What are the key numbers regarding the HIV and TB epidemics in South Africa? On this page we keep track of all the latest estimates from a variety of sources. This page is updated from time to time as new information becomes available. It was last updated in July 2018.
1. HIV prevalence: How many people in South Africa are living with HIV?
According to outputs from the Thembisa model there were around 7.2m people living with HIV in South Africa in 2017. This number is likely to keep rising since the rate of new HIV infections is higher than the rate at which people with HIV are dying. The Thembisa estimate is very close to the UNAIDS estimate of 7.1m people living with HIV in South Africa in 2016 (a number that would have risen by 2017). In his budget vote speech in May 2018 Minister of Health Dr Aaron Motsoaledi said that there are 7.1m people who are living with HIV in South Africa.
2. New infections: How many people are newly infected with HIV in South Africa per year?
According to the Thembisa model there were around 267,000 new HIV infections in South Africa in 2017. The model estimates that this number peaked in 1999 at around 570,000. The model anticipates that this number will continue to fall in the coming years. The latest UNAIDS figures indicate that there were around 270,000 new HIV infections in South Africa in 2016. Both UNAIDS and Thembisa figures translate to over 700 new HIV infections per day in South Africa.
3. HIV mortality: How many people die of Aids-related causes in South Africa per year?
According to the Thembisa model around 123 000 people in South Africa died of Aids-related causes in 2017. This is down from a peak of around 270,000 in 2005. The Thembisa estimate is slightly higher than the UNAIDS estimate of around 110 000 Aids-related deaths in South Africa in 2016. It should however be noted that these estimates have relatively wide confidence intervals and should thus only be considered to be very rough estimates. In March 2018 Statistics SA reported that HIV was indicated as the cause of death on the death notifications of 21,830 people in 2016. This number is an underestimate since death notifications do not allow for multiple causes of death to be indicated – in many cases where TB or pneumonia is written on death notifications HIV would be the underlying cause. In addition, in the Stats SA report over 57 000 deaths are categorised as “Ill-defined and unknown causes of mortality”. It seems likely that at least some of these 57 000 deaths would be because of HIV. The Statistics SA mortality and causes of death report is useful, but studies and models that use its information in combination with other sources of information are likely to produce more reliable estimates of the actual numbers of people who die of HIV and TB. While the District Health Barometer (2015-2016) focuses on years of life lost rather than deaths per year in its burden of disease chapter, it is clear on the role of HIV in deaths, stating: “In 2000, 2005, 2010 and 2015, the three leading single causes of YLLs (years of life lost) in South Africa were HIV-related conditions, TB and pneumonia, with diarrhoea ranking fourth in 2000 and sixth in 2015, suggesting that HIV-related mortality remains the leading cause of YLLs in the majority of districts in South Africa. They also address the death reports issue stating: “It is important to note that a large proportion of HIV deaths has been mis-attributed to immediate causes of death such as TB, diarrhoeal diseases and lower respiratory infections.”
4. How many people are receiving antiretroviral therapy in South Africa?
According to the Thembisa model around 4m people were receiving antiretroviral therapy in South Africa in 2017. UNAIDS estimates around 4m in 2016 – they will likely have a slightly higher estimate when their 2017 figures become available. In Motsoaledi’s 2018 budget vote speech he said that “over 4.2m” people in South Africa are on treatment.
Whichever source you believe, there are in the region of 3m people living with HIV in South Africa who are not receiving treatment. According to treatment guidelines all people living with HIV are eligible for treatment. Evidence from the START trial has definitively shown that it is better to start taking HIV treatment earlier rather than waiting for the immune system to weaken.
5. New TB cases: How many people contract or develop active TB in South Africa per year?
According to the 2017 WHO Global TB Report around 438,000 people fell ill with TB in South Africa in 2016. Of these, 182,000 were women and 256,000 were men. A new National Tuberculosis Prevalence Survey is expected to report its first findings in 2019. The study is being conducted by the South African Medical Research Council (SAMRC), the Human Sciences Research Council (HSRC) and the National Institute for Communicable Diseases (NICD). According to the District Health Barometer 2016-2017 only an estimated 72.8% of people in South Africa in 2016 with diagnosed TB were started on TB treatment.
6. New MDR-TB cases: How many people contract or develop MDR-TB in South Africa per year?
According to the 2017 WHO Global TB report, South Africa had around 19,000 cases of MDR- or rifampicin-resistant (RR) TB in 2016. Only around 11,000 of these people received treatment.
7. TB mortality: How many people die of TB in South Africa per year?
According to the 2017 WHO Global TB Report roughly 124,000 people died of TB in South Africa in 2016 – of these 101,000 were HIV positive and 23,000 were not. In March 2018 Statistics SA reported that TB was indicated as the cause of death on the death notifications of 29,513 people in 2016. As explained in the HIV mortality section above, the death notification numbers reported by Statistics SA is likely an underestimate.
What are the best sources of HIV and TB numbers?
There are a number of regular reports and models that provide information on the HIV and TB epidemics in South Africa. Below we provide a list of sources together with an indication of the kind of information they provide.
The Thembisa Model: The Thembisa model is a sophisticated and comprehensive mathematical model of HIV in South Africa developed by experts at the University of Cape Town. It combines and expands upon a number of earlier models, most notably the Actuarial Society of South Africa (ASSA) AIDS and Demographic model. The model is calibrated using a variety of real-world data sources and in our view provides the most reliable set of estimates relating to the HIV epidemic in South Africa. Detailed model outputs can be downloaded from the website in Excel format.
UNAIDS: UNAIDS publishes a series of regular reports and makes estimates available on their website. They use a different model from that used by Thembisa and often come up with different estimates.
HSRC household survey: Every few years the Human Sciences Research Council publishes its “National HIV Prevalence, Incidence and Behaviour Survey”. This report is based on HIV tests and interviews conducted by researchers visiting households across the country.
Antenatal survey: The National Antenatal HIV Prevalence Survey is published annually by the Department of Health. It is a survey of the HIV status of pregnant women attending public health antenatal clinics. Pregnant women are one of the groups of people most likely to visit public sector clinics and as a result this survey provides one of the most complete snapshots we have of HIV prevalence.
WHO Global Tuberculosis Report: Every year the WHO publishes the Global TB Report. The report contains estimates on a number of key TB indicators for South Africa and a range of other countries. Sometimes more up-to-date data is also published on the WHO’s TB site.
Stats SA: Statistics South Africa annually produces its “Mortality and Causes of Death” report. As mentioned above, this report reflects only what is written on death certificates – which, while very useful, should not be interpreted as the last word on causes of death in South Africa.
MRC: The South African Medical Research Council from time to time produces reports on various aspects of HIV and TB in South Africa.
Government: While the South African Department of Health does not often share detailed information with the general public, important statistics are from time to time provided in speeches, press statements and answers to questions posed in parliament. The health budget vote speeches of the Minister and Deputy Minister and the State of the Nation address delivered by the President of the country usually contain some top-line numbers.
SA Health Review: The SA Health Review: This annual publication of the Health Systems Trust does not track a set of indicators from year to year. It is nevertheless worth consulting since the chapters in this publication often contain rich information.
OHSC: The Office for Health Standards Compliance produces reports for parliament on the state of healthcare facilities. While these reports do not specifically deal with HIV and TB, they provide useful information on the healthcare system that is supposed to form the front-line against these co-epidemics.
Provincial and district numbers
Getting reliable statistics about healthcare in provinces and districts is difficult, but there are a few places one can look.
The District Health Barometer published by the Health Systems Trust is the best public source of information about the public healthcare system at district level. The 2016-2017 edition (published in January 2018 is available here.)
The Thembisa Model outputs (mentioned earlier) contains provincial estimates for a wide variety of HIV-related indicators.
There is provincial and district level data held by the Department of Health that is generally not released to the public.