AFFIDAVIT I, the undersigned, QUARRAISHA ABDOOL KARIM do hereby make oath and state as follows: 1) The facts deposed to in this affidavit are within my personal knowledge except where I indicate otherwise. To the extent that I rely on information supplied by others, I believe such information is true and correct. 2) I am an epidemiologist by training. I hold the following degrees BSc, BSc (Hons), MS, PhD. I have extensive experience in the field of HIV/AIDS on which I have published widely in scientific journals. From 1995 – 1996 I was the National Director of the HIV/AIDS and Sexually Transmitted Diseases programme of the Department of Health in the first democratic government. (Annexure: QAK1 –Curriculum Vitae) 3) I am currently Director of the Southern African Fogarty AIDS Training Programme, Executive Member of the HIV Prevention Trials Network (HPTN), Chairperson of the HPTN Ethics Working Group, consultant to the International AIDS Vaccine Initiative, Honorary Associate Professor in Epidemiology at the Faculty of Medicine, Nelson R. Mandela Medical School, University of Natal and Adjunct Associate Professor in Epidemiology at the Mailman School of Public Health, Columbia University. In addition, I serve as an ad hoc technical advisor to the Joint United Nations Programme on HIV/AIDS (UNAIDS) and the World Health Organisation (WHO). 4) The purpose of this affidavit is to describe the epidemiology of HIV infection in South Africa, and its implications for women and HIV transmission from mother-to-child. 5) When I describe HIV/AIDS prevalence I refer to all HIV infections, this includes old and new infections. By contrast, incidence rates refers to all new HIV infections within a defined time period INTRODUCTION 6) Women feature more strongly in this decade of the HIV pandemic compared to the first (UNAIDS, 2001). Sub-Saharan Africa is unique in having more women infected with HIV than men (UNAIDS 2001). The role of biological factors in contributing to the more efficient transmission of HIV from men to women compared to transmission rates from women to men has been well described. 7) Whilst there is increasing recognition of the role of social and political factors including gender on the increased vulnerability of women to acquiring infection with HIV, less empirical evidence exists. In countries and communities where HIV is transmitted predominantly through heterosexual contact, the increased vulnerability of women has implications for the provision of education and care and the transmission of HIV to their unborn infants. 8) The South African HIV epidemic has special features, such as: "explosive" spread; predominance in women at younger ages; and very high prevalence with no significant sign of a "saturation" plateau (Abdool Karim and Abdool Karim, 1999). 9) Prior to 1987, HIV infection was rare in the heterosexual population as shown by surveys of mineworkers (Sher,1986), voluntary blood donors (Crookes and Heyns, 1992), and from stored specimens from other community-based surveys (Abdool Karim et al, 1992, Abdool Karim and Abdool Karim 1992. By 1989, infection rates in a selection of well-attended antenatal clinics found a prevalence of about 1%, with a doubling rate of 14 months (Schall, 1990). 10) Three main sources document the HIV epidemic in South Africa in the 1990s: a) National, annual, anonymous, surveys among free, public antenatal clinic attenders (Department of Health, 2001); b) Local annual surveys amongst antenatal, STD and family planning clinic attenders in Hlabisa District, KwaZulu Province (Coleman and Wilkinson, 1997; Wilkinson et al 1997a; Colvin et al 1998; Wilkinson et al 1998; Wilkinson et al 1999a); and c) Surveys in the mining community of Carletonville mineworkers and sex workers (Williams et al, 2000a). 11) Eleven annual national antenatal clinic surveys since 1990 illustrate temporal trends across provinces among women by age and an understanding of the evolving epidemic (Department of Health, 2001). The South African government relies almost exclusively on the national, antenatal survey data for its planning and interpretation of the epidemic. 12) Hlabisa District data are more widely used, e.g. by the provincial government, the Joint United Nations Programme on HIV/AIDS, and also by the National government. In the absence of longitudinal studies, the data have provided statistically derived incidence rates (Williams et al, 2001) and from these surveys have been validation of the accuracy of the detuned assay to determine incident HIV infections in stored specimens (Gouws et al, submitted). 13) Carletonville data (Williams et al 2000a) include trends in age and gender specific prevalence rates in HIV. They are used by the Monitoring the AIDS Pandemic Working Group (MAP) in their monitoring task. 14) These assembled data have permitted development and tests of the robustness of mathematical projections that forecast longer-term effects of the epidemic (Williams et al, 2000b; Whiteside et al 1990; Doyle PR 1991). These models, useful for macrolevel short-term projections even allowing for doubtful assumptions, do not serve as well for local levels or for strata of the indigent, disenfranchised and marginalized where the epidemic rages most acutely. 15) Whilst these sources have shed light on the South African HIV epidemic, the AIDS dimension is shrouded in darkness. While we are aware of rising mortality rates and funerals, we know little or nothing about community level mortality estimates. How are families coping with the premature loss of breadwinners, with the burden of care for orphans, with the prospect and reality of intra-familial spread and child deaths? Lay press reports grow of dramatic increases in morbidity and mortality grow. Actual data (Wlkinson and Davies, 1997; Bobat et al 1990; Colvin et al 2001) remain sparse. Epidemiology – National Antenatal Surveys 16) South Africa is experiencing one of the fastest growing epidemics in the world (Abdool Karim and Abdool Karim, 1999) . HIV infection is found in all race groups in South Africa but is spreading about 10 times more rapidly amongst Black people (Crookes and Heyns; 1996). The rapidity with which HIV is spreading in the Black, heterosexual population in South Africa is most reliably demonstrated with data from the annual, anonymous antenatal surveys that have been conducted since 1990 in select public health sector facilities (Department of Health). 17) In the past 10 years, the HIV seroprevalence among first time antenatal clinic attenders has risen from 0.76% in 1990 to 10.44% in 1995 to 24.2% in 2000, with no significant sign that it has reached a plateau. Based on these surveys, it is estimated that there are currently 5 million South Africans infected with HIV. This rapidly growing HIV epidemic in South Africa is best described as explosive. 18) The geographical distribution of HIV infection is presented in Table 1. A gradient of infection can be discerned from east to west coast with the epidemic being most advanced in KwaZulu-Natal and Mpumalanga (at least two years ahead of the rest of the country) compared to the Northern Cape and Western Cape. Notwithstanding variations in prevalence between and within provinces the rate of new infections is similar across provinces. As the epidemic has matured the initial doubling time of 12-14 months has been lengthened to about 20-24 months. Table 1: Geographical distribution of HIV infection 2000 Prevalence (%) Western Cape 8.7% Eastern Cape 20.2% Northern Cape 13.2% Free State 27.9% KwaZulu-Natal 36.2% Mpumalanga 29.7% Northern Province 13.2% Gauteng 29.4% North West 22.9% Overall 24.5% 19) The age and gender differences in HIV prevalence is presented in Figure 1. Three cross-sectional, anonymous random population-based surveys of HIV prevalence (Abdool Karim et al, 1992; Abdool Karim, 2000) in 1990 (n=5023), 1991 (n=5605) and 1992 (n=5560), conducted in conjunction with the Department of Health Malaria Control Program found rapid progression from 1.2% (CI: 0.9-1.5) in 1990 to 2.5% (CI: 2.1-2.9) to 3.3% (CI: 2.9-3.7) in 1992 and stark gender disparity, with prevalence in women in 1990 fourfold that in men (CI:1.4-5.6). The disparity persisted but decreased somewhat, with risk ratios (RR) of about 2 (2.9%vs1.5%) (CI: 1.3-3.0) in 1991 and 1.9 (3.8% vs 2.5%) (CI: 1.1-2.1) in 1992. These studies showed a marked gender differences in age at infection, with a sharp and early rise beginning at 15-19 years in women, and deferred to 20-24 years in men (Figure 1). Migrants were at raised risk of HIV. In 1990, the RR for HIV infection due to migrancy was 3.1 (CI: 1.7-6.0). Here, gender disparity was reversed, unsurprising given male migratory labor concentrations. Among women, the age-adjusted RR was 2.4 (CI: 1.1-5.0) and among men, 7.3 (CI: 6.1-33.8). None of the subjects in any of the three community based surveys had antibodies to HIV-2. A community-based survey in Hlabisa in 1995 demonstrated that whilst the gender differences decreases over time, more women compared to men remain infected with HIV (Colvin et al , 1998) HIV prevalence and incidence rates in sentinel sites 20) Temporal trends point to an explosive epidemic in Hlabisa, a northern rural district of KwaZulu-Natal, and across all the similar rural districts of KwaZulu-Natal. Since 1992, anonymous HIV serosurveys in Hlabisa, (Coleman and Wilkinson 1997; Wilkinson et al 1999a) conducted among prenatal clinic attenders showed a rise in prevalence from 4.2% in 1992 to 29.9% in 1998. Incidence rose from 2.3% p.a. in 1993 to 15.0% in 1999 (Table 2) as estimated from age-prevalence by statistical methods developed by Eleanor Gouws, Brian Williams and Salim Abdool Karim (Williams et al, 2001). Table 2: Prevalence and Incidence of HIV infection among prenatal clinic attenders, aged 15-49 in Hlabisa: 1992-1999 Year N Prevalence of HIV (95% CI) Incidence per year 1992 884 4.2% (3.0-5.7) - 1993 709 7.9% (6.0-10.1) 2.3% 1995 314 14.0% (10.4-18.4) 7.2% 1997 4731 27.2% (25.9-28.5) 8.2% 1998 3166 29.9% (28.4-31.6) 9.9% 1999 3014 34.0% (32.3–35.7) 15.0% NOTE: Incidence rates were calculated using a mathematical model 21) More recently, the sensitive/less-sensitive (detuned) assay (Gouws et al, submitted) has been validated for clade C infections using panels of specimens from patients with acute infection and known dates of seroconversion. Based on this assay, incidence rates for 1999 show 20-24 year old women to be the most affected age- group (Table 3). The prevalence of HIV in the rural district of Hlabisa shows especially startling rates in 20-24 and 25-29 year old women (40% and 45.3% respectively). Table 3: Prevalence and incidence of HIV infection per age category:1999 Age group N Prevalence(%) (95%CI) Incidence (95% CI) 15-19 495 24.5 (20.7 – 28.3) 15.4 (13.0 – 18.0) 20-24 478 40.0 (35.6 – 44.4) 26.6 (22.8 – 30.5) 25-29 295 45.3 (39.6 – 50.9) 23.6(20.3–27.3) 30-34 242 32.4 (26.5 – 38.3) 16.7 (13.6 – 20.4) 35-39 155 22.1 (15.6 – 28.6) 10.9 ( 7.7 – 14.9) 40-44 46 20.7 ( 8.9 – 32.4) 29.9 ( 4.2 – 10.5) NOTE: Incidence rates were calculated using the detuned assay 22) Increasing age-specific prevalence in all age-groups from 1992 through 1998 further exemplify the dramatic progress of the HIV epidemic (Table 4). Table 4: Age trends in the prevalence of HIV infection in prenatal clinic attenders – three year intervals from 1992 to 1998 Prevalence(%) Age group 1992 1995 1998 20-24 6.9% 21.1% 39.3% 25-29 2.7% 18.8% 36.4% 30-34 1.4% 15.0% 23.4% 35-39 0.0% 3.4% 23.0% 23) In the 20-24 year age group, HIV prevalence grew dramatically from 6.9% to 21.1% three years later and to 39.3% another three years later. The effect of high incidence rates and the accumulation of HIV infected individuals from the younger groups has led to the prevalence of HIV infection to escalate from 0% in 1992 to 23% in 1998 in the 35-39 year age group (Table 4). 24) The estimated number of women and babies infected with HIV in South Africa – 1999 Age group (years) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Age-specific HIV prevalence (%) 16.5 25.7 26.4 21.7 16.2 12.1 7.5 Number of infected Women 355000 517850 475761 346172 214356 132127 67147 2108413 Estimated number of Births 302417 572401 538477 387325 208686 51077 15903 2076286 Number of babies Infected 11702 30468 24757 14176 6075 1563 278 89019 Estimated number of HIV infected persons (1999) Females Males Babies Total 2200734 606536 89019 3896289 Estimated number of HIV-infected persons, by province (1999) Adults Babies Total Western Cape 61841 1074 62915 Eastern Cape 503087 12611 515698 Northern Cape 34559 663 35222 Free State 335506 8126 343632 KwaZulu-Natal 1201832 29041 1230873 Mpumalanga 295492 7360 302852 Northern Province 246591 6472 253063 Gauteng 707736 14768 722504 North West 331198 8359 339557 Total 3717842 88474 3806316 THE SPECIFIC IMPACT OF HIV/AIDS ON WOMEN 25) Women have multiple, largely unrecognized, roles in society: amongst others, they are educators and care-givers in both the formal and informal settings, and custodians of societal values and norms. They ensure continuity of society. These contributions are difficult to measure, yet their importance will be known and felt only after their loss; many women will die of AIDS and it will take generations to recover from the loss of these women's gifts to society. In many communities, the burden of caring for sick family members falls on women; yet knowledge of a woman's HIV status often leads to ostracism, violence, and loss of security. One consequence of devaluing and repudiating women is that programmes aimed at reducing women's vulnerability to HIV infection face special challenges. 26) Women's vulnerability to HIV/AIDS stem from a range of social, economic, biological, cultural and legal factors (Whelan D, 1999). Economic dependency on men, commercial sex work, violence against women, reproductive health problems and unequal access to social services are some of the issues that illustrate the causes of this vulnerability to HIV/AIDS. This affidavit will show the impact of HIV on the health of women of reproductive age. 27) Physiologically women are at greater risk than men for HIV transmission. The risk of HIV infection during unprotected vaginal intercourse is 2-4 times higher for women than men. Women have a larger surface area of vaginal and cervical mucosa exposed to their partner's secretions during sexual intercourse. Semen infected with HIV also carries a larger concentration of the virus than vaginal fluids. 28) The South African Demographic and Health Survey conducted by Bradshaw and colleagues at the Medical Research Council for the Department of Health, confirmed that younger women carry the highest HIV burden of all age-sex groups, but showed in addition that this distribution holds for AIDS-related morbidity and mortality (Bradshaw D, personal communication). 29) In 1997 the then Minister of Health, Dr Nkosazana Zuma, established a National Committee on Confidential Enquiries into Maternal Deaths. The first Report on Confidential Enquiries into Maternal Deaths in South Africa, published October 1999, records 676 maternal deaths that occurred during 1998. The Second Interim Report on Confidential Enquiries into Maternal Deaths in South Africa was launched in November 2000. The Second Report recorded 774 maternal deaths for 1999 and increased the available data for HIV infection. 30) Maternal deaths are defined as "deaths of women while pregnant or within 42 days of the termination of pregnancy from any cause related to or aggravated by the pregnancy or its management, but not from accidental or incidental causes." The report distinguishes between direct and indirect causes, as well as women with HIV and women with AIDS. 31) AIDS-related deaths increased because of better surveillance between 1998 and 1999. In 1998, the leading recorded cause of maternal deaths was pregnancy- related hypertension (23.2%). In 1999, non-pregnancy related sepsis mainly caused by AIDS (29.6%) was the leading recorded cause. The reports hold that HIV is significantly under-diagnosed. In the Second Report 35.5% of women whose deaths were reported were tested for HIV, and 68% of them were HIV positive. The reports also suggest that as many as 2000 maternal deaths, particularly in rural areas, may not be recorded in the reports. 32) In its discussion section, the Report suggests that: "Most pregnant women who died as a result of complications of AIDS did so because of respiratory infections, although every organ system was represented. While there is very little potential for cure at present, much can be done to improve the quality of life and improve the pregnancy outcome. The use of prophylactic antibiotics in women with AIDS, the supplementation of their diet with vitamins and minerals, an altered lifestyle, specific management in labour, and the selective use of antiviral therapy to prevent vertical transmission of the virus [mother to child transmission] can all impact on the well-being of the HIV-positive woman. It is essential that national guidelines for the management of pregnant HIV-positive women and the management of pregnant women with AIDS are drawn up urgently." (p92 emphasis added) 33) The Report concludes that "South Africa has one of the fastest growing HIV/AIDS epidemics in the world, and it is important that every individual of reproductive age should know his/her HIV status. At the very least voluntary counselling and testing should be easily accessible. Guidelines for managing HIV positive women and women with AIDS during pregnancy and the puerperium are urgently required. Ethical guidelines must be developed and made available to health workers especially in the area of maternal and child care. There can be no success in combating the HIV/AIDS epidemic without a concerted effort at curbing HIV transmission. South Africa must apply the lessons learned in other areas, especially Uganda and Thailand, to stop new transmission. Young people must be the main beneficiaries of prevention programmes." (p95 emphasis added) Mother-to-child-transmission 34) The issue of addressing women's vulnerability has to be a central part of the country's response to the AIDS epidemic. 35) South Africa is experiencing an explosive epidemic. The complexities of the epidemic within and between provinces remain to be elucidated. There still remains a huge primary prevention potential. 36) As sexual transmission is the predominant mode of spread of HIV, there are major implications for transmission from infected parents to infants. The alarmingly high incidence rates and prevalence amongst women in the reproductive age group highlight this group as an important one for targeted prevention strategies that will enable those who are uninfected to maintain their status, and those already infected to prevent secondary transmission including to their unborn infants. 37) In the past 6-8 years there has been an increasing amount of empirical data that demonstrate the efficacy and effectiveness of antiretrovirals administered to the mother during pregnancy and labour and delivery and/or to the new-born infant, in reducing HIV transmission from 50% - 67%. The HIVNET 012 (conducted in Uganda), SAINT (conducted in KwaZulu-Natal) and numerous other study results demonstrate a 50% reduction in HIV transmission from infected mother to infant through the administration of one dose of Nevirapine to the mother at the onset or during labour and one dose to the infant within 72 hours of delivery. 38) This is a very feasible option for South Africa to introduce in the public sector as the majority of the deliveries take place at health facilities under supervision, and in addition most infants return to the health facility within 72 hours of delivery for postnatal assessment. Brazil has demonstrated the feasibility of such programmes by reducing its mother to child HIV transmission rate from 30% to less than 2%. 39) To augment and support a programmatic intervention in the South African public sector, rapid HIV tests are available that are stable, reliable, and accurate and can be conducted without elaborate laboratory facilities. Boehringer-Ingelheim has already made an offer to provide the drugs at no cost to the government. Surveys of health care workers and users demonstrate high levels of acceptability of the intervention. Health care worker training manuals are already available. 40) Post-partum interventions that ensure safety of continued breastfeeding are currently under investigation, increasing the feeding choices of women who are already HIV infected. Data already presented demonstrate the number of infant deaths that can be averted. The cost-effectiveness in terms both of lives saved and reduction in health service utilization has already been described. 41) We are in South Africa currently experiencing a maturing epidemic as evidenced by the increasing morbidity and mortality. The virus is spreading disproportionately among young people in their prime of their lives, in the economically active age group. These trends are already impacting on life expectancy, and it will take many years to reverse the negative human development impacts that this epidemic is having on our society. It will take many generations to recover from this devastation. 42) A vaccine, the best long-term solution, is not going to be available for at least a decade given the most optimistic scenario. 43) In contrast, we have seen in North America, Europe and Brazil how Highly Active Antiretroviral Therapy has transformed HIV/AIDS to a chronic manageable condition. Substantial reductions in drug prices, increasing availability of generic drugs and single daily doses of anti-retrovirals create an environment where even resource constrained settings (which South Africa is not) can now benefit from these therapeutic advances. Innovative strategies that involve partnerships between public and private sectors and civil society can ensure that families, communities and society can benefit from these scientific advances. 44) At a programmatic level equity is an ideal for which we all strive. What we have learnt in South Africa and other parts of the world is that incremental introduction of new interventions with careful monitoring and evaluation in the long-term ensures we get closer to equity. An essential precursor to such a phased-in/incremental approach, especially where the evidence of efficacy and effectiveness is so overwhelming, is the introduction of policy to that effect. 45) Pilot/phased-in and incremental approaches should not preclude access to the intervention where there is a need and where competent and skilled staff exist to deliver such an intervention. 46) The HIV/AIDS landscape is very fluid in terms of transmission dynamics and in terms of new advances. It demands flexibility, boldness, rapidity and innovation in our responses. The ramifications of an unchecked epidemic, already horrific, are simply untenable in a relatively wealthy young democracy such as ours. 47) A programme to prevent mother-to-child HIV transmission will not only reduce new paediatric infections. It will also reduce the maternal mortality rate and improve the quality of life for many women of reproductive age who have HIV/AIDS. ______________________________ QUARRAISHA ABDOOL KARIM PhD Director: Southern African Fogarty AIDS Training Programme Associate Professor in Community Health, Nelson R Mandela School of Medicine, University of Natal Adjunct Associate Professor in Epidemiology, Columbia University, New York. I CERTIFY THAT THE DEPONENT HAS ACKNOWLEDGED THAT HE KNOWS AND UNDERSTANDS THE CONTENTS OF THIS AFFIDAVIT WHICH WAS SIGNED AND SWORN TO BEFORE ME AT CAPE TOWN ON THIS ----- DAY OF AGUST 2001 AND THAT HE HAS NO OBJECTION TO TAKING THE PRESCRIBED OATH AND CONSIDERS SAME TO BE BINDING ON HIS CONSCIENCE. ______________________ COMMISSIONER OF OATHS References Abdool Karim Q, Abdool Karim SS. South Africa: host to new and emerging epidemics. [Editorial]. Sexually Transm Infec 1999; 75: 139-147. Abdool Karim Q, Abdool Karim SS, Singh B, Short R, Ngxongo S. HIV infection in rural South Africa. AIDS 1992; 6:1535-9 Abdool Karim Q. Women and AIDS in KwaZulu-Natal: Epidemiology and Gender Barriers to HIV Prevention. PhD dissertation, University of Natal 2000. Abdool Karim SS, Abdool Karim Q. Changes in HIV seroprevalence in a rural black community in KwaZulu. S Afr Med J 1992; 82:484. Anderson RM, Garnett GP. Mathematical models of the transmission and control of sexually transmitted diseases. Sexually Transmitted Diseases 2000; 27: 636-643. Bobat RA, Coovadia HM, Windsor IM. Some early observations on HIV infection in children in King Edward VIII Hospital, Durban. S Afr Med J 1990; 78: 524-527. Colvin M, Dawood S, Kleinschmidt, Mullick S, Lallo U. Prevalence of HIV and HIV-related disease in the adult medical wards of a tertiary hospital in Durban, South Africa. Int J STD AIDS 2001; 12: 386-389. Colvin M, Abdool Karim SS, Connolly C, Hoosen AA, Ntuli N. HIV infection and asymptomatic sexually transmitted infections in a rural South African community. Int J STD & AIDS 1998; 9:548-550. Coleman RL, Wilkinson D. Increasing HIV prevalence in a rural district of South Africa. J Acquir Immun Def Syndr Reterovirol 1997; 16: 50-53. Crookes RL, Heyns AP. HIV seroprevalence – data derived from blood transfusion services. S Afr Med J 1992; 82:484-485. Department of Health, RSA. Eleventh national HIV survey of women attending antenatal clinics of the public health services. Pretoria, 2001. Doyle PR. The impact of AIDS on the South African population. Part 1. In AIDS in South Africa: the Demographic and Economic Implications. Paper no 23. Center for Health Policy, department of Community Health, University of the Witwatersrand, Johannesburg, 1991. Gouws E, Williams B, Sheppard HW, Enge B, Abdool Karim SS. High incidence of HIV in South Africa using Sensitive/less-sensitive ELISA (submitted) Schall R. On the maximum size of the AIDS epidemic among the heterosexual black population in South Africa. S Afr Med J 1990; 78: 507-510. Sher R. Acquired immune deficiency syndrome (AIDS) in the RSA. S Afr Med J 1986; 70:23-36. Whelan D. Gender and HIV/AIDS: Taking Stock of Research and Programmes. UNAIDS March 1999 Whiteside A, Wilkins N, Mason B, Wood G. The Impact of HIV/AIDS on Planning Issues in KwaZulu/Natal. Economic Research Unit, University of Natal, Durban, 1990. Wilkinson D, Conolly C, Rotchford K. Continued explosive rise in HIV prevalence among pregnant women in rural South Africa (letter). AIDS 1999a; 13: 740. Wilkinson D, Ndovela N, Harrison A, Lurie M, Connolly C, Sturm AW. Family Planning Services in developing countries: opportunities to treat asymptomatic and unrecognized genital tract infection. Genitourin Med 1997; 73: 58-560. Wilkinson D, Wilkinson N. HIV infection among patients with sexually transmitted diseases in rural South Africa. Int J STD AIDS 1998; 9:736-739. Wilkinson D, Davies GR. The increasing burden of tuberculosis in rural South Africa – impact of the HIV epidemic. S Afr Med J 1997; 87:447-450. Wilkinson D. TB research in South Africa. S Afr Med J 1999b; 89: 155-159. Williams B, MacPhail C, Campbell C, Taljaard D, Gouws E, Moema S, Mzaidume Z, Rasego B. The Carletonville-Mothusimpilo Project: limiting transmission of HIV through community-based interventions. S Afr J Sc 2000a; 96: 351-359. Williams B, Gouws E, Wilkinson D, Abdool Karim SS. Estimating HIV incidence rates from age prevalence data in epidemic situation. Stat Med 2001: 20(13): 2003-2016. Williams BG, Gouws E, Abdool Karim SS. Where are we now? Where are we going? The demographic impact of HIV/AIDS in South Africa. S Afr J Sc 2000b;96:297-304. 1