was conducted on HIV positive, ART naïve adults initiating ART at a public (Themba Lethu Clinic) or private HIV clinic in Johannesburg between 01 January 2005 and 31 December 2011. Treatment outcomes included mortality, loss to follow up (LTFU; defined as >90 days since last scheduled visit date), failure to suppress viral load (>400copies/ml) at 6 and 12 months and absolute change in CD4 count from baseline until 6 and or 12 months after ART initiation. Survival analysis was performed using Kaplan Meir curves. Multivariate Cox proportional hazards models were used to assess predictors of mortality and LTFU. Generalized estimation equations were used to determine predictors of failure to suppress viral load while absolute change in CD4 count was analysed using the Wilcoxon rank sum test. Results: A total of 11690 patients initiated ART at the public clinic compared to 574 at the private clinic. Patients were similar in terms of age, gender and baseline viral load. Private clinic patients were less likely to die [aHR=0.39;95% CI 0.14-1.06] or to be retained on ART [aHR=1.59;95% CI 0.94-2.70], although both estimates lacked statistical significance. Public clinic patients presented with advanced HIV [WHO stage 3 or 4, p<0.001] compared to the private clinic. However, private clinic patients were 63% more likely to have a detectable viral load at 12 months of follow up [RR=1.63;95% CI 1.15-2.32]. There were no differences in the absolute CD4 changes between the private and public clinic at 6 months (median 99 IQR 43-78 vs. 103 IQR 52- 168; p=0.584) respectively. Conclusion: This study demonstrates that health systems have an influence on ART outcomes. The private sector is commended for early initiation of treatment and the availability of a variety of ARV drugs. However there is need for standardization of prescribing practices and care. Better virological responses amongst public patients can be attributed to better adherence to treatment and reduced LTFU rates compared to the private sector. Public-private partnerships are thus encouraged to address shortcomings of either sector.

M.P.H.,Faculty of Health Sciences, University of the Witwatersrand, 2011 / Introduction: Although the number of patients receiving ART is expanding nothing is really known about their adherence behaviours. Better outcomes with ART will not be achieved unless patients adhere to their treatments. It is also important to know and understand the factors that play a role in non-adherence so that the relevant interventions can be designed and implemented. Aim: The study aimed to determine adherence to ART in the clinic. By describing patient perceptions of their care, health care worker perception of challenges that they face in the clinic the study was aimed to determine the factors associated with non-adherence in the clinic as well. Materials and methods: A cross-sectional study of patients and health care workers at the Lerato clinic in Germiston Hospital. Adherence was accessed using pharmacy refill records and self-report. Both patients and health care workers were interviewed using questionnaires. The patient questionnaire had both closed and open-ended questions. Health care worker questionnaire accessed health care worker perceptions of the care that was given to patients. Triangulation produced a mixture of information that enabled not only cross- validation of the data, but minimised bias. Results: Ninety seven percent (97%) of the patients reported that they never missed their doses of medication. The adherence rate reported by pharmacy refill was 81% and the patients interviewed (n=67) had a mean age of 38.7 years and a median of ART use of 18 months. Both health care workers and patients were faced with challenges that impacted negatively on adherence. HCW’s faced workload, burnout, irregular debriefing, space constraints as well as irregular training while patients received the after effects of HCW challenges. These included among others waiting long in congested areas. Patients had financial constrains and hunger that were increased as a result of being on ART. Conclusion: Drug shortages, service availability and treatment costs did not affecting adherence to ART in the clinic. Strategies to maximise adherence in this situation should focus on meeting the needs of health care workers and patients. Special attention should be paid to addressing health care workers such as training, debriefing, the appointment system, defaulter system as well as fear of disclosure and discrimination.

Worldwide, an estimated 34 million people were living with HIV, but only 47% of the people in low- and middle-income countries eligible for antiretroviral therapy (ART) were receiving treatment at the end of 2010. The aim of this study was to investigate factors associated ART initiation as well as to determine ART uptake by age-group, gender and clinic catchment in a typical rural sub-Saharan African setting. Methods: Data from the Africa Centre 2010 population-based HIV surveillance (including C4 count measurement) was linked to the HIV treatment and care programme database. Those successfully initiating treatment (n=482) were then compared against those eligible for treatment but had not yet been initiated (n=117). The variables for analysis included gender, age, education level, employment status, number of individuals in the household, household asset index, distance of homestead from the nearest clinic, number of prior pregnancies and live births a woman ever had as well as the baseline CD4 count (at ART initiation and 2010 measurement for those not yet initiated). ART uptake by age-group, gender and across six clinic catchments was calculated using the population-based HIV surveillance from 2011 through linkage to the HIV treatment and care programme database. Results: Of the 1 308 HIV infected individuals who had CD4 count results, 599 were eligible for therapy based on a CD4 count criteria of <200 cells/mm3. Of these 80.5% (482/599) had initiated ART as of 31 October 2011. In the adjusted logistic regression model, males were 71% (OR = 0.29, p<0.001) less likely to have initiated therapy compared to females. Those in the 30 - 44 year age-group were 84% more likely to initiate therapy (OR =1.84, p=0.039) compared to those in the 15 - 29 year age-group. Individuals who had secondary and higher levels of education and those who lived far away (≥4 kilometres) from the nearest clinic were less likely to initiate ART (OR = 0.29, p = 0.001 and OR = 0.67, p = 0.337) compared to those with primary and lower levels of education and those who lived within 2 kilometres of the clinic respectively. Employed individuals were about twice as likely to initiate ART (OR 1.99, p = 0.017) compared to the unemployed. Overall the ART uptake across the study area among all HIV positive individuals was 32.5%. ART uptake and median CD4 count at initiation by clinic catchment ranged from 31.0% to 43.2% and 132 to 153 cells/mm3 respectively across the six clinic catchments. Conclusion: Although the overall rate of ART initiation was high, certain population groups were not covered well. Interventions that target younger people, males and unemployed individuals can help in reaching as many treatment eligible individuals as possible.

Thesis ((MSc(Med)(Pharmacy))--University of Limpopo (Medunsa Campus), 2011. / Background: Anecdotal evidence from fixed Voluntary Counselling and Testing (VCT) centres within the public sector indicates, that many patients are lost in the transition from VCT to Human Immunodeficiency Virus (HIV) care and treatment. The actual number of patients who are eligible for antiretroviral ttreatment (ART) after a positive HIV test, but who do not visit the antiretroviral (ARV) clinic to initiate ART, is currently not known. The need to identify the extent of this problem was therefore evident. Objectives: To investigate and describe the procedures and records used at the VCT centres under study. To identify the proportions of patients who fail to proceed through the different steps of the process from VCT to initiation on ART within a period of six months. To make recommendations for interventions aimed at improving the tracking of patients from the VCT entry point to ART initiation. Methods: The study was conducted as an operational research project at Odi and Stanza Bopape VCT centres. The design of the study was descriptive. Data were collected retrospectively and prospectively over a period of four months. Operational procedures and documentation systems at both VCT centres were observed. The records of all patients who tested HIV positive from 1 April 2009 to 30 June 2009 at Odi and Stanza Bopape VCT centres were identified from the VCT registers and selected for the study. Patients who were eligible for ART were identified based on their CD4 count. Eligible patient names were crossreferenced against the SOZO system (electronic patient database) to determine whether they had attended their pre-treatment visits at the ART clinic and whether ARV medicines had been dispensed to them for the first time. Where there was no proof that the patient attended the pre-treatment visits or finally accessed ART at an ARV clinic within six months, the patient was regarded as lost to initiation on ART. Results: The results obtained from the observational phase of the study showed differences in the procedures followed at the two VCT centres. At Odi VCT centre, patients referred for VCT by medical doctors only had an ELISA test and had to return on a different date for the ELISA test results, while patients visiting the VCT centre voluntarily first had a Rapid test and if positive they had an ELISA test on the same day. At Stanza Bopape VCT centre, patients referred by doctors and patients visiting the VCT centre voluntarily had a Rapid test and an ELISA test after a positive Rapid test. The patients at Odi had their CD4 test results interpreted by the nurse at the VCT centre while at Stanza Bopape the results were interpreted by the doctor at the ARV clinic. x iv The study included a cohort of 743 patients who tested HIV positive from April 2009 to June 2009 at Odi and Stanza Bopape VCT centres. Of these patients 344 tested at Odi VCT centre and 399 were tested at Stanza Bopape. The majority of patients at the two VCT centres were female (55% at Odi VCT centre and 59% at Stanza Bopape VCT centre), unemployed and single. At both VCT centres, patients were expected to return for collection of CD4 results within two weeks of the HIV test. At Odi VCT centre, 159 (49.4%; n=322) patients did not return to collect their CD4 results. Of those who returned, only 41.1% (67; n=163) returned within one month. At Stanza Bopape VCT centre 52.8% (210; n=399) patients did not collect their CD4 results. Of the patients who collected their CD4 count results, 51.3% (97; n=189) collected within one month. The Fisher’s exact test revealed no statistically significant difference (P=0.410) between the two VCT centres in terms of patients who returned for their CD4 results collection and those who did not return. More than half of the patients with accessible CD4 counts at Odi presented late for VCT. This was shown by 65.4% (n=275) of patients with CD4 count 200 cells/mm3 during HIV diagnosis. At Stanza Bopape VCT centre 46.6% (n=386) also had CD4 count 200 cells/mm3. The difference in terms of late presentation between the patients from the two clinics was statistically significant (P<0.001; Fisher’s exact test). The ART initiation rate at both VCT centres was found to be low. More than half of the patients eligible for treatment (CD4 200 cells/mm3) at both VCT centres did not initiate ART. This was shown by 59.4% (n=180) of patients at Odi VCT centre and 67.8% (n=180) of patients at Stanza Bopape VCT centre who did not initiate ART. There was no significant difference (P=0.317; Fisher’s exact test) between the two VCT centres in terms of the patients who did not initiate ART. Conclusion: A high percentage of patients who presented for VCT and were eligible for treatment were lost to initiation on ART. The majority of these patients did not return to collect their CD4 results and thus were lost immediately after VCT. These results suggest a need for an urgent intervention that will improve ART uptake. Recommendations: Patients referred by doctors for VCT at Odi VCT centre should have a Rapid test, and if positive they should have an ELISA and CD4 test on the same day to prevent the loss of patients before they even identify their HIV status. The option of a ‘one stop’ VCT and immediate CD4 results, should be further explored due to the unacceptable patient default rates at both VCT centres. A CD4 count machine which x v will provide results immediately on the same day of the test should be utilised. There must be sufficient personnel and equipment to follow-up on patients who do not return for their CD4 results, pre-treatment counselling and ART. The SOZO system should be integrated between the VCT centres and the ARV clinics to improve the flow of patient information between the VCT centre and the ARV clinic. A qualitative study should be conducted to explore reasons for patients not returning to collect their CD4 results. Key words: VCT; loss to initiation; non-uptake; lost in transition; HIV and AIDS

Thesis (MPH) -- University of Limpopo (Medunsa Campus), 2011. / Background Available and affordable second-line antiretroviral treatment regimens are a key component of universal access to treatment and its continuation. However, providing second-line ART is affected by a number of factors including infrastructure, skills and cod competency of available personnel, cost and availability of second-line drugs. Like first-line antiretroviral agents, second-line drugs have also inherent toxicities. While these have been described in adults, few studies reported this in children. Hence, the need to conduct this study. The aim of the study was to characterize the adverse effects on second-line antiretroviral therapy among HIV infected adults and adolescents and children treated at Mildmay Uganda (MUg). Method This was a cross-sectional study based on the review of patients' records. Data was extracted from client medical charts of patients on second-line antiretroviral treatment regimens treated at Mildmay Uganda Centre from January 2000 to December 2008. Results In total, 247 cases, the majority (90.7%) of the study participants were aged 13 years and above, female (53.0%), and single (68.0%). With regard to reasons that led to the second-line regimen, the main reason for switching was treatment failure (83%), followed by toxicity (5.7%), and other reasons such as the onset of Kaposi Sarcoma disease, maintaining regimen after transfer, and nevirapine pre-exposure. Overall, 55 out of 247 patients suffered from one or more adverse effects, a prevalence of 22.3%. The mean number of adverse drug reactions (ADR) was 1.3 per patient. Of the 55 who experienced adverse effects, 76.4% experienced one type of adverse effect. The three most common adverse effects were gastro-intestinal, followed by hematological and peripheral neuropathy. Based on age, while adults experienced a broad range of adverse 11 drug reactions including metabolic and hepatic ADRs besides the three cited above; children experienced two types only, gastrointestinal, and hematological ADRs. Hematological ADRs were significantly more prevalent in children than adults (66.7% versus 21.4%); they were also more prevalent in females than males (30.8% versus 15.2%). In contrast, while metabolic ADRs were equally distributed, peripheral neuropathy was more prevalent in males than females (30.3% versus 20.5%). Based on the number of ADRs experienced, more males than females (19.8% versus 14.5%) experienced one type of adverse effect whereas more females than males (6.1% versus 4.3%) experienced two to three adverse drug reactions to second-line regimens. Clinically, the didanosine-based regimens were the most associated with ADRs in both children and adults. In Children, the regimen, ABC/DDI/LPV/R, was responsible of 66.7% of ADRs documented; while in adults it was involved in 60.3% of ADRs. Similarly, TDF-, zidovudine-, and stavudine-based regimens were also involved respectively in 52.9%, 17.8%, and 14.7% of ADRs. Based on gender, didanosine-containing regimen, with either TDF or stavudine with lopinavir/r were responsible of 66.7% of ADRs reported in males; while TDF-based regimen with lamivudine or FTC were involved in 21.1% of ADRs also in males. In females, didanosine-, TDF-, zidovudine-, and stavudine-based regimens were also involved respectively in 57.9%, 36.9%, 23.7%, and 7.9% of ADRs. The regimen, TDF/DDI/LPV/R, was also the most associated with the occurrence of ADRs in females as it was in males. Moreover, although no difference was found among patients whose bodyweights increased significantly and those whose did not, patients whose CD4 counts increased significantly experienced more ADRs than those whose had not (8.8% versus 1.3%, p= 0.03). Conclusion Overall, second-line regimens seem to be well tolerated as the overall prevalence of adverse effects was 22.37%. The didanosine-containing regimens were most associated with the occurrence of ADRs. Hematological ADRs were more prevalent in children than 12 adults; and more in females than males. In contrast, while metabolic ADRs were equally distributed, peripheral neuropathy was more prevalent in males than females. These findings emphasize the need to individualize treatment based on the characteristics of the patient.