SUMMARY REPORT

NATIONAL HIV AND SYPHILIS SERO-PREVALENCE SURVEY IN SOUTH AFRICA

2001

Directorate: Health Systems Research, Research Coordination and Epidemiology

VERSION 2

Table of Contents

Preamble
Acknowledgements
1. Introduction
2. Objectives of the Survey
3. Survey Methodology

3.1 Study design
3.2 Study administration
3.3 Quality Assurance
3.4 Sentinel population
3.5 Sampling methodology
3.6 Inclusion criteria
3.7 Laboratory testing

3.7.1 HIV testing
3.7.2 Syphilis testing

3.8 Data processing and analysis

4. Results

4.1 Total sample size
4.2 HIV prevalence

4.2.1 National HIV prevalence
4.2.2 Provincial HIV point prevalence estimates
4.2.3 HIV point prevalence estimates by age group
4.2.4 Extrapolation of HIV infection to the population of South Africa

4.3 Syphilis prevalence

4.3.1 National syphilis prevalence
4.3.2 Syphilis point prevalence by age group

5. Discussion
6. Conclusions
7. References

PREAMBLE

This report is the 12th in a series of national antenatal surveys designed to track the magnitude of the HIV epidemic in South Africa. The results of this survey indicate that South Africa, like most of the African region is still confronted with a major health problem with vast social, economic and development implications.

The antenatal survey findings over the last several years along with data obtained on STI prevalence further support trends and patterns observed in recent years: that the pace at which the HIV epidemic was growing between 1990-1998 has slowed down and has begun to level off. Intervention programmes seem to be effectively taking ground amongst teenagers where we have seen an encouraging decrease in HIV prevalence between 2000-2001.

The findings of this report provide renewed optimism in what has become one of the greatest challenges our country has ever faced. They place us in a better position to tackle the challenges of care and support while sustaining the momentum in prevention activities. The initiatives in treatment, care, support and prevention against HIV and AIDS, TB, STIs and other opportunistic infections by different role-players is well appreciated.

Whilst we acknowledge the contribution of all in responding to prevention and intervention activities to combat the HIV and AIDS epidemic, it is the responsibility of all South Africans in all sectors and walks of life to participate actively in the fight against HIV and AIDS. Social mobilisation and intersectoral collaboration will ensure that our interventions also translate in decreased mortality and morbidity due to HIV and AIDS.

The emerging trends mark the beginning in successes in the fight against HIV and AIDS. South Africa has positioned itself to be effective in this endeavour and is committed in partnership with its National and International initiatives to combat HIV and AIDS, STI and TB, and thus reach its social and economic development targets.

DR M E TSHABALALA-MSIMANG
MINISTER OF HEALTH
DATE

ACKNOWLEDGEMENTS

I would like to thank all those made this survey possible. In particular I would like to thank the Provincial Heads of Departments for their support to this important exercise. Special thanks to all provincial survey co-ordinators, F Meidany & A Ntoto (EC), L van der Bank (FS), M Mpama and D Monnakgotla (GP), Z Ahmed & A Smith (KZN), B Harris (MP), R Rabie (NC), H Metsileng and M Naicker (NW), M Mogoswane (LP) and N Shaikh (WC) as well as all the nurses and staff in all the 421 clinics which participated in the 2001 survey.

My gratitude goes to the HIV testing laboratories; the Virology Departments of the Cape Town and Natal Universities, the Microbiology Department of the Medical University of Southern Africa (MEDUNSA), the National Institute for Communicable Diseases (NICD) and the National Health Laboratory Services (NHLS) who have ensured quality laboratory support to the survey.

Many thanks to colleagues at the National Department of Health for all their support and in particular I would like to thank the Directorate that is responsible for co-ordinating the survey. The Chief Director: Health Information, Evaluation and Research Dr Lindiwe Makubalo; Ms Pakiso Netshidzivhani, Dr Rose Mulumba, Ms Matsie Ratsaka-Mothokoa, Ms Rika Du Plessis, Mr Leonard Mudzanani, Ms Caron Johnson, Mr Fred Shikweni and Ms Lusanda Mahlasela.

Finally I would like to acknowledge all the women who participated in the study without whom the survey would not have been possible

DR AYANDA NTSALUBA
DIRECTOR-GENERAL: HEALTH

1. INTRODUCTION

HIV and AIDS is currently one of the most devastating health conditions affecting the health of millions throughout the world. UNAIDS estimates that approximately 60 million people have been infected since the beginning of the epidemic and in 2001 alone an estimated 5 million would have acquired HIV infection1. A significant proportion of these individuals reside in Sub Saharan Africa.

The health and social implications of HIV and AIDS on human development are extensive2 and have resulted in an expanded national and international effort to respond to the HIV and AIDS epidemic with increased funding and expansion of prevention, care and intervention activities. Notable initiatives include the Global Fund against HIV and AIDS, TB and Malaria, and The New Partnership for Africa's Development (NEPAD) programme. In South Africa activities have expanded and funding of HIV and AIDS has increased significantly from the 2001/2 to the 2002/3 financial year.

Effective monitoring and surveillance of the HIV and AIDS epidemic is an essential component of a dynamic and effective national programme. National and even global estimates of HIV infection are limited by relatively weak monitoring systems. In many countries monitoring systems are not designed to give a representative sample of the total population and in countries with large populations, where one province or state would have the population of a country such as South Africa, the prevalence estimates from sentinel sites could all but lose meaning.1

In South Africa the Department of Health (DOH) instituted a mechanism to annually monitor the HIV epidemic since 19903. This surveillance system was reviewed, and the sampling design modified to give nationally representative estimates of HIV prevalence. A more comprehensive understanding of the epidemiology of HIV and AIDS will be possible with information obtained from the Department's expanded second generation surveillance programme which will indicate new infection rates as well as associated behavioural factors4.

This report presents highlights of the 2001 antenatal survey conducted in the 9 provinces of South Africa.

2. OBJECTIVES OF THE SURVEY

The primary objectives of the 2001 survey were:

The secondary objective of the 2001 study was:

3. SURVEY METHODOLOGY

3.1 Study design

This survey was conducted concurrently across all nine provinces from the 1st to 31st October 2001. The study was an anonymous, unlinked, cross-sectional survey5. The study population included pregnant women who attend antenatal clinics in the public health sector of South Africa.

3.2 Study administration

A national workshop of provincial co-ordinators was held to review the protocol, review study methods, plan field logistics review standard operating procedures, and review field logistics, (e.g. Provision of 5 ml Vacutainers with SST gel and clot activator, paired bar-code labels, and data capture sheets). It also prepared the team for data analysis and processing procedures.

Participating laboratories and courier services were prepared to ensure that appropriate testing procedures would be adhered to and blood specimens safely and timely transported from each antenatal clinic to testing laboratories.

The national department's co-ordinating office ensured the overall co-ordination of the survey including support visits to the nine provinces, conducting the procedural audit of the survey countrywide, re-processing provincial estimates, processing of the national data set, modelling further HIV estimates and compiling a singular national report. The procedural audit included investigation into preparation of sentinel sites, logistics and adherence to the study protocol in general.

3.3 Quality Assurance

In the course of this survey, careful laboratory and data management quality assurance was conducted. For each of the participating laboratories, the ELISA and RPR tests for HIV and syphilis testing were internally quality assured. The National Institute for Communicable Diseases (NICD) performed external quality control for the HIV testing and the Medical University of Southern Africa (MEDUNSA) Microbiology laboratory conducted quality control for the syphilis testing. Double data-capture was conducted on provincial data sets.

3.4 Sentinel population

The sentinel population for the study includes pregnant women attending a public sector antenatal clinic for the first time during the current pregnancy. The choice of the first antenatal visit is made to minimise the chance for one woman attending two clinics and being included in the study more than once. The survey builds on routine screening for blood grouping, syphilis, and full blood count (FBC) that takes place during a woman's first antenatal visit.

3.5 Sampling methodology

The probability proportional to size (PPS) sampling method was used to determine the sample size for the 2001 antenatal HIV survey5. The clinic formed the primary sampling unit in the sampling frame; this method is adopted to ensure a representative sample that is weighted for rural urban distribution. A total sample of 16 743 specimens was included in the survey.

Provinces with the biggest population sizes (of women in the reproductive age) yielded the biggest sample sizes, whereas those with smaller population sizes had smaller sample sizes. A total of 421 sentinel sites (clinics) participated in the 2001 survey. All sites were required to consider for inclusion in the sample the first forty antenatal attendees who qualify for inclusion in the study.

3.6 Inclusion criteria

Consecutive first time antenatal clinic attendees from the 421 sentinels sites were included in the survey until the target of 40 specimens was reached or the 31st October, whichever came first. Women attending each of the participating clinics for the first time in the current pregnancy were eligible for inclusion in the study. Blood was taken as part of the routine screening done for pregnant women when they attend the clinic for their "booking visit".

As the survey is anonymous no personal identifiers (names, ID number, address, etc) are used on the blood sample of the participant. Instead bar-coded numbers are used to ensure anonymity of the participants, to facilitate laboratory procedures and minimise the chances of errors during the handling of the blood specimens.

3.7 Laboratory testing

Laboratory testing was conducted in accordance with the national standardised survey protocol. Participating laboratories included the NHLS labs in Bloemfontein, Johannesburg, Kimberley, Nelspruit, Port Elizabeth as well as the Virology laboratories of the University of Natal-Durban and MEDUNSA. As already mentioned the NICD and MEDUNSA laboratories conducted quality controls.

3.7.1 HIV testing

The HIV testing methodology employed is the World Health Organisation (WHO) recommended procedure for antenatal surveys6. As required by that protocol blood specimens were tested with one ELISA test (Abbot Axysm System for HIV-1/HIV-2) in all provinces except the Western Cape where two tests were performed. According to the WHO protocol low HIV prevalence as in the Western Cape province (less than 10%), requires that two ELISA tests be used. Sera found to be reactive on the first assay were retested with the second ELISA test, whereas those shown to be non-reactive on the first test were considered HIV antibody negative and therefore not retested. All positive HIV sera were stored for the pilot HIV incidence testing.

3.7.2 Syphilis testing

Syphilis testing was conducted using the RPR test. 16 701 specimens were tested and results for 42 specimen were excluded from RPR testing for quality reasons.

3.8 Data processing and analysis

Data analysis was conducted using the STATA and SPSS software packages. Data entry was standardised using the EPI-INFO programme designed and piloted with the provinces during the planning session. Data analysis was conducted at both the provincial and national levels to verify analysis.

4. RESULTS

4.1 Total sample size

A total number of about 16 743 pregnant women participated in the survey (HIV results were not recorded for 13 specimens and therefore not included in the analysis, whilst 42 were excluded for RPR). Thus 16 730 specimen were tested for HIV and 16 701 were tested for syphilis.

4.2 HIV prevalence

4.2.1 National HIV prevalence

Based on the 16 730 blood samples tested for HIV during the survey, it is estimated that nationally, 24.8% of pregnant women were infected with HIV by the end of 2001. This is in comparison with a prevalence rate of 24.5% recorded for 2000. Whilst this rate of infection is high and a significant public health problem the findings indicate that there is not a statistically significant growth in the epidemic from the previous year. Figure 1 shows that the national prevalence rate of increase has slowed in its increase since 1998. This is referred to as a levelling off, plateau or stabilisation in growth.

Figure 1 National HIV prevalence trends among antenatal clinic attendees in South Africa: 1990-2001

4.2.2 Provincial HIV prevalence estimates

Provincial prevalence rates show the geographic variations in the HIV epidemic. Table 1 show the HIV rates over the last three years.

Table 1: Provincial HIV prevalence estimates: Antenatal clinic attendees, South Africa 1999-2001

PROVINCE
HIV pos. 95% CI
HIV pos. 95% CI
HIV pos. 95% CI
1999
2000
2001
KwaZulu-Natal (KZN) 32.5 (30.1-35.0) 36.2 (33.4-39.0) 33.5 (30.6-36.4)
Mpumalanga (MP) 27.3 (25.2-30.7) 29.7 (25.9-33.6) 29.2 (25.6-32.8)
Gauteng (GP) 23.9 (21.7-26.0) 29.4 (27.2-31.5) 29.8 (27.5-32.1)
Free State (FS) 27.9 (24.7-29.8) 27.9 (24.6-31.3) 30.1 (26.5-33.7)
North West (NW) 23.0 (19.7-26.3) 22.9 (20.1-25.7) 25.2 (21.9-28.6)
Eastern Cape (EC) 18.0 (14.9-21.1) 20.2 (17.2-23.1) 21.7 (19.0-24.4)
Limpopo (LP) 11.4 (9.1-13.5) 13.2 (11.7-14.8) 14.5 (12.2-16.9)
Northern Cape (NC) 10.1 (6.6-13.5) 11.2 (8.5-13.8) 15.9(10.1-21.6)
Western Cape (WC) 7.1 (4.4-9.9) 8.7 (6.0-11.4) 8.6 (5.8- 11.5)
National 22.4 (21.3-23.6) 24.5 (23.4-25.6) 24.8 (23.6 - 26.1)

N.B. The true value is estimated to fall within the two confidence limits, thus the confidence interval is important to refer to when interpreting data.

Figure 2 HIV prevalence by province among antenatal clinic attendees in South Africa, 2000-2001

Figure 2 above shows HIV prevalence by province in 2000 and 2001. The trends reflect the HIV prevalence rates the provinces have reached. It is worth noting however that HIV prevalence decreased but not statistically significantly in KwaZulu-Natal between 2000 and 2001. HIV prevalence rates did not change significantly (statistically) in Mpumalanga, Gauteng and Western Cape provinces between 2000 and 2001. Provinces which had marginal increases (also not statistically significant) include Limpopo, Free State and the Eastern Cape provinces. There were however larger increases in the Northern Cape and North West Provinces; the increase in the Northern Cape was marginally significant statistically.

4.2.3 HIV point prevalence estimates by age group

There were different trends between young and older women. HIV prevalence among teenagers was estimated at 15.4%. This represents a decline in HIV prevalence between 2000 and 2001. However this is not statistically significant. HIV prevalence in the 20 to 29 year age group has not shown an increase from the previous year whilst there is a significant increase in HIV prevalence amongst women in the 30 to 39 year age category.

Table 2: HIV prevalence trends by age group among antenatal clinic attendees in South Africa 1999-2001

AGE GROUP
Est HIV+ (95%CI)
Est HIV+ (95%CI)
Est HIV+ (95%CI)
1999
2000
2001

<20

 16.5 (14.9-18.1) ) 16.1 (14.5 -17.7 15.4 (13.8 -16.9)
20-24 25.6 (24.0-27.3) 29.1 (27.4-30.8) 28.4 (26.5 -30.2)
25-29 26.4 (24.6-28.3) 30.6 (28.8-32.4) 31.4 (29.5 -33.3)
30-34 21.7 (19.1-23.8) 23.3 (21.5-25.1) 25.6 (23.5 -27.7)
35-39 16.2 (14.1-18.3) 15.8 (13.9-17.7) 19.3 (17.0 -21.5)
40-44* 12.0 (8.5- 15.6) 10.2 (6.9- 13.3) 9.1 (6.2 -11.9)
45-49* 7.5 (-.77-15.9) 13.1 (2.1-24.0) 17.8 (4.3 -31.4)

*The true value could lie anywhere between the Confidence Intervals (CI)

N.B. The sample size for women in the 45 to 49 year age group is small. Thus the prevalence rate in this group should be read with caution, as confidence intervals are wide.

Figure 3 HIV prevalence by age group among antenatal clinic attendees in South Africa, 2000-2001

Figure 4 HIV Prevalence by Age, South Africa Antenatal Clinics. 1991 - 2001

4.2.4 Extrapolation of HIV infection to the population of South Africa

The Department of Health uses a model, which it developed to generate estimates of HIV infection in the general South African population7 On the basis of this model, it is estimated that in 2001 2.65 million women between the ages of 15 to 49 were infected with HIV and 2,09 million men between 15 and 49 were infected with HIV. It is estimated that 83 581 babies became infected with HIV through the mother-to child transmission route.

It is estimated that by 2001 approximately 4.74 million people in South Africa between 15- 49 years of age had become infected with the HI virus. This is in comparison with 4.70 million who were infected by 2000.

4.3 Syphilis prevalence

4.3.1 National syphilis prevalence

This study found that 2.8% of women who are pregnant and presenting at public maternal health facilities had syphilis infections. Though still a concern, this figure represents a decline from 4.9 % in 2000. The syphilis prevalence trend is very notable as it has shown a constant decline over the last several years (Table 3 and Figure 5). Syphilis prevalence in the Northern Cape requires closer monitoring as the Northern Cape has shown a significant increase in syphilis prevalence between 2000 and 2001.

Figure 5 National Syphilis Prevalence among antenatal clinic attendees in South Africa: 1997-2001

Table 3: Syphilis prevalence estimates among antenatal clinic attendees in South Africa, 1999-2001

PROVINCE
RPR pos. 95% CI
RPR pos. 95% CI
RPR pos. 95% CI
1999
2000
2001
KwaZulu-Natal (KZN) 3.8 (3.1 - 4.5) 2.6 (1.6 - 3.7) 1.3 (0.9 - 1.7)
Mpumalanga (MP) 9.7 (8.0 - 11.5) 3.7 (2.4 - 5.0) 2.0 (1.3 - 2.7)
Gauteng (GP) 9.6 (8.3 - 10.8) 9.6 (8.2 - 11.0) 2.7 (2.0 - 3.3)
Free State (FS) 15.8 (13.5 - 18.0) 4.8 (3.6 - 6.0) 2.0 (1.1 - 2.9)
North West (NW) 5.6 (4.2 - 7.1) 3.6 (2.1 - 5.2) 4.0 (2.8 -5.2)
Eastern Cape (EC) 4.4 (3.5 - 5.4) 3.3 (2.5 - 4.1) 3.3 (2.4 - 4.2)
Limpopo province (LP) 8.6 (6.8 - 10-5) 4.2 (3.2 - 5.2) 4.9 (3.6 - 6.2)
Northern Cape (NC) 3.8 (2.0 - 5.6) 5.1 (3.1 - 7.1) 6.2 (4.0 - 8.5)
Western Cape (WC) 4.4 (3.0 - 5.8) 5.2 (4.3 - 6.1) 2.9 (2.1 - 3.7)
National 7.30 4.90 2.80

Figure 6 Syphilis prevalence estimates among antenatal clinic attendees by province in South Africa, 2000-2001

Provincial prevalence trends (see Figure 6) across the nine provinces show overall decline in syphilis trends with the exception of the Northern Cape between 2000 and 2001.

4.3.2 Syphilis point prevalence by age group

Figure 7 shows a decrease in prevalence between 2000 and 2001 in all age groups. The lowest syphilis prevalence rate in 2001 was noted among women aged 45-49 years.

Figure 7 Syphilis prevalence trends by age group among antenatal clinic attendees in South Africa, 2000-2001

5. DISCUSSION

The findings of the 2001 antenatal survey show that HIV and AIDS is a significant health problem in South Africa. It is estimated that 4.74 million individuals between 15-49 years are infected with the HI virus despite the intensive HIV prevention programmes that are in place. These high prevalence rates have significant implications on the future burden of HIV associated disease and the ability of the health system to cope with provision of adequate care and support facilities.

The most striking observations that can be made on HIV prevalence estimates however is that the prevalence of HIV though still high, has not increased between 2000 and 2001. These findings confirm the observation that the rapid growth of the South African epidemic may be slowing down8

Whilst HIV prevalence trends have their limitations as a marker of overall reduction in new HIV infections9, this slowing down factor is supported by what is emerging from syphilis trends and HIV trends in teenagers. Syphilis infection is an important indicator of greater biological susceptibility to HIV infection and sometimes an indication of possible high-risk behaviour of the infected individual. The 2001 antenatal survey has shown an ongoing reduction in rates of syphilis in women at antenatal care facilities. As shown in this report the decrease since 1998 is very notable. It is reasonable to associate the reduction in syphilis at antenatal clinics with a host of intervention activities and in particular the intensified programme on syndromic treatment and management of STIs in antenatal service provision points.10

The second set of data, which provides information supportive of the argument that the HIV epidemic in South Africa is stabilizing relates to HIV prevalence rates in women/girls under 20 yrs. HIV trends in teenagers are considered a good indicator of behavioural change aimed to reduce HIV infection such as a delay in sexual debut and condom use. As demonstrated in this report the early decline and stabilization are very encouraging.

HIV trends

The sampling design for the national HIV prevalence study is made to enable reliable estimates of HIV at provincial level. As expected, the provinces are at different stages of the HIV epidemic and the epidemiological pattern on HIV is not the same for all provinces12

To illustrate these differences, the observation is made that KwaZulu-Natal which had reported the fastest growing epidemic and which still has the highest provincial prevalence rate is now stabilising at 33.5 %. As these studies yield prevalence and not incidence estimates, the reasons for this decrease may be numerous. The Western Cape on the other hand, has the lowest HIV prevalence rate in the country, but its HIV rate may also be stabilising at this relatively low HIV level.

Syphilis trends

Syphilis sero-prevalence in South Africa is at its lowest level ever(2.8%)since the beginning of surveillance. Syphilis trends have decreased for the third consecutive year. A remarkable decline has been observed in almost all provinces except the Northern Cape.

Effective prevention and treatment is key to successes in this area. The STI syndromic management programme and the screening and treatment of syphilis are routine activities conducted at antenatal clinics. This programme alongside others is beginning to show the effectiveness of some intervention programmes. These findings refer to the occurrence of new syphilis cases only and other surveillance activities such as the PHC routine surveillance programme will shed more light on overall STI infection trends.

6. CONCLUSIONS

Up until 1998 South Africa was described as having one of the fastest growing epidemics in the world. The 2001 survey confirms that the trend seen since the 1998 survey indicates that this is no longer the case.

The findings of this survey are however an important pointer to the magnitude of the HIV and AIDS problem that South Africa confronts, where an estimated 4. 74 million individuals are infected with HIV. The full participation of all sectors of society and the strengthened intersectoral involvement in implementing the HIV and AIDS and STI strategic initiatives will be critical to an effective national programme and maintaining the gains that South Africa is beginning to observe. The epidemiological trends in HIV prevalence should at this stage be a source of encouragement and leave no room for complacency as the numbers of individuals who are acquiring HIV infections each day is still high and the health care implications of the current infections are enormous.

This HIV survey and the expanded surveillance programme which places emphasis on high risk behaviour monitoring and monitoring HIV incidence will continue to be important in underpinning areas to strengthen the national response to HIV and AIDS.

7. REFERENCES

1. UNAIDS epidemic update 2001: Joint United Nations Programme on HIV and AIDS (UNAIDS)/WHO.
2. Department of Health Technical Report, 1998. HIV and AIDS in South Africa: Impacts and Priorities.
3. Kustner H G, Swanevelder J P, & van Middelkoop A 1998. The South African HIV epidemic, reflected by nine provincial epidemics, 1990-1996. South African Medical Journal.88 (1). 1320-1235.
4. Department of Health, 1998. The South African HIV and AIDS/STD Second Generation Surveillance System.
5. Department of Health Technical Report, 1997. Annual Antenatal Survey Protocol.
6. WHO Global Programme on AIDS (1989). Unlinked anonymous screening for the public health surveillance of HIV infections. International Guidelines.
7. Department of Health (1996), Epidemiological Comments Vol. 23 no. 1
8. Makubalo L Simelela N, Mulumba R, Levin J: 1999 Antenatal Survey results: Little Room for Pessimism. SAMJ Nov. 2000, Vol 90, no.11
9. UNAIDS/WHO (2000) WORKING group on HIV and AIDS & STI surveillance. Guidelines for 2nd Generation HIV surveillance
10. Department of Health, 2000. HIV and AIDS/STD Strategy plan for HIV and AIDS and STI (2000-2005).
11. Makubalo L, Levin J, Zaba B, Mulumba R, Bikitsha N, Mahlasela L, 2000. Recent trends in HIV prevalence in South Africa. Aids Conference 2000 - Durban.
12. Allen D, Simelela N, Makubalo LE (2000): Epidemiology of HIV and AIDS in South Africa. Journal of Science.

LIST OF CONTRIBUTORS

  1. Dr LE Makubalo- Cluster Manager: Health Information, Evaluation and Research (HIER)
  2. Ms PM Netshidzivhani- Director: Health Systems Research, Research Coordination and Epidemiology (HSRRCE)
  3. Dr R Mulumba- Director: HIV and AIDS and STI's
  4. Dr J Levin- Medical Research Council
  5. Ms H du Plessis- HIER
  6. Ms M Ratsaka-Mothokoa- HSRRCE
  7. Ms L Mahlasela- HSRRCE
  8. Mr L Mudzanani- HSRRCE
  9. Ms C Johnson- Communicable Disease Control
  10. Mr F Shikweni- M& E Unit