‘Prevalence’ and ‘incidence’ are terms frequently used as a measure of the extent of the HIV/AIDS epidemic. However not everyone is aware of the distinction between these terms and what their significance is. Different information can be obtained from these two measures of the HIV epidemic.

Prevalent cases of HIV are the people who are infected at a particular time, and therefore include people who have been infected for some time as well as those who are newly infected. In contrast, incident cases are the newly infected people only.

In the report on the demographic impact of HIV/AIDS in South Africa for 2006¹, the South African Medical Research Council, in collaboration with the Centre for Actuarial Research and the Actuarial Society of South Africa, defines these terms as follows:

In other words, prevalence gives the total number of all cases up till now (less those who have died), whereas incidence tells us how many recent infections have occurred. Prevalence thus gives us the ‘big picture’ of the situation as a whole i.e. the overall extent of the epidemic, whereas incidence is a reflection of speed at which the epidemic is spreading. Thus the incidence will be reflective of current behaviour and other factors leading to HIV infection. Furthermore, if we want to assess how well our preventive programmes are working, incidence will give us a better measure than prevalence.

Until recently, due to the extreme difficulty in measuring the incidence in a population, estimation of HIV incidence in South Africa has been carried out indirectly by mathematical models, such as that of Dorrington et al. (2006). But in the last few years, laboratory-based methods to distinguish between recent and established long-term infections, have been developed and provide a direct measure of the incidence. One of these, the BED capture enzyme immunoassay optimised for dried blood spot specimens, has been used in a study by the Human Sciences Research Council². In the study, about 16 000 people were tested in 2005.

Some findings regarding prevalence (The ‘big picture’)
The following are just some of the figures that emerged from the projections of the statistical model of Dorrington et al. (2006):

• Total number of people HIV positive in mid 2006: 5.4 million
• Percentage of total population: 11%
• Percentage of HIV positive people sick with AIDS: 11%
• Regarding youth:
  • Number of infected people under 25 years of age: 1.3 million (This is 24% of the total of 5.4 million infected).
  • Prevalence among young women (aged 15-24): 16.9%
  • Prevalence among young men in the same age group: 3.7% (The prevalence among women was thus nearly five times higher  than that among young men in this age bracket)
• The prevalence in the country and most provinces is reaching a plateau, with KwaZulu Natal the highest (estimated plateau of around 40% for women attending antenatal clinics), and the Western Cape the lowest (estimated antenatal plateau of around 17%).
  

The effect of Antiretroviral Therapy rollout on prevalence figures
As ART rollout proceeds, more and more people will be on medication, and fewer people will be dying of AIDS. This will mean that the total number of surviving people with HIV will be larger, compared with the time when people were not receiving ART. In other words, the prevalence will increase, because of two factors:

• relative increase in number of people with HIV and
• decrease in deaths due to AIDS.

If prevention measures work, the prevalence should eventually fall, but it will take many years to see this trend, as people with HIV will be living much longer than before.

In the short term, prevalence figures are not a good reflection of prevention measures because of the above effects of ART.

The only effective way to assess the success or otherwise, of prevention measures, is to measure incidence (new, recently-acquired infections).

Some findings regarding incidence
The study by Rehle et al. (2007) on the incidence (new cases) of HIV infection indicated that:

The total number of new infections per year was 571 000 - an average of 1 500 new infections per day in 2005.

• Of these, 192 000 or one third, were young people in the age group 15-24.

The highest incidence occurs in women in the 20-29 year age band at 5.6%, compared with an incidence rate in all persons 15 – 49 of 2.4%.

• Women are more vulnerable than men in the same age group:
  • the incidence in women aged 20-29 (5.6%) is more than 6 times the incidence found in males in this age group (0.9%); and
  • Among youth aged 15-24 years, females account for 90% of the recent HIV infections.
• The incidence in children aged 2-14, was 0.5%, with a total number of 69 000 new infections. These infections are probably not linked to mother-to-child transmission but to other factors, including child sexual abuse.
• Widows had a high incidence of HIV – 5.8%
• Pregnant women aged 15-49 had an incidence of 5.2%, compared with 3.8% for non-pregnant women.

Studies of incidence figures can give an indication of areas where the most research is needed: as can be seen above, groups needing urgent attention are young women, especially those who are pregnant. In addition, the high incidence among widows and young children needs attention.

In conclusion, there is no ‘right’ or ‘wrong’ regarding the use of prevalence versus incidence to assess the HIV epidemic, but correct interpretation will depend on people’s understanding of these two terms.

Sources

1. Dorrington R E, Johnson L F, Bradshaw D and Daniel T. The Demographic Impact of HIV/AIDS in South Africa. National and Provincial Indicators for 2006. Cape Town: Centre for Actuarial Research, South African Medical Research Council and Actuarial Society of South Africa.
2. Rehle T. et al. 2007. National HIV incidence measures – new insights into the South African epidemic. South African Medical Journal 97:194-199.

Author: Mary Mattheyse
Contact: afroaidsinfo@mrc.ac.za
Edited by Prof Debbie Bradshaw
Date: May 2007