Master of Public Health - MPH / Human Immunodeficiency Virus (HIV) prevalence and new infections rate among young people in Southern Africa is high despite various programmes implemented to address general population prevention and treatment. The youth has a low HIV testing uptake even though there seems to be high HIV prevalence among this age population group. Youth focused interventions have proven to be a success in encouraging young people to have an HIV test done as part of improved health-seeking behaviours. In South Africa, such an intervention was initially implemented by LoveLife and later adopted in 2006 by the government as Adolescent and Youth Friendly Services (AYFS) for a larger-scale implementation in Primary Health Care facilities as a strategy to improve youth Sexual and Reproductive Health.

The main question addressed in this dissertation is: Are the legislation and policies of Ethiopia, Botswana and Uganda providing for various modalities of HIV testing consistent with human rights as enshrined under the African Human Rights system? Critically investigating the African human rights HIV testing instruments as well as the relevant domestic legislation and policies of the three countries. / Thesis (LLM (Human Rights and Democratisation in Africa)) -- University of Pretoria, 2007. / A Dissertation submitted to the Faculty of Law University of Pretoria, in partial fulfilment of the requirements for the degree Masters of Law (LLM in Human Rights and Democratisation in Africa). Prepared under the supervision of Professor Julia Sloth-Nielsen of the Faculty of Law, University of the Western Cape, Cape Town, South Africa. / http://www.chr.up.ac.za/ / Centre for Human Rights / LLM

Despite the availability of services, parents on antiretroviral treatment do not routinely get their children HIV tested and treated. The study aimed to document knowledge, attitudes and practices of parents towards HIV testing and treatment among children aged 0-14 years. A quantitative cross-sectional analytic study design was applied and a structured questionnaire used to collect data from 192 respondents. Findings show that 71.4% of respondents had correct knowledge of HIV transmission in children. Negative attitudes towards HIV testing of children were expressed by 42.2% of respondents while practices were reported by 16.7%. Practices were associated with self-efficacy, stigma, number of children and marital status. Knowledge and attitudes did not affect parental practices of HIV testing of children. Parents with low self-efficacy felt stigmatised, had negative attitudes, poor perception of availability of services and were unlikely to get children HIV tested. Recommendations to increase HIV testing among children included provision of skills and support to parents for HIV disclosure; localized care models to reduce stigma and identify children in need; and improving quality of services. / Health Studies / M.A. (Public Health)

Background: The HIV epidemic is a serious public health concern globally. There are 1,692,242 million known non-citizens in South Africa; this is equivalent to 3.3% of the total South African population (Statistics South Africa [STATS SA], 2011) this reflects global trends relating to number of non-citizens living in foreign countries (Vearey, 2008). Migration is an important demographic process to consider when studying HIV transmission as it increases migrants’ susceptibility to HIV (International Organisation for Migration [IOM], 2010). International migration, which is the movement of people across international borders, can result in migrants finding themselves in spaces of vulnerability which may lead to risky sexual behaviour (IOM, 2010). Furthermore, access to healthcare may be limited due to the dynamics of living in a foreign country. Internal migration, defined as the movement of people within the borders of a country (IOM, 2010) may result in ‘intra-urban’ inequalities that inhibit access to basic services such as housing and healthcare (Nunez et al, 2011). Despite it being the smallest province in the country, Gauteng has the highest level of in-migration, with an estimated net inflow of 367 100 internal migrants as for the period 2006–2011 (STATS SA, 2011). Globally, international migrants are more seriously considered as a concern for HIV transmission. However, in South Africa, internal migrants are equally as concerning particularly due to their circular migratory patterns. Knowledge of one’s status is a crucial first step in management of HIV. Voluntary HIV testing remains a challenging aspect of public health interventions, especially amongst key populations such as migrants (WHO, 2010). Although numerous studies have been conducted around migration and HIV, there remained a need for an investigation into the factors that influence HIV testing among Johannesburg residents. This is particularly significant, given the rapidly increasing levels of migration into the city, as well as the high urban HIV prevalence, which has been found to be twice high as that in rural areas and highest within urban informal settlements (Vearey, 2010). Therefore, this study set out to examine factors associated with HIV testing among residents of Johannesburg, in an attempt to determine whether migration status matters or not. Methods: This is a quantitative study with a sample size of 487 Johannesburg residents. International (n=150) and internal migrants (n=293) were examined in relation to each other and a comparative group of Johannesburg natives (n=44). STATA version 11 was utilised to conduct secondary data analysis of the RENEWAL survey (2008). This data, which was collected using a cross-sectional study design, was acquired from the African Centre for Migration and Society (ACMS) at the University of the Witwatersrand. Univariate descriptive analysis, bivariate chi-squared test and multivariate, logistic regression models were employed. Results: Levels of HIV testing were found to be higher amongst internal migrants (56%) when compared to international migrants (42%), (x2(1) =0.62; Pr=0.004). There was only a slight difference between Johannesburg natives and internal migrants who reported a 55% chance of HIV testing (x2(2) =8.32; Pr=0.016).These findings were only significant at the bivariate level. Overall, factors that were significantly associated with HIV testing amongst residents are: sex (95% CI 2.01 to 4.88; p=0.000); type of residence (95% CI 0.29 to 0.76; p=0.003); knows where to locate a testing facility (95% CI 1.41 to 3.50; p=0.001) and knows that anti-retroviral treatment (ART) is free (95% CI 1.93 to 4.83; p=0.000). Income was significantly associated with HIV testing amongst migrants (95% CI 0.40 to 0.90, p=0.016). Females were three times more likely (3.14) to test for HIV when compared to males. The odds of getting an HIV test by those who resided in informal settlements were less (0.48) when comparing with those who stay in formal housing. Residents who knew where to locate a HIV testing facility were twice (2.22) as likely to get tested for HIV as compared to those who did not know where to find one. Residents who knew that ART is free were three times as likely (3.05) to get tested for HIV as compared to those who did not. Those who were not earning a salary were less likely (0.61) to get tested for HIV compared to those who were earning a salary. Conclusion: The fact that migrant status, that is -internal versus international migration as a variable is not significant against HIV testing at the multivariate level indicates that there are far more important mediating factors that determine HIV testing than migration. More importantly, a more detailed and focused exploration into the length of stay of migrants in the city as well as the effect of urban inequalities on health, is needed.

Background and Objectives: South African businesses are feeling the brunt of HIV & AIDS and experiencing losses in productivity and profitability due to high levels employee absenteeism, sick leave and disability as well as the added costs of recruiting and retraining. These losses have had a negative impact on the national economy and in response some employers have agreed to recognize that HIV is a business issue and introduced HIV workplace programmes. It is not really known whether the employees that are most at risk are actually testing through these employer funded programmes in order for them as well as the employer to gain maximum benefit. There is a need to identify barriers to HIV testing so that workplace programmes can design better targeting strategies. Methods: This study was a retrospective review and analysis of the 2005 records of tested and untested employees in 8 companies registered on the Direct AIDS Intervention (DAI) Programme. Results: Overall there was very poor utilization of Voluntary Counselling and Testing (VCT) services with an average uptake of 13%. Multivariate logistic regression analyses showed that race, sex, period of employment and sector were significantly associated with HIV testing while there was no association with age and marital status. In general, Blacks were more likely to be tested for HIV (OR(95%CI)1.47(1.24 – 1.74);p<0.0001) and so were Coloureds (OR(95%CI) 1.79;(1.48 – 2.18); p<0.0001) and Indians(OR(95%CI)1.35(1.04 – 1.76);p=0.03) when compared to Whites. Males were less likely to have an HIV test (OR 0.69;p<0.0001) compared to female employees. Those who had been employed for more than one year were more likely to test (OR(95%CI); 1.83(1.37 – 2.43);p<0.0001) than newer employees. Employees who worked within a manufacturing company were more likely to have an HIV test (OR(95%CI) 2.39(1.96 – 2.92);p<0.0001) and so were those employed by a health/research companies (OR(95%CI) 2.83(2.11 – 3.81);p<0.0001) compared to those that were employed by a services sector company. Conclusions: The low uptake of VCT in this study is attributed to stigma which if not addressed will to continue to have a negative impact on the success of workplace programmes. Employers need to develop specific education activities in order to protect employees from discrimination and thus build confidence in the independence of the programmes thereby encouraging utilization.

Conhecer os motivos de busca de serviços para a realização do teste anti-HIV é umfator importante para a prevenção da aids entre a população geral. Apesar ddisponibilização de teste e aconselhamento gratuitos no Brasil, há lacunas quanto cobertura. Assim, este estudo estimou a associação entre fatores contextuais individuais e a realização do teste anti-HIV entre a população brasileira. Para isto, foram analisados dados de um inquérito domiciliar realizado com 4.760 moradorede regiões urbanas. A amostra final foi composta de 2.566 (51,9 por cento ) mulheres e 2.194(48,1 por cento ) homens que tinham iniciado a vida sexual. O modelo teórico foi baseado noquadro da vulnerabilidade e para efeito de análise as variáveis relativas às respostados indivíduos foram consideradas de nível individual (dimensão individual e sociada vulnerabilidade) e aquelas referentes ao município de moradia de nível contextual(dimensão social e programática da vulnerabilidade). As variáveis do nível individuaforam: características sociodemográficas e da saúde sexual e saúde reprodutivainformação sobre o tratamento de aids, conhecer alguém com aids e ter opiniõesobre práticas de segregação em relação às pessoas infectadas pelo HIV (questões que expressaram a atitude de apartação e exclusão do convívio social dos portadorede aids). Para o nível contextual, utilizamos o índice de desenvolvimento humano, prevalência de aids e a presença de Centro de Testagem e Aconselhamento no município de moradia. A variável dependente foi categorizada em não realizou teste, realizou por busca espontânea e realizou por solicitação. Para estimaçãdos fatores associados foram realizados três modelos de Poisson multinível deintercepto aleatório, sendo dois para mulheres (busca espontânea e por solicitação) 1 para homens (busca por solicitação); e um modelo de Poisson sem considerar osconglomerados para os homens (busca espontânea). Nos quatro modelos a categoriade referência da variável dependente foi não realizou o teste. No teste por buscaespontânea, observamos que os fatores associados que foram comuns entre mulheree homens foram do nível individual: idade, uso de preservativo na primeira ou núltima relação sexual, autopercepção de risco e conhecer alguém com aids. Asvariáveis associadas, a este tipo de teste, que foram diferentes entre os sexos foram:entre as mulheres no nível individual (ser solteira ou separada, início a vida sexual até 15 anos, ter tido três ou mais parceiros(as) sexuais na vida e ter informações sobre o tratamento para aids) e contextual (IDH alto e presença de CTA); entre os homens: somente do nível individual (ter ensino médio e superior e ser homo ou bissexual). Já na realização do teste por solicitação, as variáveis similares entre os sexos foram do nível individual (saber ler e escrever, ter filhos de até seis anos de idade, conhecer alguém com aids e ter informação sobre o tratamento de aids) e do contextual (presença de CTA no município de moradia). As variáveis associadas ao teste por solicitação, que foram diferentes entre os homens e as mulheres foram do nível individual: entre as mulheres (idade de 16 a 55 anos, ser casada ou em união consensual, início a vida sexual até 15 anos, uso de preservativo na primeira ou na última relação sexual e ter tido três ou mais parceiros (as) sexuais na vida) e do contextual (IDH alto no município de moradia). Entre os homens as variáveis do nível individual foram: idade entre 26 e 45 anos, ter tido DST na vida, sofrimento de violência sexual e não ter ideias de prática de segregação em relação à aids. Concluímos que o motivo de realização do teste ocorre mais frequentemente quando a epidemia é percebida como próxima. Concluímos também que, independentemente do sexo, a busca espontânea por teste se dá por fatores ligados à vulnerabilidade individual, ao passo que no teste solicitado agrega-se variável do plano programático. Concluímos que há marcadas diferenças de gênero, estando mulheres casadas e homens heterossexuais desprotegidos por não buscarem espontaneamente o teste anti-HIV. A realização do teste por solicitação é coerente com as estratégias da resposta brasileira à epidemia de HIV/Aids que prioriza a prevenção da transmissão vertical, a testagem quando da detecção de outra DST e na ocorrência de violência sexual / Understanding HIV test-seeking motivation is important to AIDS prevention in the general population. Although free counseling and testing are available, coverage gaps are found in Brazil. This study estimated contextual and individual factors associated with HIV testing among Brazilians. Data analysis was based on a household survey carried out with 4,760 residents of urban areas. The final sample corresponded to 2,566 (51.9 per cent ) women who had ever had sex and 2,194 (48.1 per cent ) men who had ever had sex. The vulnerability theoretical framework was developed to analyze variables regarded as being of individual level (individual and social dimensions of vulnerability), and those related to the city of residence were considered contextual level (social and programmatic dimensions of vulnerability). Individual level variables included socio-demographic characteristics, sexual and reproductive health, information on AIDS treatment, knowing someone with AIDS, and AIDS-related segregation ideas. Human development index (HDI), AIDS prevalence, and the presence of counseling and testing facilities in the city of residence were contextual level variables. Outcome was defined as not taking the test, client-initiated testing, and provider-initiated testing. Three multilevel Poisson models with random intercept were developed to estimate associated factors. Two models were estimated for women (client and provider-initiated testing) and one for men (provider-initiated testing), as well as one Poisson model excluding men clusters (client-initiated testing). Not taking the last HIV test was the reference category in all models. At the individual level, for both males and females, factors associated with client-initiated testing were: age between 26-35 years; inconsistent condom use; self-perception of risk; and knowing someone with AIDS. Different variables were associated with client-initiated testing for each sex. Among women the following variables were associated at the individual level: being single or divorced; first sexual debut until age of 15; having had three or more sexual partners; and information on AIDS treatment. The following were associated at the contextual level: high HDI; and presence of counseling and testing facilities. Among men only individual level variables were associated: high school and college education; and being homo or bisexual. As for provider-initiated testing there were some differences between the variables associated for each sex. For both sexes, some individual level variables were associated literacy, having children under 6 years old, knowing someone with AIDS, and information on AIDS treatment and one contextual variable presence of counseling and testing facility in the city of residence. As for gender differences in provider-initiated testing, the following were associated for women: at the individual level age between 16-55 years; being married or living with partner; first sexual experience under the age of 15; inconsistent condom use; and having had three or more sexual partners and at the contextual level high HDI of city of residence. As for men, the associated individual level variables were: age between 26-45 years; history of STD; being victim of sexual assault; and absence of AIDS-related segregation ideas. We conclude that HIV testing is more frequent among individuals who perceive AIDS epidemic as something familiar. We also conclude that client-initiated testing is associated with individual vulnerability and that provider-initiated testing is associated with programmatic vulnerability for both men and women. Test-seeking motivation is affected by gender inequality: married women and heterosexual men are unprotected as they present HIV testing-seeking behavior less frequently. Provider-initiated testing is in accordance with the Brazilian response to HIV/AIDS epidemic, which prioritizes HIV testing in strategies for preventing vertical transmission, and in the context of STD diagnosis and report of sexual violence

Conhecer os motivos de busca de serviços para a realização do teste anti-HIV é umfator importante para a prevenção da aids entre a população geral. Apesar ddisponibilização de teste e aconselhamento gratuitos no Brasil, há lacunas quanto cobertura. Assim, este estudo estimou a associação entre fatores contextuais individuais e a realização do teste anti-HIV entre a população brasileira. Para isto, foram analisados dados de um inquérito domiciliar realizado com 4.760 moradorede regiões urbanas. A amostra final foi composta de 2.566 (51,9 por cento ) mulheres e 2.194(48,1 por cento ) homens que tinham iniciado a vida sexual. O modelo teórico foi baseado noquadro da vulnerabilidade e para efeito de análise as variáveis relativas às respostados indivíduos foram consideradas de nível individual (dimensão individual e sociada vulnerabilidade) e aquelas referentes ao município de moradia de nível contextual(dimensão social e programática da vulnerabilidade). As variáveis do nível individuaforam: características sociodemográficas e da saúde sexual e saúde reprodutivainformação sobre o tratamento de aids, conhecer alguém com aids e ter opiniõesobre práticas de segregação em relação às pessoas infectadas pelo HIV (questões que expressaram a atitude de apartação e exclusão do convívio social dos portadorede aids). Para o nível contextual, utilizamos o índice de desenvolvimento humano, prevalência de aids e a presença de Centro de Testagem e Aconselhamento no município de moradia. A variável dependente foi categorizada em não realizou teste, realizou por busca espontânea e realizou por solicitação. Para estimaçãdos fatores associados foram realizados três modelos de Poisson multinível deintercepto aleatório, sendo dois para mulheres (busca espontânea e por solicitação) 1 para homens (busca por solicitação); e um modelo de Poisson sem considerar osconglomerados para os homens (busca espontânea). Nos quatro modelos a categoriade referência da variável dependente foi não realizou o teste. No teste por buscaespontânea, observamos que os fatores associados que foram comuns entre mulheree homens foram do nível individual: idade, uso de preservativo na primeira ou núltima relação sexual, autopercepção de risco e conhecer alguém com aids. Asvariáveis associadas, a este tipo de teste, que foram diferentes entre os sexos foram:entre as mulheres no nível individual (ser solteira ou separada, início a vida sexual até 15 anos, ter tido três ou mais parceiros(as) sexuais na vida e ter informações sobre o tratamento para aids) e contextual (IDH alto e presença de CTA); entre os homens: somente do nível individual (ter ensino médio e superior e ser homo ou bissexual). Já na realização do teste por solicitação, as variáveis similares entre os sexos foram do nível individual (saber ler e escrever, ter filhos de até seis anos de idade, conhecer alguém com aids e ter informação sobre o tratamento de aids) e do contextual (presença de CTA no município de moradia). As variáveis associadas ao teste por solicitação, que foram diferentes entre os homens e as mulheres foram do nível individual: entre as mulheres (idade de 16 a 55 anos, ser casada ou em união consensual, início a vida sexual até 15 anos, uso de preservativo na primeira ou na última relação sexual e ter tido três ou mais parceiros (as) sexuais na vida) e do contextual (IDH alto no município de moradia). Entre os homens as variáveis do nível individual foram: idade entre 26 e 45 anos, ter tido DST na vida, sofrimento de violência sexual e não ter ideias de prática de segregação em relação à aids. Concluímos que o motivo de realização do teste ocorre mais frequentemente quando a epidemia é percebida como próxima. Concluímos também que, independentemente do sexo, a busca espontânea por teste se dá por fatores ligados à vulnerabilidade individual, ao passo que no teste solicitado agrega-se variável do plano programático. Concluímos que há marcadas diferenças de gênero, estando mulheres casadas e homens heterossexuais desprotegidos por não buscarem espontaneamente o teste anti-HIV. A realização do teste por solicitação é coerente com as estratégias da resposta brasileira à epidemia de HIV/Aids que prioriza a prevenção da transmissão vertical, a testagem quando da detecção de outra DST e na ocorrência de violência sexual / Understanding HIV test-seeking motivation is important to AIDS prevention in the general population. Although free counseling and testing are available, coverage gaps are found in Brazil. This study estimated contextual and individual factors associated with HIV testing among Brazilians. Data analysis was based on a household survey carried out with 4,760 residents of urban areas. The final sample corresponded to 2,566 (51.9 per cent ) women who had ever had sex and 2,194 (48.1 per cent ) men who had ever had sex. The vulnerability theoretical framework was developed to analyze variables regarded as being of individual level (individual and social dimensions of vulnerability), and those related to the city of residence were considered contextual level (social and programmatic dimensions of vulnerability). Individual level variables included socio-demographic characteristics, sexual and reproductive health, information on AIDS treatment, knowing someone with AIDS, and AIDS-related segregation ideas. Human development index (HDI), AIDS prevalence, and the presence of counseling and testing facilities in the city of residence were contextual level variables. Outcome was defined as not taking the test, client-initiated testing, and provider-initiated testing. Three multilevel Poisson models with random intercept were developed to estimate associated factors. Two models were estimated for women (client and provider-initiated testing) and one for men (provider-initiated testing), as well as one Poisson model excluding men clusters (client-initiated testing). Not taking the last HIV test was the reference category in all models. At the individual level, for both males and females, factors associated with client-initiated testing were: age between 26-35 years; inconsistent condom use; self-perception of risk; and knowing someone with AIDS. Different variables were associated with client-initiated testing for each sex. Among women the following variables were associated at the individual level: being single or divorced; first sexual debut until age of 15; having had three or more sexual partners; and information on AIDS treatment. The following were associated at the contextual level: high HDI; and presence of counseling and testing facilities. Among men only individual level variables were associated: high school and college education; and being homo or bisexual. As for provider-initiated testing there were some differences between the variables associated for each sex. For both sexes, some individual level variables were associated literacy, having children under 6 years old, knowing someone with AIDS, and information on AIDS treatment and one contextual variable presence of counseling and testing facility in the city of residence. As for gender differences in provider-initiated testing, the following were associated for women: at the individual level age between 16-55 years; being married or living with partner; first sexual experience under the age of 15; inconsistent condom use; and having had three or more sexual partners and at the contextual level high HDI of city of residence. As for men, the associated individual level variables were: age between 26-45 years; history of STD; being victim of sexual assault; and absence of AIDS-related segregation ideas. We conclude that HIV testing is more frequent among individuals who perceive AIDS epidemic as something familiar. We also conclude that client-initiated testing is associated with individual vulnerability and that provider-initiated testing is associated with programmatic vulnerability for both men and women. Test-seeking motivation is affected by gender inequality: married women and heterosexual men are unprotected as they present HIV testing-seeking behavior less frequently. Provider-initiated testing is in accordance with the Brazilian response to HIV/AIDS epidemic, which prioritizes HIV testing in strategies for preventing vertical transmission, and in the context of STD diagnosis and report of sexual violence

Despite the availability of services, parents on antiretroviral treatment do not routinely get their children HIV tested and treated. The study aimed to document knowledge, attitudes and practices of parents towards HIV testing and treatment among children aged 0-14 years. A quantitative cross-sectional analytic study design was applied and a structured questionnaire used to collect data from 192 respondents. Findings show that 71.4% of respondents had correct knowledge of HIV transmission in children. Negative attitudes towards HIV testing of children were expressed by 42.2% of respondents while practices were reported by 16.7%. Practices were associated with self-efficacy, stigma, number of children and marital status. Knowledge and attitudes did not affect parental practices of HIV testing of children. Parents with low self-efficacy felt stigmatised, had negative attitudes, poor perception of availability of services and were unlikely to get children HIV tested. Recommendations to increase HIV testing among children included provision of skills and support to parents for HIV disclosure; localized care models to reduce stigma and identify children in need; and improving quality of services. / Health Studies / M.A. (Public Health)

Magister Public Health - MPH / Background: Early HIV testing is a crucial step for pregnant women in preventing mother-to-child transmission of HIV. In the public sector nearly all pregnant women presenting at antenatal clinics are screened for HIV. However, according to a large medical-aid administrator in South Africa, only 21.96% of pregnant women on their medical aid claimed for an HIV test as part of their antenatal care in 2012. Despite having frequent opportunities when consulting with pregnant women, general practitioners tend to be reluctant to offer HIV screening to these privately insured patients. In South Africa, private sector general practitioners are reimbursed for their services at pre-determined, negotiated rates. Previous studies indicate that monetary incentives over and above the negotiated rate may motivate health providers to promote screening to patients, and this may lead to increases in the uptake of testing. Due to limited resources within the public health sector, general practitioners are seen as key resources in a public private partnership to assist government achieve strategic health outcomes such as improved access to quality healthcare and improved compliance to treatment plans. Methodology: A quasi-experimental, ‘before and after’ study design, was conducted among 2,934 Metropolitan Health network general practitioners in South Africa who managed a pregnant woman on a medical aid. The same populations of general practitioners were used in the pre and post analysis with the general practitioners receiving information about the benefits of HIV testing in pregnant women before and after. The only difference was with the intervention related to a new HIV Counselling and Testing incentive process. Data was extracted from the billing system of a private medical insurance company in South Africa>. Quantitative data and stratification was analysed using the Statistical Package for the Social Science software, version 16.0 and Epi Info version 7.1.0.6. The effectiveness of the intervention was assessed by comparing the pre intervention period between April 2011 and September 2012, and post intervention period between March 2013 and August 2014. A subgroup analysis was done to determine variations in the name it, by general practitioners and patient characteristics. Results: There was no significant difference in HIV testing by general practitioners in this network preand post the intervention (21.99% vs. 21.96%, p=0.939). Compared to general practitioners aged 25-44 years, general practitioners older than 65 years old were 13% less likely to test (OR 0.87, CI: 0.74-1.01) and general practitioners between 45 and 65 years were 9% less likely to do an HIV test (OR 0.91, CI: 0.85-0.98). This study found that as patients’ age increased, they were more likely to be tested: beneficiaries aged 35- 44 years were 15% more likely to be tested compared to beneficiaries aged 15-24 years (OR 1.15, CI: 1.1-1.21). Beneficiaries who had a vaginal delivery were less likely to be tested compared to women who chose caesarean as a delivery method (OR 0.87, CI: 0.84-0.9). Medium income beneficiaries were more likely to be tested compared to low income beneficiaries (OR 1.09 CI: 1.03-1.16) and beneficiaries from the “high income” scheme grouping were less likely to be tested (OR 0.87, CI: 0.82-0.92) compared to the low income scheme grouping. The timing and frequency rates of HIV testing, for both caesarean and vaginal deliveries, occurred most between months two and six, peaking at month four. Overall, Eastern Cape and Mpumalanga had the lowest testing rates compared to all the other provinces (OR 0.96 CI: 0.89-1.05). Conclusions: Most general practitioners’ HIV testing rates of pregnant women in the private sector behaviour analysed in this study remained the same, despite the presence of a financial incentive. This study’s findings suggest that healthcare provider behaviour to comply with clinical guidelines and best practice, has no association with the presence of financial incentives, especially with increased administration tasks to access the incentive. These study findings emphasise the need to continue to strive for improved compliance especially by older general practitioners’ to adhere to clinical best practice and national HIV screening guidelines of pregnant women. The aspiration of achieving the highest quality of care in both private and public sector are principles that should continue to be pursued especially where private sector general practitioners’ will be used to offer public health services in the future National Health Insurance.

Master of Public Health - MPH / Increased uptake of VCT services by pregnant women may be attributed to the development of counseling services and increased availability of rapid tests at the study clinics by the Namibian Health and Social Services. A high knowledge and understanding of HIV and VCT services by pregnant women also probably contributed. / South Africa