AIM: This thesis examines the adherence to therapy as part of a clinical trial to determine the effectiveness of fully intermittent therapy for childhood tuberculosis. OBJECTIVES: These were to determine 1) the effectiveness of fully twice weekly therapy in childhood pulmonary tuberculosis, 2) whether adherence rates would be affected by twice weekly dosing and 3) whether certain socio-demographic factors influenced adherence.

Introduction Autism spectrum disorder (ASD) is an important global health problem. It has been shown that early diagnosis and intervention can improve the outcomes in affected children. Lack of knowledge about ASD among health care practitioners can delay the identification of children with ASD as well as early intervention. Currently, a dearth of epidemiological information exists regarding ASD in Africa. The aim of this study was to assess the knowledge of general practitioners (GPs) in Libya regarding ASD and to identify perceived challenges by GPs to early diagnosis and intervention of children with ASD in Libya. The study included the working GPs in the city of Tripoli. Methodology This cross-sectional descriptive study was conducted between 1st June 2017 to 31st August 2017 and utilized an electronic platform, the Research Electronic Data Capture (REDCap) from the University of Cape Town to collect study information. The Knowledge about Childhood Autism among Healthcare Workers (KCAHW) is validated questionnaire that explores health care worker knowledge about ASD across four domains. The KCAHW was used in combination with a more detailed questionnaire which explored challenges to early identification and interventions by GP’s in Libya. Descriptive statistics are reported in percentages and the Wilcoxon rank sum test was used to explore relationships between domain knowledge scores with age and gender as well as years of experience as a GP. Results The number of surveys returned to the REDCap application survey site was 215 which represent a response rate of 58.1%. However, the number of complete responses which could be included in data analysis was found to be only 62, indicating an overall response rate of 16.7%. The median age of the participants was 30 years old, IQR (29-33) years with a female predominance amongst participants (n= 55, 89%). The total median score from the knowledge questionnaire was 10, IQR (8- 12). No statistical differences could be found between domain knowledge scores and participants' age, gender and work practice. Participants showed good knowledge about stereotypical movements and repetitive behaviours in ASD, but 48% were not aware of comorbidities associated with ASD. The main challenges identified by GPs in the management of ASD were lack of awareness, insufficient experience, and health infrastructure in Libya as well as social stigma associated with developmental disorders. Conclusion and recommendations GPs in Libya have limited knowledge about certain aspects of ASD. A number of challenges were identified by GPs which hamper the early identification and management of ASD in Libya. There is a need to improve undergraduate medical training about ASD as well as to offer ongoing medical education to GP’s to improve recognition and management of ASD in Libya. Further research is indicated to explore the epidemiology, clinical spectrum and severity of ASD as well as intervention in Libya

OBJECTIVE: To assess the knowledge of Doctors and Clinical Nurse Practitioners (CNPs) with regard to the case management of ARI in children under the age of 5 years in the Khayelitsha District of Western Cape Province. METHODS: The study was a descriptive survey of all the doctors and CNPs in Khayelitsha district. A questionnaire with closed and open-ended questions was self-administered. The international WHO case management guidelines were used as the gold standard except where the Paediatric Handbook was used as the reference with respect to questions on tuberculosis, bronchiolitis and croup. A comparative analysis was carried out on the knowledge scores of the doctors in the public sector, the private General Practitioners (GPs) and the CNPs. Correlation between years of experience and scores were measured. A further comparison was performed between doctors with and those without a post-graduate qualification in Paediatrics. The association of the use of WHO/ARI/90.5 manual and the Red books (University of Cape Town Paediatric handbooks) with the knowledge scores was also evaluated. RESULTS: 34 Doctors and 34 CNPs participated in the study. All the questionnaires administered were returned - 100% response rate. The median percentage scores of the doctors and the CNPs were 68.5 % and 63.0% respectively, p = 0.007. Twenty six percent of CNPs scored less than 50%. The median scores of the doctors with and of those without a post-graduate qualification in Paediatrics were 78.5% and 67.0% respectively, p = 0.005. The median scores of the doctors in the public sector and the private GPs were 71.0% and 65.0% respectively, p = 0.03. The median scores of participants who had seen and made use of the WHO/ARI/90.5 manual and of those who had not were 68.5% and 66.5% respectively, p = 0.13. The median scores of participants who had seen and made use of the University of Cape Town Paediatric handbooks and of those who had not were 67.0% and 64.0%, respectively, p = 0.37. The rank correlation co-efficients of doctors' and CNPs' scores with years of experience were 0.26 (p = 0.13) and -0.01 (p = 0.94) respectively. CONCLUSIONS AND RECOMMENDATIONS: (i) There was a wide range in knowledge scores, with approximately one quater of CNPs scoring less than 50%. Training in management of ARI is therefore necessary. (ii) The poor correlation between years of experience and knowledge indicates that experience is not by itself associated with improvement in knowledge, and supports the need for a service - oriented training programme. (iii) The impact of the recommended training programmes should be evaluated using this study as a baseline for comparison. (iv) There is a need for similar Knowledge, Attitude and Practice (KAP) studies on ARI in other areas to inform local trainers and to enable prioritisation of training.

Background: The most prevalent, most lethal of the inherited dystrophies is Duchenne Muscular Dystrophy (DMD) and globally, the incidence is 1 in 3500 live male births. Currently, DMD has no cure, the latest care guidelines, especially on corticosteroids, cardiac interventions, and non-invasive ventilation, are all associated with improved muscle function, survival and quality of life. This reflects the fact that the natural history of DMD has been changed by these effective measures. Despite these advances, the progression and disastrous outcome of the disease cannot be modified. Potential therapeutic approaches that target the causative genetic mutations raise hopes of promising treatment for DMD. Many clinical trials of molecular genetic therapies have been planned and conducted for DMD. In South Africa, even though mutational characteristics of South African DMD/BMD patients have been described in several studies, the development of experimental therapies faces many challenges due to the lack of epidemiological data, the natural history of the disease and information about clinical care amongst Africans. Understanding the disease course of the local population can lead to better care approaches, further with the possibility of gene therapy becoming available, patients that would qualify for such treatment need to be identified. Hence the need for a DMD specific disease registry. Objective: This study aims to describe the concept and design of the first DMD disease registry of South Africa using Research Electronic Data Capture (REDCap) Methods: A comprehensive literature review was undertaken to identify the key areas of DMD, which must be recorded to permit comparison across disease expression and intervention variables. The registry was developed using REDCap's web based online designer accessed through the Clinical Research Centre (CRC) in the Faculty of Health Sciences at the University of Cape Town, and the workflow methodology was adopted to manage the registry. Clinical data from DMD patients form the database and consists of seven parts: 1) Enrolment details, 2) Background data, 3) Current disease, 4) Schooling, career prospects, and life style/psychological details, 5) basic activity of living scale, 6) power chart, 7) current motor function/symptoms. Electronic case report forms were created from these clinical data by the use of REDCap and for specific variables serial entries were possible relating to disease progression. We adopted international data standards proposed by TREAT-NMD, a global network of registries on DMD to ensure our data is internationalised and comparable to other registries. Results: Retrospective data entry combined with dynamic prospective recording of data was utilized in this project. Building on an existing basic database, 100 confirmed DMD boys are currently eligible for inclusion into the registry. The registry database consists of 7 forms collecting information on clinical and genetic information, which is subdivided into 100 items making a total of 210 variables. As our registry is an on-going study, sequential analysis of accumulated data will be done going forward to review trends on our DMD patients. Conclusions: This work describes the concept and design of our DMD registry and the steps followed to its establishment with REDCap. The focus is to consolidate clinical and genetic information on South African DMD patients that will translate to clinical research and form the basis for this patient information to be linked nationally and internationally. It is the hope that such an effort can be replicated in the conceptualisation of new disease registries.

Includes abstract. / Includes bibliographical references. / The purpose of this quantitative, descriptive, cross sectional survey was to determine whether maternal socio-economic status has an influence on infant feeding practices, nutrition and growth status of HIV-exposed infants at Delft Community Health Centre. The aim of this study was to describe the influence of maternal socioeconomic status on infant feeding practices and infant anthropometric measurements. Information was collected from 125 mother-infant pairs who presented at the health clinic with infants aged between six weeks to six months. The WHO anthropometry calculator was used to determine the z scores of the anthropometric measurements. One hundred and twenty five Case Report Forms of mother-infant pairs were analyzed. Few infants were underweight if their mothers’ personal income or total household income were more than R800.00 per month, 12.7% and 1% respectively. Nearly twice as many infants (49.6%) of the single mothers were underweight as compared to infants (19.8%) whose parents were married. Similarly, twice as many infants (50%) were underweight if their mothers walked to the health facility compared to 23.8% of infants’ whose mothers’ used taxis. Education and employment status of mothers appear to prevent infants from becoming underweight as twice as many infants (55.8%) were underweight when their mothers did not complete secondary school compared to 23.3% of infants whose mothers did complete secondary school. Nearly four-fold more infants (59.5%) were underweight if their mothers were unemployed compared to those infants (14.9%) whose mother were employed. Housing, the presence of a flush toilet or running tap water in the house did not improve the body mass index of infants. A total of 57.4% of infants whose mothers resided in brick houses, 71.9% of infants whose mothers had access to flush toilets and 57.5% who had running tap water in the house were still underweight. Infants whose mothers lived in houses with less than two rooms or where 3-4 people occupy the house had higher risk of being underweight (54.6% and 40.5% respectively). Underweight children were still prevalent even if the room were occupied by only one person (50%) of 1-2 children (67.2%). All women chose to formula feed their infants after receiving infant feeding counselling. Despite the availability of free replacement feeds there were evidence that infants were not properly fed.

Introduction: A continuous and concerning increase in the number of deliveries at Mowbray Maternity Hospital (MMH) has been noted over the years and now comprises a greater proportion of deliveries compared to deliveries conducted by midwives at midwife obstetric units (MOUs). To date there have been no studies assessing the changes in the pattern of deliveries at MMH. This study describes the antenatal and intrapartum referrals at MMH in 2005 and 2013, to identify any changes and whether or not referrals are appropriate. Method: This is a descriptive study with an analytic component involving review of a sample of hospital folders (138 for 2005 and 246 for 2013) of women who delivered at MMH from January to December 2005 and 2013. Results: The mean age of referred women was 27.259 (SD ± 6.277) years and 27.326 (SD ± 6.025) years in 2005 and 2013, respectively, with no significant statistical difference (p = 0.918). There was also no significant statistical difference (p=0.056) in the proportion of coloured, black or white women who delivered at MMH during 2005 and 2013. In 2005, a total of 27 (52.2 %) delivered women were single, 54 (39.1%) were married and two (1.4%) were divorced. In 2013, a total of 178 (72.4%) women were single, 65 (26.4%) were married and three (1.2%) were divorced, with a significant statistical difference (p < 0.001). In 2005, 75 (54.3%) women were unemployed and 46 (33.3%) were employed, whereas in 2013, 172 (69.9%) women were unemployed and 69 (28%) were employed, which shows a significant statistical difference (p < 0.001). In 2005, women mostly resided in Mitchell's Plain (32.6%), Gugulethu (28.3%) or in Khayelitsha (27.5%). In 2013, most women resided in Mitchell's plain (33.7%), Gugulethu (24.4%), Retreat 48 (19.5%) and Southern Peninsula 31 (12.6%), which represents a significant statistical difference (p= 0.001). 2 The median parity for 2005 sample was 1 (IQR: from 0 to 2), while in 2013 it was 1 (IQR: from 0 to 1). Although most women (94.2% versus 95.1%) booked at antenatal clinics in 2005 and 2013 respectively, with no significant statistical difference (p=0.697), the gestational age at first ANC differed significantly (p < 0.001) (median 24 versus 19 weeks). In 2005, the median number of ANC visits was five (IQR: from 4 to 7) visits, whereas the median was six (IQR: from 5 to 8) visits in 2013, with a significant statistical difference (p= 0.013). Over half of referred women (55.8% and 50.8%) in 2005 and 2013 respectively were delivered by normal vaginal delivery. The remainder had either a caesarean section or assisted delivery, with no significant statistical difference (p=0.139). Most women were referred from MOUs in both 2005 and 2013, at 90.6% and 85.45% respectively, with a significant statistical difference (p < 0.001). During both years virtually all pregnancies were considered high risk and the most common reason for referral was previous caesarean section (18.8% versus 19.9% respectively). For both years most pregnancy referrals experienced one, or more, antenatal risk factors, mainly previous caesarean section at 31 (12.5%), obesity at 27 (11%), prelabour rupture of membranes at 26 (10.6%) and HIV at 24 (9.8%) in 2005 and previous caesarean section accounted for 56 (11.1%), prolonged pregnancy for 51 (10.1), obesity for 50 (9.9%), HIV for 50 (9.9%) and tobacco use for 42 (8.3%) in 2013. The difference was statistically significant (p < 0.001). In 2005, the main intrapartum risk factors were fetal distress (23.6%), failure to progress and preterm labour (18% each). In 2013, fetal distress was most common (36.2%), followed by failure to progress (16.7%). The difference is statistically significant (p=0.034). Conclusion: The Cape Town Metro West health system features a functional maternity referral system. Midwives perform well in referring pregnant women who meet the criteria for high risk. There has been an increase in the number of women referred to the MMH over the study period but in this study group all referrals were found to be appropriate and were compliant with relevant obstetric management protocols. It appears there have not been large shifts in the demographics of referred women over the period reviewed. In addition, the change in the referral pathway has seen Southern Peninsula and Retreat referring to MMH but Khayelitsha no longer referring there. Furthermore, there are emerging risk factors that reflect the epidemiological changes currently being observed in the Cape Town Metro West region.

Includes bibliographical references (leaves 67-82). / This study deals with sexual behaviour of adolescents in a rural area of South Africa. The primary objective is to explore the sexual behaviour of rural adolescents, specifically with regard to the first time they engage in sexual intercourse, the number of sexual partners, their use of contraceptives and their knowledge of HIV/AIDS and how this influences their sexual activity. This study also examines how alcohol drinking, cigarette smoking, drug and substance use affect adolescent sexuality. Lastly, it explores the influence of religion and family connectedness on adolescent sexuality.

Aims: 1. To establish the extent of Syphilis in Pregnancy and the association of syphilis with unbooked status and perinatal deaths. 2. To determine why the diagnosis was not made in forty cases of Early Congenital Syphilis. Objectives: 1. To establish the percentage of patients with syphilis at delivery and possible association between unbooked status and positive syphilis serology. 2. To determine the Perinatal Mortality Rate and establish what percentage of perinatal death s occurred in unbooked patients and in those with positive syphilis serology. 3. To establish booking status, place of delivery, whether treated or not, serological findings, signs and symptoms of infants with Early Congenital Syphilis. Study Design: 1. A descriptive, retrospective study of deliveries in the month of March 1994. 2. A descriptive, prospective study of perinatal deaths over six-month period January to Jun e 1995. 3. A descriptive, retrospective study of Early Congenital Syphilis patients over five-year period 1990 - 1994. Setting: Maternity and paediatric wards at Provincial Hospital Uitenhage, East Cape. Patients and Methods: 1. Records were studied of 154 maternity patients delivering in March 1994 in order to ascertain booking status and serology results. 2. Data on sixty-one perinatal deaths weighing over 500 grams was examined to establish maternal booking and serological status. 3. Folders of forty patients with Early Congenital Syphilis admitted to the paediatric ward were examined. Results: 1. Syphilis in Pregnancy: i) Prevalence of syphilis at delivery was 9% for patients with titres =/> 1 :8. ii) Unbooked patients totalled 47% of deliveries (73/154). iii) Fifteen percent of deliveries had no syphilis serology tests performed and were discharged without screening (23/154). iv) No significant association was found between unbooked status and positive serology (p=0.35). 2. Perinatal Deaths: i) In instances of a perinatal death, a significant association was found between unbooked status and positive RPR serology (p=0.017). ii) Perinatal death rate due to syphilis totalled 10.8/1000 deliveries. 3. Early Congenital Syphilis: i) Eighty-five percent of cases (34/40) delivered at PHU and were mismanaged by personnel. ii) Eighteen of the thirty-four mothers had attended antenatal clinic and were untreated or inadequately treated by the time of delivery (53%). vii iii) Sixteen of the mothers had been unbooked and were discharged without screening or treatment (47%). iv) Four booked patients had negative sousveillance during the antenatal period. Conclusion: Provincial Hospital Uitenhage serves a poor community. This is reflected in the high prevalence of syphilis at delivery and the high proportion of unbooked patients. Time-consuming and inefficient methods of sero-surveillance plus separate clinics and staff for antenatal and Sexually Transmitted Diseases compounded existing problems. The standard of care anticipated at a Level 2 Referral Hospital was not delivered by Health Workers, largely due to lack of in-service training and guidelines for the management of Syphilis in Pregnancy. Recommendations: 1. In-service staff training by the Perinatal Education Programme (PEP) and adoption of protocols of management. 2. Rapid ("same-day") availability of serology results and initiation of treatment at Antenatal Clinic. 3. RPR at booking visit and repeated at delivery in all patients. 4. Monthly evaluation by Perinatal Problem Identification Programme (PPIP) and Obstetric/Paediatric meeting to monitor implementation.

Background: Malawi has one of the highest HIV/AIDS prevalence rates in sub-Sahara Africa. It has the ninth largest HIV burden in the world. Following the 2010 WHO PMTCT recommendations Malawi started providing lifelong ART to HIV-infected pregnant and lactating women regardless of clinical stage or CD4 count (option B+) in July 2011. Aim To assess the outcome of pregnant and lactating mothers receiving ART (option B+) and their infants less than 24 months in a rural health district of Malawi. Methods: A retrospective cohort study of option B+ women who were initiated on ART between 1st July 2011 and 31st December 2012 was conducted in Ntchisi district. Their exposed infants were also enrolled in the study. The study participants were followed up to 31st December 2013. Data was mainly collected from ART registers, ANC registers and ART patient master cards using structured questionnaires. Data analysis was done using Microsoft Excel and Statistical Package for Social Science (SPSS). Results: A total of 201 option B+ mothers, 136 pregnant women and 65 lactating mothers were enrolled in our study. Their median age was 32 years. 19.9% of HIV pregnant mothers started ANC at less than 12 weeks gestation and 21% attended the recommended four ANC visits or more. The proportion of pregnant and lactating women tested for HIV was 89.6%. Uptake of ART in HIV positive pregnant and lactating women was 80.1%. Of 54 option B+ mothers enrolled in the July 2011-December 2011 cohort, 70.4%, 64.8%, 57.4% and 55.6 % were retained at 3, 6, 12 and 24 months respectively, and 73.5%, 66% and 65.3% of 147 option B+ mother enrolled in the January 2012-December 2012 cohort were retained at 3, 6 and 12 months respectively. Out of 126 option B+ who remained in care in December 2013, 89 (70.6%) had adherence rate of 95% or more in the last visit of the October-December 2013 quarter. Of all women who commenced option B+ during pregnancy, 56/ 77 (72.7%) who remained in care during the October-December 2013 quarter had adherence of at least 95%, while 33/49 (67.3%) of women who commenced option B+ during lactation and who remained in care during the October - Dec ember 2013 quarter had adherence of at least 95% or more. This difference was not statistically significant, OR = 1.2, 95% CI: 0.6-2.8. A total of 198 exposed infants were enrolled and their median birth weight was 3.2 kg. Uptake of PCR/rapid test for the infants was 73.7 %. 163/198 (82.3%) received NVP. Out of 53 exposed infants enrolled in Jul y 2011-Dec ember 2011 birth cohort, 81.1 %, 67.9 %, 5 1 % and 17 % were retained at 3, 6, 12 and 24 months respectively. In the January 2012-Dec ember 2012 cohort the proportion of exposed infant s retained were 89 %, 81.2 % and 47.6 % at 3, 6 and 12 months respectively. Of all infants tested for HIV infection during the study period, a higher proportion who were enrolled in the July-Dec ember 2011 birth cohort became HIV-infected compared to those enrolled in the January-December 2012 cohort, 7/ 3 4 ( 20. 6 %) versus 4/ 112 ( 3.6 %), OR = 7.0, 95% CI: 1.9 -25.7. A significantly higher proportion of HIV-exposed infants born to mothers who initiated ART during lactation acquired HIV infection than those born to mothers who initiated ART during pregnancy, 7/43 (16.3%) versus 4/103 (3.9%), OR = 4.8, 95% CI: 1.3 - 17.4. Conclusion: Our research findings suggest that the PMTCT programme in the Ntchisi district can be improved. Late booking during pregnancy, initiation of ART late during pregnancy or only during lactation, low retention in care for HIV pregnant and lactating mothers and their HIV-exposed infants, inadequate HIV testing of HIV-exposed infants and low ART adherence rate of HIV pregnant and lactating mothers should be addressed in order to optimize the administration and effectiveness of option B+.

Includes bibliographical references / Alcohol use and alcohol use disorders contribute a significant proportion of the burden of disease in low, middle, and high-income countries. As a result, fetal alcohol spectrum disorders (FASD) represent one of the most common preventable causes of intellectual disability globally. Understanding the core brain areas of susceptibility to prenatal alcohol as they manifest in early life is key to developing strategies for early focused identification and intervention. This thesis explored the relative impact of prenatal alcohol exposure on the brain in infants as measured by multimodal brain imaging and the relationship of these findings to early neurobehavioral and developmental status. The specific aims the thesis addressed included leveraging structural magnetic resonance imaging (MRI), diffusion tensor imaging (DTI), proton magnetic resonance spectroscopy (1H-MRS) and resting sate functional MRI (rs-fMRI) scans in approximately 100 infants (50 alcohol exposed and a matched number of control, unexposed babies) at 2-4 weeks of age, to assess group differences in early brain development. Correlations between multimodal neuroimaging measures and neonatal neurobehavioral assessments and associations between early structural imaging findings and later infant developmental, as measured by the Bayley III assessment at 6 months, were further explored in the same group of infants. These studies addressed the hypothesis that maternal alcohol use in pregnancy would result in quantitative MRI abnormalities demonstrable at 2-4weeks of age and that these changes would correlate with early indicators of neurobehavior and development. Chapter 1 presents the rationale and outline of the thesis. The burden of fetal alcohol spectrum disorders (FASD) is described in the context of different resource settings around the world with detailed reference to South Africa. Chapter 2 presents a published systematic literature review of published studies of MRI in children and adolescents with prenatal alcohol exposure. Chapter 3 provides an overall description of the methods and context for this study. Although the results chapters each include a focused methods section, the word restrictions of journal articles did not allow for adequate contextual detail for the project as a whole.