A research report submitted to the Faculty of Health Sciences, University of Witwatersrand, Johannesburg, in partial fulfilment of the requirements for the degree of Master of Science in Child Health-Neurodevelopment. January 2017 / Aim and objectives: The aim of this study was to describe the hearing screening profile of children between 0-6 years living with HIV/AIDS currently on HAART at a virology clinic within a tertiary hospital in Gauteng using an audiological screening protocol. The objectives were to describe the demographic profile of children on HAART undergoing hearing screening, to determine the relationship between CD4 percent and the duration on HAART, to document and describe the occurrence of possible outer ear abnormalities, to document and describe the occurrence of possible middle ear pathologies and to document and describe the occurrence of possible inner ear pathologies. Methodology: This was a cross-sectional, prospective descriptive study; using purposive criterion sampling. It was conducted at a tertiary provincial hospital in Gauteng. A questionnaire and a hearing screening protocol was used to obtain data. Consent was obtained from the parent/caregiver of all participants. Ethical approval was obtained from the hospital and the University of Witwatersrand Medical Ethical Committee prior to the study. Results: There was the presence of possible ear pathologies detected by the hearing screening. The possible outer ear abnormalities existed in 26% of ears, possible middle ear pathologies existed in 29% of ears and possible inner ear pathologies existed in 1% of ears. However as the frequency increased the number of refers obtained in DPOAE screening also increased, which could be indicative of early cochlear pathology (inner ear pathology) in the high frequencies. Conclusions: Audiological screening in infants and children living with HIV/AIDS may be essential, as there may be a wide range of possible hearing deficits. If undiagnosed or not identified and managed early these deficits may result in language and cognitive delays. / MT2017

Perceptions of HIV and AIDS are inextricably linked to stigma and discrimination and perceptions of HIV positive mothers are particularly complex. In order to obtain a perspective on social perceptions of HIV positive mothers this study interviewed eight HIV negative, workingclass mothers. By focusing on HIV negative mothers, who shared the social category of motherhood (thereby providing an insider perspective), HIV was fore-grounded and social perceptions of HIV was explored from an outsider perspective. A vignette was used to facilitate a semi-structured interview which explored issues around general perceptions of the HIV positive individual (with a particular focus on issues around stigma and blame); perceptions of HIV positive motherhood and perceptions of the impact of maternal HIV on the child. Interview data was analysed using a qualitative thematic analysis. Results highlighted that motherhood from the perspective of the mother was perceived differently to motherhood from the perspective of the child. From the perspective of the mother, respondents identified with the HIV positive mother as a black, working-class mother and viewed her as ‘normal’ and ‘ordinary.’ Identifying infected mothers as part of the group of black, working-class women, respondents drew on their similarity of being powerless in heterosexual relationships. Thus respondents viewed HIV positive mothers as blameless victims when they became infected as their perception was that promiscuity was not a part of motherhood. When motherhood was considered from the perspective of the child, respondents viewed the HIV positive mother as a soon-to-be absent mother who continued to infect her child. The infected mother was perceived to have a relatively short lifespan and as such was perceived to fail in her duty as mother when she was not available to ensure that her child developed normally. Respondents perceived that the absent and unavailable infected mother produced children who are socially, morally and developmentally disadvantaged when they are teased and ostracized by society, fail to grow and develop normally and eventually turn into criminals.

Student Number : 0002189H - MA(Clin) research report - School of Human and Community Development - Faculty of Humanities / This study explored the perceptions of loveLife counselors on how HIV and AIDS has impacted on adolescent sexuality. The study further examined whether loveLife counselors thought that HIV and AIDS had changed how adolescents expressed their sexuality. The counselors’ perceptions regarding adolescents’ use of condoms during penetrative sex, being faithful to one sexual partner or abstaining from sexual relationships were also explored. In-depth interviews were conducted with loveLife counselors to understand how they think HIV and AIDS has affected adolescents’ sexual behaviour. The sampling method utilized was the nonprobability, purposive sampling. The participants consisted of five loveLife counselors, who work with adolescents. Data was analyzed by means of thematic content analysis. The loveLife counselors’ believe that adolescents have changed their attitudes about how they express sexuality as a result of HIV and AIDS. According to the counselors, HIV and AIDS infection among adolescents have decreased. HIV and AIDS have a positive impact on how adolescents sexually conduct themselves. The counselors think that adolescents are taking serious the warnings against the dangers of HIV and AIDS. This is due to the fact that they are afraid of contracting the virus and consequently suffering from the stigma that people have attached to those who are infected with the virus. As a result of that fear, the perceptions of the counselors are that the majority of adolescents are using condoms during penetrative sex, while others are being faithful to only one sexual partner. According to the counselors in most cases female adolescents initiate condom use as a way of preventing pregnancy rather than as a protective method against contracting HIV and AIDS. Adolescents are, however, finding it very difficult to abstain from sexual relationships.

MMed (Paediatrics), Faculty of Health Sciences, University of the Witwatersrand / Aim: Case fatality rates for child severe malnutrition have remained high globally and in South Africa. It has been postulated that much of this excess mortality is due to HIV infection. This study sought to examine case fatality rates in children with and without HIV infection, and with different forms of malnutrition. Methods: A prospective, observational study was undertaken at three academic hospitals in Johannesburg, South Africa. Severely malnourished children were identified and their anthropometric details, clinical features, laboratory findings and admission outcomes analysed. Nutritional status was categorised using the Wellcome and WHO classifications. All children had their HIV status established. Results: The case fatality rate in 113 severely malnourished children was 11.5%. Fifty one percent of children were HIV infected. Most (44%) of children had kwashiorkor, with 26% having marasmus and 20% classified as marasmic kwashiorkor. HIV positive children were significantly more likely to die than negative children (19% vs 3.6%, OR 6.2, 95% CI 1.2–59, p=0.02). Marasmic children were more likely to have HIV than those with kwashiorkor or marasmic kwashiorkor (83% vs 33%, OR 9.7, 95% CI 3.5–29.1, p< 0.001). Half (51%) of all HIV negative children whose mother’s status was known had an HIV positive mother. TB was suspected and treated in 24% of children, although confirmed in only 19% of these. Factors associated with an increased mortality included hypothermia (OR 9.7), hypoglycaemia (OR 9.7), shock (OR 7.2), thrombocytopaenia (OR 5.7), raised INR (OR 9.8) and the intravenous administration of fresh frozen plasma or packed red blood cells (OR 9.7 and 7.8 respectively). Conclusion: The HIV pandemic has altered the face of malnutrition in the study setting. Case fatality rates remain unacceptably high in HIV positive malnourished children. Specific guidelines for the management of severe malnutrition in HIV positive children and improved tuberculosis, growth monitoring and growth promotion programmes could reduce this impact.

HIV-associated sensory neuropathy (HIV-SN) is one of the most common neurological problems of HIV. It is frequently painful and reduces quality of life. HIV-SN can be caused both by HIV itself and by exposure to neurotoxic antiretrovirals such as stavudine. The South African Department of Health now recommends use of tenofovir in place of stavudine as first line treatment. However many people remain on stavudine and or live with the side effects. Stavudine is still prescribed in many other resource-poor countries. This thesis presents the first systematic study of clinical and genetic risk factors for the development of symptomatic HIV-SN in Black Southern Africans. I recruited 404 Black HIV-positive Africans from the Virology Clinic of the Charlotte Maxeke Academic Hospital, Johannesburg and assessed HIV-SN using the AIDS Clinical Trials Group (ACTG) Brief Peripheral Neuropathy Screen. HIV-SN was defined as present if the patient had both symptoms and signs of peripheral neuropathy. If present, the distribution and intensity of symptoms were recorded. Of those exposed to stavudine, 57% (226/395) had HIV-SN. Pain was the most common symptom and was experienced by 74% (172/226). Of these, 76% (128/172) reported their pain as moderate to severe. As in previous studies, increasing age and height were independently associated with risk of HIV-SN. However nadir and current CD4 T-cell counts and sex were not associated with SN. Patients donated blood for DNA extraction and single nucleotide polymorphisms (SNPs) were selected from the literature and genotyped using Illumina Golden GateTM technology. 342 individuals were assessed for genetic associations with HIV-SN and a subset of 159 positive for HIV-SN were assessed for associations with painful HIV-SN. I completed four genetic analyses: SNPs and haplotypes from TNF and adjacent genes from the major histocompatability complex on chromosome six were assessed for association with HIV-SN. I found no association with TNF-1031, even though this had associated with risk of HIV-SN in Caucasian, Chinese and Malay cohorts. Novel associations were identified between HIV-SN protection and 5 other SNPs (BAT1 rs3130059, rs2523504; ATP6V1G2 rs2071594; NFKBIL1 rs2071592, rs2071591). Associations were also found with haplotypes: FV15-23 weakly associated with risk and FV30-31 associated with protection against HIV-SN in this cohort. Analysis of 8 SNPs not previously assessed produced two novel associations with LTA SNPs (rs1041981, rs909253), where the minor alleles conferred protection against HIV-SN. Analysis of linkage disequilibrium (LD) suggests that there is linkage disequilibrium within the TNF block, that it differs between ethnicities and that TNF-1031 is unlikely to be a causative SNP for risk of HIV-SN. SNPs from other cytokines and chemokines implicated in the pathogenesis of HIVSN and the associated pain were assessed in Chapter 5. The major allele of the antiinflammatory gene IL4 (rs2243250) associated with risk of HIV-SN. This allele has been associated with higher CD4 T-cell counts, so I have proposed a role for high IL- 4 in early stage HIV-SN. A 3-SNP haplotype of IL10 associated with protection against HIV-SN whilst another IL10 haplotype showed a trend for risk of painful HIVSN. These data and the involvement of TNF haplotype (Chapter 4) suggest an inflammatory etiology for HIV-SN. Polymorphisms of UCP2 (rs659366) and UCP3 (rs1800849) have previously associated with risk of diabetic neuropathy. These SNPs encode uncoupling proteins 2 and 3 which regulate reactive oxygen species and may affect development of neuropathy via the effects of oxidative stress and mitochondrial dysfunction. Alleles of these SNPs did not associate with HIV-SN in this cohort. Patterns of linkage disequilibrium may differ between the two ethnicities or UCP2 and UCP3 may associate with a mechanism particular to diabetic neuropathy. I also assessed a ‘pain protective haplotype’ and SNPs of GCH1 which have been associated with decreased pain intensity in radicular pain following lumbar discectomy. Associations of the 3-SNP ‘pain protective’ haplotype (rs10483639*C, rs3783641*A and rs8007267*T) and a 6-SNP haplotype containing this motif with protection against pain were significant but dependent on age, sex and CD4 T-cell count. Association of another 3-SNP haplotype (rs10483639*G, rs3783641*T and rs8007267*C) with increased risk of pain in HIV-SN was also not independent of age, sex and CD4 T-cell count. The weaker associations here compared to Caucasian cohorts may be a result of differing LD between ethnicities or demonstrate different pain mechanisms between HIV-SN and radicular pain following lumbar discectomy. My results highlight the prevalence of HIV-SN and frequency of pain in this Southern African cohort. The genetic studies identify a likely inflammatory component and identify genes worthy of further investigation both in HIV-SN and the associated pain.

A research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in partial fulfillment of the requirements for the degree of Master of Public Health. Johannesburg, June 2017 / Introduction South Africa is grappling with the challenge of providing lifelong treatment, care and support to young people living with HIV (YPLHIV). The 2012 HIV Prevalence, Incidence and Behaviour Survey placed HIV prevalence among young people aged 15-24 at 7.1% (Shisana et al. 2014). Concerns abound around young people, and centre around low testing rates in this age group (Shisana 2014), their propensity for risky behaviours such as unprotected sex, age disparate relationships (Shisana et al. 2014) and high loss to follow up of ART-eligible youth (Cornell et al. 2010, Nglazi et al. 2012; Wang 2011). In the Western Cape province where HIV prevalence is 4.4% (Shisana et al. 2014) Clinic X provides integrated adolescent and youth-friendly health services including HIV counseling and testing, sexual and reproductive health, ART and other services for adolescents and youth aged 12-25 years. Services include facilitator-led support group meetings and private WhatsApp and Facebook groups for HIV-positive youth (van Cutsem et al. 2014). The aim of the study was to explore how social support is experienced by HIV positive young people from Clinic X who participate in support groups at the clinic, on WhatsApp and Facebook and to understand how support group membership is perceived to impact their HIV-related behaviours, if at all, during 2015 and 2016. The specific objectives were to (1) identify dimensions of social support experienced by YPLHIV, aged 18-25, in clinicbased and online support groups at Clinic X; (2) explore perceived social cohesion among YPLHIV aged 18-25 who participate in online support groups compared to those who only participate in clinic-based support groups and (3) explore perceptions of HIV positive young people at Clinic X in Khayelitsha regarding the influence of support groups on their behaviour during 2015 and 2016. Methods A cross-sectional exploratory qualitative study was conducted, namely focus group discussions (FGDs) and in-depth interviews (IDIs). Ethical clearance was obtained from the Human Research Ethics Committee (Medical) of the University of the Witwatersrand in October 2015. The researcher collected data for the study during December 2015 and November 2016 at Clinic X, which is one of two youth clinics in the Khayelitsha Health District of the Western Cape Metro Region. Participants were purposively sampled for the FGD component of the study from the population of young people aged 18-25 years old who attended support group meetings at Clinic X. The researcher purposively sampled interviewees for the IDIs from participants identified during the FGDs. Male and female individuals aged 18 to 25 years old who had been diagnosed HIV-positive, were enrolled in pre-ART or ART care at Clinic X Youth clinic and were participating in a youth club were included. Two FGDs were held, one for users of the online platforms and another for participants in face-to-face youth clubs meetings only. Twenty participants in total were included in the FGDs. Another nine in depth interviews were held with participants purposively sampled from the FGDs. Thematic analysis was used to analyze all data, which were recorded electronically, professionally transcribed and coded using MaxQDA. Key themes from data analysis included types of social support experienced, sense of belonging, desire to remain in the support groups and susceptibility to peer influence. Inductive themes included HIV-related behaviours, barriers to use of online platforms and sources of social support. Results Exploring the dimensions of social support revealed that peers within the youth clubs experienced emotional and informational support during youth club meetings as well as online. Emotional support was underpinned by the youth’s positive affect for each other based on shared experiences and the resultant feelings of being understood by their peers. The youth also received emotional and instrumental support from various other sources: health workers, family, and intimate partners as well as friends. The youth opened up about barriers to their seeking or accepting support from family or online as well as their perceptions about the consequences of the attitudes of individuals in their lives and community members about HIV. Participants’ perceptions about various indicators of social cohesion within the group showed that, while there was a sense of belonging, there were also limits in the extent to which youth felt susceptible to influence from peers in the youth clubs. Youth expressed a desire to remain in face-to-face youth clubs but were ambivalent about continued participation in the virtual clubs. Findings on the third objective revealed that the youth felt that the youth clubs had improved their knowledge and they were confident to maintain health-protecting behaviours. Seeing other youth looking healthy and overcoming challenges also motivated the youth. The behavioural intentions of YPLHIV were underpinned by gendered motivations for maintaining health, with young women being especially motivated by the desire for motherhood. Discussion To the body of evidence, this study contributes a qualitative understanding of how young people living with HIV experience social support online, showing that the dimensions of social support experienced are the same as those of participants of online health forums for other chronic conditions, namely information and emotional support (Gaysyknsky et al. 2014). Further, the study shows that for YPLHIV, online support groups may act as a barrier to the exchange of instrumental support, which research (Atukunda 2017) identifies as a salient need in this context. In seeking to understand why there may be disjuncture between the intention of pre-ART YPLHIV in the study to remain in care and the actual retention reported in numerous studies, the chapter proposes that social comparison, structural determinants and low risk perception may play a role. The major limitations of the study are the short time of engagement and exclusion of adolescents below 18 years of age who access services at the youth clinic. Conclusion and recommendations In conclusion, online peer groups foster peer support among YPLHIV, with implications on their emotional health. Still, more needs to be done to change perceptions about the use of online platforms for social support, design community-focused programmes to address gaps in support and reduce the financial burden on young people using mobile phones for health. / MT2017

Research report to be submitted to the Faculty of Health Sciences, University of Witwatersrand, Johannesburg, in partial fulfillment of the requirements for the degree of Masters of Medicine in the branch of Obstetrics and Gynaecology Johannesburg, 2017 / Introduction Cervical carcinoma is the second most common malignancy worldwide after carcinoma of the breast and most common in the developing world1,2. In Africa, the population of women who are 15 years and older is estimated to be 267.9 million with approximately 78 897 diagnosed with invasive cervical carcinoma annually and a 78% mortality1. Aim The aim of the study was to ascertain whether HIV sero-positive women in South Africa present with a more advanced disease of invasive cervical carcinoma than their HIV sero-negative counter parts as well as to assess the degree of immuno-suppression and its effect on the stage of the disease at initial presentation. Is there a difference between the studies done then and what is presented now? Methods This was a descriptive retrospective record review. A total of 1300 cases of cervical carcinoma were seen at Charlotte Maxeke Johannesburg Academic Hospital, Combined Oncology Clinic from 2009 to 2010. Variables analyzed were patient age, HIV status, ARV standing, CD4 count, parity, race, papsmear result, cell type FIGO staging. This was done using the SPSS (Statistical Package for Social Sciences) version 13. RESULTS The mean age of the patients analyzed was 50.74 +- 13.08. There were 436 (37.1%) HIV seropositive patients, with a mean CD4 count of 357.59 +- 361.15. The mean age of presentation for HIV sero-positive patients was 55.4 +- 11.8 and for sero-negative patients 42.1+- 9.5 (p=0.000). A majority of the patients presented with stage IIIB disease. The HIV status had no bearing on the stage of the disease at presentation (p=0.363), nor the degree of immunosuppression (p=0.999). Due to the HIV pandemic, sero-positive patients presented with invasive cervical carcinoma 10 years earlier than their sero-negative counterparts. Black patients were mostly affected when compared to other races with a (p= 0.004). Antiretroviral seemed to make no difference on clinical staging at presentation (p=0.152) Conclusion HIV sero-positive patients presented with invasive cervical carcinoma 10 years earlier than their sero-negative counterparts. The degree of immunosuppression and HIV sero-positivity has no bearing on severity of the disease. / MT2017

Background Gert Sibande District has the highest HIV prevalence among women attending public antenatal health clinics. Multiple sexual partnerships (MSP) enhance heterosexual HIV transmission, which is the main form of transmission in South Africa. There is need, therefore, to identify important factors associated with MSP for the development of strategic intervention policies and programmes. Objectives To determine the prevalence and associated factors of multiple sexual partnerships (MSP) in men and women in Gert Sibande District (GSD) in 2010. Methods This is a secondary data analysis of data collected through a cross-sectional multi-stage study using cluster probability sampling technique in GSD in 2010. The study collected data on 750 adult men and women aged 16 to 55 years through an interviewer-administered standardised questionnaire. Simple descriptive statistics and chi square analysis were used to determine the prevalence and patterns of the MSP in the study population. Multiple logistic regression models were built to determine factors that were independently associated with MSP. Results The analysis included 592 sexually active respondents: 200 men and 392 women. A fifth of the respondents had had their first sexual encounter before the age of 16 years. Condom use was higher among men than women. Condom use was lowest with most recent partners (56.6%) than in second (74.6%) and third sexual partners (78.6%). Alcohol use was high, with more men (72.0%) than women (33.2%) having ever consumed alcohol (p<0.001) and among these alcohol users, more men (44.5%) than women (8.7%) were involved in risky drinking. The overall reported MSP prevalence was 22.0%, 95% CI: 19 - 25%. Men (44.0%, 95% CI: 37 - 51%) reported significantly higher levels of MSP than women (10.7%, 95% CI: 8 - 12%). Levels of MSP decreased with age and were highest among young adults, men (20 to 24 years) and women (15 to 19 years), those who were never married, and among men in the intermediate socio-economic group. There were significant associations between MSP and underlying socio-demographic factors (age, socio-economic status and marital status), and with intermediate sexual behavioural factors (age at first sex, condom use at last sex), sex under the influence of alcohol and transactional sex in the past 12 months). Among men, young age (AOR 3.0, 95% CI: 1.0 - 9.3) socio-economic status (AOR 3.1, 95% CI: 1.7 - 5.6) predicted having MSP. The strongest positive correlation of MSP among men occurred with the sexual behavioural factors, particularly age at first sex (AOR 9.7, 95% CI: 2.3 - 41.4) and having sex under the influence of alcohol (AOR 4.5, 95% CI: 1.9 - 9.7). There was a 4.5 times likelihood of MSP with transactional sex in the past 12 months. Among women, being never married (AOR 10.9, 95% CI: 1.3 – 90.3), condom use at last sex (AOR 2.4, 95% CI: 1.1 – 5.6), transactional sex in the past 12 months (AOR 12.0, 95% CI: 3.9 – 37.1) and having sex under the influence of alcohol (AOR 9.3, 95% CI: 4.4 – 19.6) were significantly associated with increased odds of MSP. Conclusion The findings of this study showed a high prevalence of MSP compared to the reported prevalence of MSP in the South African national and sub-national surveys (SABSSM, NCS and DHS). The prevalence was ever higher across some sub-groups of the population. The findings highlight the need for interventions that will address socio-economic factors influence MSP in GSD, especially among young adults and unmarried adults of GSD. Among this largely black population, the occurrence of several sexual risk factors, including early age at first sex, transactional sex, and high alcohol use, indicate the need for group-specific interventions. This study also provides a basis for future research to allow for the comparison of changes in MSP levels among adults of GSD and for prevention interventions targeting partner reduction.

Abstract Introduction HIV serodiscordant couples are presenting more often to their doctors as the epidemic continues to spread. There have been various theories as to why some couples remain discordant, but none have yielded a conclusive answer as yet. The negative partners in discordant couples have previously been ignored. However, more of them are now presenting with psychosocial problems of their own. They are increasingly being recognized as ‘hidden’ patients and as potential resources. The researcher would like to discover what issues and problems these negative partners have so that we can understand and help them. Aims To explore seronegative partners’ experiences and emotions in HIV discordant couples. Methodology Using qualitative interview methods to explore the range of emotions and experiences that seronegative partners in discordant relationships experience at the time of disclosure of the different results and thereafter. Results Three main themes emerged from the research. These were: 1) Emotional problems-These were further sub-divided into expected responses and unexpected responses, based on the researchers own views and from literature reviews.2) Coping strategies- these were further sub-divided into positive and negative coping strategies based on the beneficial or harmful effects on the individual, family or the relationship as well as the increase in risk behaviour associated with each action. 3) Future plans-these were according to the participants and included what they wanted for their future. Conclusion Negative partners in discordant couples have been ignored for too long. They face many challenges and problems but they can also assist in the management of their partners. This was explored in more detail. From the research, a host or experiences and reactions were noted and based on these, recommendations for doctors were drawn up. The recommendations include the following: 1. Doctors should not ignore the negative partner in discordant couples. They are also patients and in need of help and support. They can in turn help the doctor in future treatment of the positive partner. By just acknowledging them and that they are also going through difficult times, allows them to bring out their problems and assist in future management. 2. Always discuss the possibility of discordant results before testing a couple. This possibility can be a source of immense stress and confusion to the couple if they have not been alerted to the fact that it can occur. 3. Both partners in the discordant couple must be screened for depression as there is a high rate of this illness in both groups. Too often, this depression is not actively looked for in the negative partner. Doctor’s who fail to do this, will be neglecting the wellbeing of the family unit. 4. All discordant couples should be offered ongoing counseling sessions, either separately or together as a couple. This will give the often ignored negative partner an opportunity to air their views and problems, not only to the doctor, but if they participate in the sessions as a couple, they may be able to speak to their partners, indirectly, by speaking to the doctor. 5. When faced with a discordant couple, the topic of future parenthood should be raised by the doctor. If this is ignored, it may force some couples to engage in high risk behaviour without them knowing the repercussions of it. The doctor must present all the options available to the couple, including adoption and assisted reproductive techniques as well as the role of antiretroviral medication. 6. Doctors must keep abreast of the latest developments in HIV and treatment as patients expect them to provide answers. This would include the doctor improving his/her computer skills as well as learning and practicing the art of critical reading so that he/she can access valid information for patients when the need arises. 7. Doctors must pay attention to the setup at their surgeries so that they always ensure privacy and confidentiality. This includes the position of the waiting room in relation to the consulting rooms so that what is discussed cannot be heard by others. The doctor must also not leave patient records where subsequent patients can see them. Lastly, the doctor must ensure that his staff members, who do see patient records, are informed of their obligations to patient confidentiality. 8. Doctors should avoid using the term undetectable when reporting on the positive partner’s viral load. This term may be misunderstood by both partners to mean that the virus was no longer present and thus safe sex practices were on longer needed. A better word to use would be “further reduced” or “ in the acceptable range showing good compliance”.

MMed (Paediatrics), Faculty of Health Sciences, University of the Witwatersrand, 2009 / Aim: To describe the prevalence of HIV infection, and the disease profile and outcome of 440 HIV-infected children admitted to the general paediatric wards at Chris Hani Baragwanath Hospital (CHBH). Methods: A comprehensive list of all paediatic patients admitted to the general wards between October and December 2007 was compiled using hospital admission records. Hospital folder and laboratory records were used to determine HIV prevalence. A retrospective review of inpatient hospital records was conducted for all confirmed HIV-infected paediatric patients admitted during the study period. Results: The prevalence of confirmed HIV infection amongst paediatric admissions at CHBH during the study period was 29.5% (95% CI 27.2 -31.9%). Of these children, 54.1% were newly diagnosed with HIV during the current hospital admission. Despite the majority (92.7%) of admissions having advanced HIV disease (WHO Stage 3 or 4), only 17% were accessing ART. Of the 202/440 (45.9%) children known to be HIV-infected before hospital admission, only 74/202 (36.6%) were currently receiving ART. Of the remaining 128/202 children known to be HIV-infected before hospital admission, 121/128 (94.5%) had WHO HIV stage 3 or 4 disease and thus were eligible for ART. Only 19% of children had a normal weight. Amongst infants aged less than 6 months uptake of PMTCT interventions was poor - only 36% of mother-infant pairs received single dose nevirapine and 28% of infants received cotrimoxazole prophylaxis. Respiratory illness was the principal reason for hospitalization in 37.5% of admissions. Gastroenteritis, sepsis and tuberculosis accounted for 22%, 19.5% and 21% of principal diagnoses respectively. The overall case fatality rate was 12% (95% CI 9.2–15.5%), with deaths in HIV-infected children contributing 58% of all deaths in the general paediatric wards. Over half (52%) of all deaths in the HIV-infected group occurred in infants younger than 6 months of age. vi Conclusion: HIV infection remains a major contributor to morbidity and mortality among paediatric admissions at CHBH. Poor uptake of PMTCT interventions, late diagnosis of HIV infection and delay in accessing ART are immediate barriers to improved care in HIV-infected children at CHBH. The underlying reasons for poor accessibility and under- utilisation of paediatric HIV-related services requires further investigation. Efforts to reduce mortality amongst HIV-infected paediatric admissions at CHBH should focus on early diagnosis of HIV infection and prompt initiation of antiretroviral treatment, especially in infants under 6 months of age.