1. South African National HIV Prevalence, Incidence, Behaviour and Communication Survey, 2008
Impact of HIV/AIDS

2. Outline of the presentation
Objectives of the study
Methods
Ethics
Results
Conclusion
Recommendations

3. Objectives of the report
To present data for mid-term review of SA NSP
To describe trends in HIV prevalence, HIV incidence, and risk behaviour in SA
To assess exposure to major national HIV communication programmes
To propose indicators to be used to monitor the South African HIV & AIDS epidemic and its management

4. Method Study Design: Cross-sectional national population Survey
Population: All in 2008; 2002 & 2005:≥2 years
Sampling: a multi-stage disproportionate, stratified sampling approach, 1000/ EA
Mapped the EAs
We stratified the sample by province, type of settlement and age.
We disproportionately sampled populations with smaller populations such as whites and Indians.
We provide a map to show the geographic distribution of the sampled area

5. 2007 Master Sample used in 2008
The sample drawn is spread across the country, with a higher concentration in metropolitan areas and less concentrated in the sparsely populated areas of the country such as Northern Cape and rural areas.

6. We flew a plane on to the sampled areas for ease of sampling households

7. Sampling participants
Select primary sampling unit (EA)
Select secondary sampling units (15 VP’s/households in each sampled EA)
Refer to aerial photos and data kits on EAs
Select ultimate sampling units (at most 4 individuals from each sampled VP, 1 from each age group)
Children under 2 years
Children aged 2–14 years
Youth aged 15–24 years
Adults aged ≥25

8. Questionnaires Visiting Point Questionnaire
Questionnaire for parent/guardian of children aged under 2 years
Questionnaire for parent/guardian of children aged 2-11 years
Questionnaire for children aged years
Youth Questionnaire for persons aged years
Adult Questionnaire for persons aged ≥25 years

9. Blood specimens DBS specimens taken
Tested for HIV antibodies – prevalence
Confirm all positives
Confirm 10% of negatives
Used BED for incidence (not reported in this study)
Detected ARVs (also not reported in this presentation)
BED and ARV results will be presented in later publications.

10. Ethics Ethics approval
HSRC’s Research Ethics Committee (REC 2/23/10/07)
CDC’s Institutional Review Board (IRB)
Global AIDS Programme
Informed consent and child assent
Linked anonymous HIV testing

11. Results
Prevalence
Behaviour
Knowledge

12. HIV Prevalence Results
This section covers information on the existing HIV cases as well as new infections.
Trend data is provided as well as for 2008 only where appropriate.

13. HIV Prevalence Year HIV Prevalence (%) 95% CI 2002 11.4 10.0-12.7 2005
10.8
2008
10.9
The study results show that HIV prevalence in 2008 was 10.9%, no different from the last two surveys.
The epidemic has stabilised at this level of nearly 11% and remained so since 2002.
In the next slide we take a look at the 2008 South African epidemic curve

14. HIV prevalence among 15 - 49 year olds by province, South Africa 2008
This slide shows HIV prevalence in the population aged years for 2008.
Kwazulu Natal continues to have the highest prevalence, followed closely by Mpumalanga.
Western Cape and Northern Cape have the lowest HIV prevalence
With the rest of the provinces in the middle

15. HIV prevalence, by sex and age, South Africa, 2008
This slide shows the prevalence of HIV by sex and age group for 2008.
The results show
First, HIV prevalence continues to be high among females compared to males
Second, the HIV prevalence peaks at ages for females earlier than that of males, which occurs at aged
The peak for women remains stubbornly high at 33% for all the three surveys, and for men it peaks at a new high at 25.8%.
Third, South Africans older than 50 have high HIV prevalence, with 1 in 10 aged being HIV positive. This may be a cohort effect, that is people infected earlier now living longer because of ARV treatment.

16. HIV prevalence among the most-at-risk populations, South Africa 2008
95% CI
African females 20-34
1395
32.7
African males
944
23.7
Males 50 years and older
946
6.0
Men who have sex with men 86
9.9
People who are high-risk drinkers
965
13.9
Persons who use drugs for recreational purposes
490
10.8
People with disabilities
458
14.1
One of the ways of containing the spread of HIV is to identify the high risk groups and develop interventions targeted at them. In many countries the most-at-risk groups are defined as sex workers, injection drug users and men who have sex with men.
In South Africa, the most at risk groups in addition to the one I just listed, are African females aged 20-34, African males aged and so on.
This slide shows HIV prevalence in these groups.. The results show
African females to have the highest HIV prevalence
Followed by African males.
Alcohol users as well as drug users have high HIV prevalence.
The people with disabilities are included as high risk in line with the NSP.
Now moving on to a new topic
HIV prevalence tells us about the percent of the population with HIV at the given time. It does not tell us when people were infected and whether the programmes are working or not. To understand whether the programmes are working we need incidence. The NSP requires that a calculation be made on HIV incidence for young people aged

17. Behavioural Results
So far we have presented evidence on the extent of HIV in South Africa in different age groups and different provinces, we have also presented information on HIV incidence.
What we now want to present are the behavioural determinants that increase risk of HIV.
We know that the most common mode of HIV transmission in South Africa is through heterosexual sex. The following section presents a set of key NSP indicators related to sexual behaviour risks for HIV infection, namely,
the age of sexual debut,
multiple sexual partnerships,
unprotected sexual intercourse and
age mixing

18. Sexual debut by sex among 15-24 year olds, South Africa 2002, 2005 and 2008
On sexual debut
Early sexual debut increases vulnerability to HIV infection among young people, especially females. It is thus important to know the age at which sexual debut occurs in order to inform HIV prevention interventions targeted at young people.
This slide shows that <10% of year olds had started having sex before the age of 15 years.
About twice as many males were found to have started having sex earlier in comparison to females
Trend remained the same for the entire period

19. Within 5 years of own age (%) Partner is 5+ years older (%)
Intergenerational sex among year olds, South Africa 2005 and 2008
2005
Within 5 years of own age (%)
Partner is 5+ years older (%)
Male
98.0
2.0
Female
81.4
18.5
Total
90.4
9.6
2008
98.5
0.7
72.4
27.6
85.1
14.5
The second behavioural determinant is intergenerational sex.
Intergenerational sex, or age mixing, is an important social determinant of HIV infection. One way of determining age mixing is to calculate the age differential. For example, youth who have partners 5 years and older than themselves expose themselves to HIV, as it exposes them to a higher prevalence age group.
This slide shows the rates of teenage intergenerational sex for 2005 and 2008:
The findings show that:
Among people who reported having partners who were 5 years older than themselves there was a substantial increase from 9.6% in 2005 to 14.5% in The same pattern of findings were also found amongst females and the percentage increased substantively from 18.5% in 2005 to 27.6% in 2008.

20. Multiple sexual partners, South Africa 2002, 2005 and 2008
The third behavioural determinant is having multiple sexual partners.
Having multiple sexual partners substantially increases the risk of acquiring HIV as it contributes to sexual networks that allow for pathways for HIV transmission.
This slides presents data on multiple sexual partners by age in SA from The results show:
Young males tend to have more multiple sexual partners than females and the rates increased in each survey
In 2008 five times more males (30.8%) reported having had more than one sexual partner in the past 12 months than their female counterparts (6.0%)
Older females tend to have fewer multiple partners compared to the rest of the groups on the slide.
The next slides examines the same phenomenon by province

21. Why Multiple Partners
Change in values- normalization of multiple sexual partnerships
Transactional reasons – exchanging of sex for resources – mainly young girls – driven by poverty / materialism
Sugar daddies & mommies – prefer younger partners while they have partners their own age
Married males who are unhappy at home – “strengthening their homes”
Some males my prefer to have sex with other males but because of expectations from society hide their sexual preferences
Power dynamics in relationships – men and women who have multiple partners because they can

22. Most-at-risk population with multiple sexual partners, South Africa 2002, 2005 and 2008
We further analysed the rates of multiple sexual partnership in the high risk groups
This slide presents the findings, which show that the high risk drinkers and those who use recreational drugs have high rates of multiple sexual partners.
African males aged also have high rates of multiple sexual partners compared with African females aged
These suggest that interventions on responsible alcohol use and harm reduction programmes are needed

23. Condom use by age group and sex, South Africa 2002, 2005 and 2008
The fourth behavioural determinant under study is condom use.
Consistent and correct condom use is one of the most effective means available for preventing HIV infection today. One of the indicators used widely to measure this is the level of condom use during the last sexual intercourse.
This slide shows reported condom use by sex and age group in the three surveys.
The graph shows that the same linear trend of increases in condom use was seen for both the and age groups from 2002 to 2008.
The pattern among those 50 years and older was slightly different as the rates of condom use reported by both sexes were unchanged from 2002 to 2005 but increased by as much as five-fold in 2008.
Further analysis of condom use is done by province and among the most at risk population

24. Condom use by most-at-risk populations at last sex, South Africa 2002, 2005 and 2008
This slide also shows that those who at high risk are also heeding HIV prevention message in so far as condom use is concerned. They all have increased condom use rates.
We have now presented information on behaviour determinants of HIV, we turn attention to HIV testing.

25. Respondents aged 15 years and older who ever had an HIV test, South Africa, 2002, 2005 and 2008
Yes %
Male
21.4
Female
Total
2005
27.6
32.9
30.5
2008
43.0
56.7
50.8
VCT is important as an entry strategy for both prevention and access to treatment, care and support services. Increasing knowledge of HIV status is important as it has been linked to an increase in prevention behaviours especially for those who test positive through VCT.
This slide shows the proportion of respondents aged 15 years and older who reported having had a history of HIV testing.
It shows that there large increases in HIV testing rates from 2002, through 2005 to from 21.4% to 50.8%.
It is interesting to note that while HIV testing was equal between the two sexes in 2002, there was a significant sex disparity seen in both 2005 and 2008 with more females being tested that their male counterparts.
This shows that the message on HIV testing, coupled with increased VCT sites and PMTCT programmes have contributed to interest in getting tested and access to testing. It is possible that ARV availability has also motivated people to get tested.
The next level of analysis is on knowledge of HIV transmission and prevention

26. HIV knowledge by sex and age group, South Africa 2005 and 2008
Knowledge of various aspects of HIV/AIDS allows for appropriate actions to be taken in relation to prevention, among other aspects.
This slide presents the percentages of adults who both correctly identify ways of preventing the sexual transmission of HIV and who reject major misconceptions about HIV transmission by age group. Using 2005 as a baseline compared to 2008, there were decreases in accurate knowledge about HIV transmission among all age groups from mainly over 40% to below 32% on average except among those males 50 years and older which remained unchanged at 32.0% and 27.9% respectively .

27. Conclusions: successes
Epidemic has stabilized at high levels
Reduction in HIV prevalence among children suggesting that PMTCT is making is making a difference
Slight reduction of HIV in the youth
Increased awareness of HIV sero-status especially among women
HIV prevalence decreased among
adults in the Western Cape and Gauteng provinces
Substantial increase of condom use among youth and all other age groups including women

28. Conclusions: challenges
Continued rising HIV prevalence among adults in KwaZulu-Natal and Eastern Cape
HIV prevention knowledge has declined- Multiple sexual partners
High rates of HIV prevalence among women in the reproductive age.

29. Recommendations
Need for programmes to help people to have pregnancy without risking HIV
Implement anti-multiple and intergenerational sexual partnerships campaigns that are community driven
Address high sexual partner turnover and intergenerational sex by changing community norms

30. Recommendations cont’d
Implement provider-initiated routine
HIV testing in all health care facilities

31. Research consortium Human Sciences Research Council (HSRC)
Medical Research Council (MRC)
Centre for AIDS Development, Research and Evaluation (CADRE)
National Institute for Communicable Diseases (NICD)

32. Funded by
US Centers for Disease Control and Prevention (CDC) through Funding Opportunity Announcement Number:
CDC-RFA-PS06-614
(Catalog of Federal Domestic Assistance Number: ) Program to Improve Capacity of an Indigenous Statutory Institution to Enhance Monitoring and Evaluation of HIV/AIDS in the Republic of South Africa as Part of the President’s Emergency Plan for AIDS Relief (PEPFAR)

33. Thank you for your attention