Includes bibliographical references. / The recent scale-up of ART coverage in resource-limited settings has greatly improved access to treatment. However, literature indicates that patients on ART have high rates of non-adherence (32% to 67%), virological failure (5.2%, to 47%) and resistance to ARVs (5.4% to 80%). These patients are failing first- and second-line ART, leaving no treatment options for successful virus suppression. Yet, literature addressing facilitators and barriers faced by patients on secondline ART is scarce. This study examined factors affecting adherence to second-line ART from the perspective of clinic staff as well as second-line failure patients, assessing both individual and structural barriers. Research was conducted at a large primary-care clinic in Khayelitsha, a periurban township in Cape Town, South Africa. A literature review was conducted to present 1) rates of non-adherence, virological failure and resistance and 2) to present known facilitators & barriers faced by patients on ART, with a specific focus on second-line patients. The literature was found via PubMed and Cochrane Central Register of Controlled Trials (CENTRAL), with a preference for studies in low and middle income countries and those including second-line ART populations. The primary research used participants who were drawn from an MSF-run program to support patients failing second-line treatment.

Includes bibliographical references (leaves 149-176). / Diabetes is a growing problem worldwide. It is estimated that at least 171 million people have diabetes worldwide and this figure is projected to more than double by 2030. Mauritius has not been spared from diabetes. Indeed, it has one of the highest diabetes prevalence in the world. It is creating tremendous pressure not only on households but on the health care system.

Includes abstract. / Includes bibliographical references (leaves 55-60). / Bacterial vaginosis (BV) is a common cause of vaginal symptoms among women of childbearing age, although some women may be asymptomatic. The etiology of BV has been uncertain for many years, but several studies have related an increased frequency of BV to multiple diverse factors such as smoking, sexual behavior, vaginal douching, black race, sexually transmitted infections, and certain socio-economic factors. BV infection may lead to adverse reproductive health outcomes and may also increase the susceptibility of women to human immunodeficiency virus (HIV) and sexually transmitted infections (STIs). It is therefore of importance to know the risk factors for BV among a population so that mechanisms of prevention can be developed.

Objectives: This article will review qualitative data from intervention-based counselling sessions to explore barriers to partner notification for men and women who have contracted sexually transmitted infections (STIs) in a township community in South Africa. It will further assess the intervention's impact on participants' motivation and skills to notify their partners about their STI status. Methods: Relying on recorded counselling sessions from an intervention run by a parent study, this substudy reviewed 30 transcripts from enhanced counselling sessions with 15 men and 15 women. Participants were adults between the ages of 19-41 (mean age = 28.4) who live within the catchment area of a South African township. Recordings were chosen based on verbal responsiveness of the participant and were manually coded for analysis. In addition, two programme counsellors were interviewed to enhance rigour and to reduce potential bias. Results: By the conclusion of the intervention session, both male and female participants were motivated to notify their partners face-to-face about their positive STI status. Despite this, misperceptions about the etiology and transmission of STIs, as well as inadequate support from the clinical level and power imbalances amongst men and women emerged as major barriers for the prevention of future STIs. Conclusions: While the intervention appears to be successful in facilitating partners' intentions to notify, the data shows significant social and structural barriers that will create difficulties for the prevention of future STIs. Participants' persistent concerns about acquiring HIV or their current positive status affect decision-making and therefore, could be a window of opportunity for health-care providers or lay counsellors to discuss STIs in high prevalence areas.

Violence has been declared a public health problem by the World Health Organisation (1996). Violence prevention strategies are actively used in efforts to rehabilitate offenders during their incarceration, and are intended to prepare them for reintegration into society and to prevent recidivism. Given the lack of research undertaken on violence prevention strategies in the gang-ridden context of Cape Town, South Africa, this dissertation looks at a prison intervention programme, Help I am Free, that uses performance arts as a case study to investigate the potential impacts of such intervention strategies. The paper does so by outlining the experiences of five male gang members before, during, and after their time in prison. Using a cross-sectional qualitative research design, the dissertation explores participants’ life histories and seeks to shed light on the ways in (and degrees to) which such programmes may influence the men’s attitudes towards themselves and others, and the likelihood of their successful reintegration into society. The theoretical frameworks used draw on applied theatre theories, social-cognitive theories, and theories of masculinities. The paper argues that programmes such as Help I am Free may impact participants’ perceptions of self and others strongly, and may also lead to an increased sense of self-worth and motivation to change lifestyle. However, it also indicates the limitations of the programme to achieve sustained success, given the structural barriers which afflict South Africa. The paper outlines a number of challenges that the men face while on parole, and concludes that while prison intervention programmes, and art programmes specifically, may trigger a sense of healing and encourage more positive notions of self, there are more comprehensive strategies in post-release urgently needed in order to prevent recidivism and to achieve authentic social re-integration.

South Africa is one of six countries worldwide that has the highest national burden of tuberculosis (TB) and the largest number of HIV-infected people in the world. HIV infection, Mycobacterium tuberculosis (M.tb) infection and TB disease is most common during a woman’s reproductive age, particularly in South African women. HIV co-infection increases the risk of TB disease either by facilitating reactivation of a latent TB infection or by favouring the progression of a recently acquired TB infection towards active disease in HIV-infected patients. Globally, HIV-TB co-infected adults are 19 times more likely to develop TB disease than HIV-uninfected adults, in the absence of preventive therapy. In South Africa 61% of TB cases are reported to be HIV-infected. HIV-infected pregnant women with latent TB infection are more likely to progress to active TB disease and women in the early postpartum period are twice as likely to develop TB as non-pregnant women, usually at 3 months post-delivery. More pregnant women die from TB disease than from any other pregnancy or childbirth related causes, particularly in South Africa. This risk is greater in HIV-infected, pregnant women, who account for 29.7% of pregnant women attending public antenatal clinic services in South Africa. Infants of pregnant women with TB have increased risks of mortality and morbidity compared to infants of women without TB, and these risks are even higher in pregnant women co-infected with HIV and TB. The risk of M.tb exposure, infection and TB disease in HIV-exposed, uninfected infants is high. An analysis is presented on the relationships between sociodemographic and clinical risk factors and M.tb infection and TB disease in HIVinfected mothers and HIV-exposed infants examined in the setting of an infant TB vaccine clinical trial. Prevalence of maternal M.tb infection and the incidence rate of maternal TB disease and infant M.tb infection and TB disease in this cohort is also investigated. The protocol (Part A) outlines the study design and the methodology of the research for this sub-analysis. The literature review (Part B) provides an overview of recent and current literature on the prevalence and incidence rate of M.tb infection and TB disease in HIV-infected pregnant and post-partum women and their HIV-exposed infants in resource-limited settings, particularly in sub-Saharan Africa and specifically in South Africa. Literature on the risk factors associated with the exposure and progression to M.tb infection and TB disease in these susceptible populations is described. The results of the sub- analysis are presented as a manuscript (Part C). The main findings are the incidence rate of maternal TB was 1.36/100 person-years and incidence rate of infant M.tb infection and TB was 2.47 and 3.62/100 personyears respectively. Maternal CD4 count >350 cells/mm³ was strongly associated with QFT positivity that may have affected the estimate of maternal M.tb infection. Infant M.tb infection was driven by new household TB contact(s) as was infant TB disease in addition to higher QFT values (IU/ml) and maternal smoking. Determining which pregnant or postpartum HIV-infected women and their infants are at the highest risk of becoming M.tb infected and developing TB disease, by improving active TB screening of mother-infant pairs, could be an important public health means to reducing the burden of disease and death caused by TB, particularly in HIV endemic areas of South Africa where Prevention of Mother to Child Transmission coverage is greater than 95%.

Includes bibliographical references / Chemical hazard communication is intended to alert users of the potential hazards of chemicals. Apart from the fact that hazard information should be understood, it is also important that it should be recalled. Recall of hazard communication is critical when the written form of the information is not available at the time it is required. A cross-sectional study investigating predictors of the recall of chemical safety information on labels and safety data sheets amongst 315 workers (industry, transport and agriculture sectors) and 87 consumers in two provinces of South Africa was conducted. The recall of participants was tested using two modules (module 2 which relates to the demographics, and module 3 which relates to the recall of the participants) from the Hazard Communication Comprehensibility Testing (CT) Tool developed by the Centre for Environmental and Occupational Health at the University of Cape Town. Respondents were predominantly male (67.7%), the median age was 37 years (IQR: 30-46 years) and less than half of the participants completed high school (47.5%). The majority of participants were blue collar workers outside of industry (55.5%). The skull and crossbones symbol was the label element most recalled, both unprompted (79.6%) and prompted (94.8%), and the first aid and treatment measures the least frequently recalled (6.0%-29.9%). Multivariate analysis showed the predictors that were found to increase the recall of all the label elements were, industrial sector, transport sector, agriculture sector, gender, home language English and Afrikaans, reading English and Afrikaans, completing high school and non-industry white collar occupations. The predictors that were found to decrease recall were further education, not wearing glasses and non-industry blue collar occupations. This study found demographic factors to influence the recall of hazard information. Policy should ensure the implementation of procedures that promote the recall of hazard information to protect workers from hazardous exposures.

This project focused on the recent global reforms in TB diagnostic policy and the implementation of Xpert MTB/RIF (GeneXpert) diagnostic technology into the health system, as a case to assess the extent to which software issues - particularly the human qualities of the system – mediates policy implementation. It centres the experiences of frontline workers in local implementation contexts as imperative because of frontline workers’ have discretionary power and influence in their practice. The premise of this mini-dissertation is that researchers and policy makers should centre the lived experiences of service delivery level health workers when implementing policy or programmatic reforms. This may deepen people-centred approaches which is essential for health systems strengthening. This mini-dissertation is structured into three parts: Part A: This is the research protocol that was submitted for ethical review and approval to the Faculty of Health Science Ethical Review Committee (FHSERC). The protocol frames the study objectives and the initial intentions of the research study. The justifications for the research question, theoretical framework, the research design, methods for data collection and analysis and timelines are clearly presented and discussed. Part B: Using GeneXpert policy reform implementation as a pathfinder, this section presents an undertaking of a structured narrative review of the existing literature addressing the major barriers and enablers for health systems implementation reform. This review assesses the extent to which people issues and people-centred practices are considered in policy implementation research of GeneXpert. The aim of this section of the dissertation is to identify and map-out literature considering the human experiences and relationships of frontline health workers and how these may intersect with hardware, contextual and social systemic factors, that may potentially mediate the implementation of GeneXpert TB diagnostic policy. Part C: This section presents the background, methodology, findings and interpretations from the research, as a journal-ready manuscript. This paper seeks to contribute to the policy implementation literature in the field of HPSR from the perspective of centering nurses' lived experience – especially nurses who are overburdened and undervalued – as imperative in the field of inquiry. The main findings reflect that nurses are burdened by the pressure to meet policy targets, the encumbrance to enforce administrative and bureaucratic procedure, and the minimal platforms or pathways to input on challenges and innovations back to higher level management and decision makers. Within the context of top-down, target-driven and highly structured and standardized operational processes for diagnosing TB, nurses navigate multiple overlapping and contradictory modes of being in their interactions with patients as a response to these pressures. This paper seeks to offer voice to nurses’ experiences of implementing TB diagnostic policy in PHC settings in SA considering its relationship with broader systemic and contextual influences. It also raises particular issues about tensions between efforts to achieve efficiency and effectiveness through enforcing the system, and facilitating people-centered and responsive practices in implementation.

BACKGROUND: Since gaining independence in 1971, the total fertility rate in Bangladesh has decreased from 6.9 births per woman to 2.3. However, Bangladesh ranks second in the world for greatest number of pregnancies by age 18. Modern contraceptive use among married adolescent girls is low at 47%, whereas the rate of unintended pregnancy is high at 21%. The United Nations and the Government of Bangladesh have indicated that provision of adolescent-friendly reproductive health services is a critical next step in family planning programming. METHODS: Guided by the Social-Ecological Model, I applied mixed methods to determine factors that influence modern contraceptive use among married adolescent girls in Bangladesh. For the quantitative component, I used a logistic regression approach and the 2014 Bangladesh Demographic and Health Survey (BDHS) data to develop models for modern contraceptive use among nulliparous married adolescent girls (n=641). For the qualitative component, I conducted nine key informant interviews with family planning experts at the Ministry of Health and non-governmental organizations, and facilitated focus group discussions with 60 health workers. Data were collected in Khulna division where total fertility is low and in Sylhet division where total fertility is high. FINDINGS: Contrary to expectations, contraceptive use was not associated with access to family planning services or with female autonomy. Continuing education after marriage, however, was associated with contraceptive use. These associations were robust to the inclusion of common correlates of contraceptive use, including: girl’s awareness of emergency contraception, older age at first sex, younger age, girl’s preference for more children, termination of pregnancy, husband’s desire for fewer children than the girl, marriage to a husband 30 years of age or greater, and wealth. Health workers’ personal biases against adolescents’ contraceptive use, and the belief that hormonal contraceptives cause infertility are barriers to service provision and contraceptive use among adolescents. RECOMMENDATIONS: To reduce early childbearing in Bangladesh, health workers require training to provide family planning services to adolescents. Increasing community knowledge of modern contraception is equally important. Efforts are required to enroll and retain married adolescent girls in secondary education.

CONTEXT: Health is influenced by a broad range of social, economic, and environmental factors beyond the typical remit of professional public health. It is therefore increasingly recognized that multiple sectors need to be engaged in efforts to improve population health. Health in All Policies (HiAP) is an approach to systematically consider health across policies and programs. HiAP has not been implemented in Delaware, despite interest from stakeholders. Using Delaware as a model state where HiAP could be implemented comprehensively, this research sought to answer two questions: 1) “What HiAP adoption and implementation models are appropriate for the state of Delaware?” and 2) “Using the Intervention Mapping framework, how can such models be adapted to the state?” The study provides a method to operationalize HiAP – which can be used by researchers and practitioners globally – and gives Delaware a concrete plan to move HiAP forward. METHODS: The study used a cross-sectional design to achieve its aims. Methods included document review, key informant interviews, focus groups, and a questionnaire. Thematic analysis was used to analyze qualitative data and descriptive statistics were generated to examine questionnaire data. Intervention Mapping, a framework for program design, implementation and evaluation, provided the study’s guiding structure. FINDINGS: Three key findings emerged regarding HiAP broadly: 1) HiAP practitioners do not adequately use strategic communications to increase buy-in across sectors; 2) the scope and reach of HiAP is influenced by the degree of institutional power held by the lead organization and cross-sector partners; and 3) practitioners do not fully recognize the importance of being adaptable throughout HiAP implementation, which hinders sustainability. The study also synthesized best practices and identified Delaware’s relevant contextual factors. The research demonstrated the value Intervention Mapping can bring to operationalizing HiAP. CONCLUSIONS: Scholars need to refine the essential elements of HiAP to add: 1) strategic communications across sectors and 2) flexibility throughout HiAP implementation. Practitioners and researchers seeking to advance HiAP in a jurisdiction should use contextual factors and Intervention Mapping to systematically create a theory-based, practical approach with measurable objectives. Future research should further examine the roles of strategic communications and institutional power in HiAP.