Background: Tuberculosis (TB) can be prevented through vaccination with the Bacillus Calmette-Guerin (BCG) vaccine, but it still poses a major global health problem. South Africa, with an estimated annual TB incidence of 301,000 cases in 2018, remains one of the most TB burden countries. Current estimates show that the eight most burden TB countries—India (leading), China, Indonesia, the Philippines, Pakistan, Nigeria, Bangladesh, and South Africa—account for two thirds of the total TB burden. The 2018 figures for South Africa translate into an increase of 400% over the preceding 15 years. Furthermore, South Africa still falls below its targets of a case detection rate of at least 70% and successful treatment of 85% among new smear positive cases initiated on treatment. Monitoring and measuring progress in achieving these targets requires robust and high quality TB surveillance systems. In this way, policies and intervention programmes regarding the TB burden in South Africa will be derived using accurate TB data. It was in this context that the current research was set to contribute by: • Exploring the extent of the completeness of TB surveillance systems in South Africa; • Describing and modelling spatial epidemiology of TB; • Investigating the impact of Antiretroviral Treatment (ART) on TB notifications and mortality; and • Reviewing the quality of TB, the Human Immunodeficiency Virus (HIV), and ART policy guidelines in South Africa. Method: This research is a quantitative dominant mixed methods study which used secondary TB surveillance data which is routinely collected by the Department of Health, and TB mortality data (for the period 2005-2015) that is collated by Statistics South Africa. The TB data sets were the Electronic TB Register (ETR), n= 3,474,320 and the Civil registration system, n= 776, 176. In addition, 14 South African TB and HIV/ART policy guidelines for the period 2004-2016 were reviewed. An evaluation of the TB surveillance system was done using the updated Centres for Disease Control (CDC) guidelines on how to evaluate a surveillance system and then an epidemiological analysis of the two TB data sets was conducted. This was followed by the geospatial analysis of both data sets using the Global and Local Moran’s Indices to identify spatial autocorrelation and clustering respectively, the Local G to identify hot spots, and Spatial Lag Regression to identify spatial dependency. Further analysis was performed for the concordance between ART and TB. Finally, TB and HIV/ART policy guidelines were analysed using the Appraisal of Guidelines for Research and Evaluation (AGREE) II tool. Results: Evaluation of the TB system revealed that variables in the ETR data for date of birth, HIV status, and ART had low levels of completeness (5.1-62.1%, 20.6-90.6%, and 0.01-19.7% respectively) for the period under study. Furthermore, only two variables for the TB data from the civil registration system—smoking status and level of education of the deceased—had low levels of completeness. Other findings were a delay in initiating TB treatment in Western Cape province. Furthermore, the study revealed that the total number of TB deaths recorded in the ETR were not the same as those in the civil registration system; in fact, the ETR identified approximately 39% of the total TB deaths in the civil registration system. The epidemiological analysis of the mortality TB data in South Africa for the period 2005-2015 showed a significant decrease in the cumulative annual TB death rate. Furthermore, TB affected more males than females, with more deaths in the 35-44 age group for both males and females. The geospatial results revealed that some district municipalities had higher TB notification and death rates than others. The study further identified a positive association between TB notification and death rates, and the South African Multidimension Poverty Index (SAMPI) and HIV. Other findings were a spatial autocorrelation in the TB notification and death data for 2005 and 2010. Furthermore, this study demonstrated a presence of hot spots for the overall, male and female crude and age sex standardised TB notification and death rates in 2005 on the north western part and in 2010 in the eastern region of South Africa. Results of the concordance between ART and TB showed a decrease in the TB case notification and TB death rates that may be attributed to ART uptake. The AGREE II results showed that in all the reviewed TB and HIV/ART policy guidelines, the ‘Clarity of presentation’ domain had the highest scores; however, none of the reviewed policy guidelines could be recommended for use, therefore requiring improvement. Results for the extracted recommendations for data management, monitoring, and evaluation revealed that 50% (seven) of the policy guidelines did not have any recommendations that could be associated with data management, monitoring, and evaluation. Those that did, recommended accurate data recording, data collation, and analysis by the different levels in the health system (health facility, sub district, district municipality, province and national). Conclusion: This study has demonstrated that there are data quality issues for the South African TB surveillance system, particularly the completeness of some variables and the timeliness. Furthermore, the study identified TB hot spots in certain parts of the country. In addition, the study confirmed the utility of the Geographic Information System (GIS) to demonstrate any presence of areas with high risk of TB. It has further demonstrated that the quality of the reviewed TB and HIV/ART policy guidelines was poor. This information is useful for the South African TB programme. Furthermore, the findings indicated a disproportionate distribution of the TB burden which means that certain groups of people in South Africa (e.g., males and those in the 35-44 age group), including those living in locations with high levels of SAMPI and HIV, had a higher burden of TB during the period of study. The existence of these identified inequalities in the burden of TB underscores the importance of developing targeted public health interventions and policies to be directed towards the most vulnerable populations who happen to be those with observed high TB notification and death rates and are located in district municipalities with hot spots. As mentioned earlier, this research identified gaps in the TB surveillance data. Gaps in the surveillance TB data have major implications for TB surveillance; for example, if the TB mortality data have more deaths than the TB surveillance data, as was the case for this study, this means that there is a possibility that a number of people infected with TB were never identified by the surveillance system. Furthermore, it may imply that there is an under reporting of TB notifications which may have ramifications in terms of continued undetected TB infections in the community. TB can be prevented and cured, but it can be argued that the continued existence of poor TB data may, in part, contribute to the factors that have sustained TB to remain a challenge for South Africa. Addressing the identified challenges facing TB surveillance in South Africa will require engaging with the vulnerable groups right from the time when policies are being developed, as suggested in the AGREE 11 tool attribute of stakeholder involvement, followed by directing enough resources into TB prevention activities and surveillance data management at a district municipal level.