A Tale Of Two Epidemics Within TWO Countries

Abstract

Pettifor et al's recent study concludes that “the remarkable prevalence of HIV in young people in South Africa cannot be ascribed to exceptional risk taking behavior” [[1]Pettifor A.E. Levandowski B.A. Macphail C. et al.A Tale of two countries: Rethinking sexual risk for HIV among young people in South Africa and the United States.J Adolesc Health. 2011; 49: 237-243Abstract Full Text Full Text PDF PubMed Scopus (29) Google Scholar]. This and the accompanying editorial [[2]Jaspan H. The wrong place at the wrong time: Geographic disparities in young people's HIV risk.J Adolesc Health. 2011; 49: 227-229Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar] argue that behavioral change campaigns have “failed,” and that more emphasis needs to be placed on biological interventions.This conclusion is based on their comparative study of sexual behaviors in representative survey samples of 18–24-year-olds in the United States (Add Health) and South Africa (South African Youth Survey—SAYS). Their study found that all the risk factors evaluated (excluding couple age discordance) to be more prevalent in the United States—the low HIV prevalence area.We believe this conclusion is unwarranted for two reasons. First, their study ignores the literature that explores the reasons for the large differences in HIV prevalence by race within both these countries. Second, they do not evaluate any network-level risk factors. The Add Health Survey revealed that for every category of individual-level risk behavior, non-Hispanic blacks had considerably higher sexually transmitted infection rates (up to 28 times higher) compared with non-Hispanic whites [[3]Hallfors D.D. Iritani B.J. Miller W.C. Bauer D.J. Sexual and drug behavior patterns and HIV and STD racial disparities: The need for new directions.Am J Public Health. 2007; 97: 125-132Crossref PubMed Scopus (302) Google Scholar]. This does not however mean that behavioral factors do not determine these racial differences. It may just be that these differences are due to network-level factors that have not been assessed. Indeed, a further analysis of the Add Health and three other surveys from the United States showed that differences in rates of concurrency and assortative mixing explained a 2.6-fold difference by race in HIV prevalence [[4]Morris M. Kurth A.E. Hamilton D.T. et al.Concurrent partnerships and HIV prevalence disparities by race: linking science and public health practice.Am J Public Health. 2009; 99: 1023-1031Crossref PubMed Scopus (238) Google Scholar].Our group has shown a similar phenomenon in the Cape Area Panel Survey—a representative sample of 14–22-year-olds in Cape Town, South Africa. HIV prevalence in South Africa differs by racial group by over an order of magnitude [[5]Shisana O. Rehle T. Simbayi L. et al.South African National HIV Prevalence, HIV Incidence, Behaviour and Communication Survey. Human Sciences, Research Council, Pretoria, South Africa2005Google Scholar]. We found that individual-level risk factors were unable to explain these large differences in HIV spread, but that network-level factors could [[6]Kenyon C. Dlamini S. Boulle A. et al.A network level explanation for the differences in HIV prevalence in South Africa's racial/ethnic groups.Afr J AIDS Res. 2009; 8: 243-254Crossref PubMed Scopus (35) Google Scholar]. For example, the sexually active white respondents had had the greatest number of lifetime partners, but partner concurrency was more commonly reported by the black respondents—41% of black men compared with 4% of white men stated that they had engaged in a concurrent sexual partnership. The SAYS study found that the biggest “risk factor” for HIV acquisition was race—black women, for example, had an adjusted odds ratio of 8.3 for being HIV-infected compared with white women [[7]Pettifor A.E. Rees H.V. Kleinschmidt I. et al.Young people's sexual health in South Africa: HIV prevalence and sexual behaviors from a nationally representative household survey.AIDS. 2005; 19: 1525-1534Crossref PubMed Scopus (452) Google Scholar]. However, they did not collect data on network parameters such as concurrency. In a separate study using the Cape Area Panel Survey data, we described that sexually transmitted infection symptom prevalence remained similarly associated with race when controlling for individual-level parameters only [[8]Kenyon C. Badri M. The role of concurrent sexual relationships in the spread of Sexually Transmitted Infections in young South Africans.South Afr J HIV Med. 2009; 10: 29-36Google Scholar], but not when additionally adjusting for concurrency.These findings caution against concluding, on the basis of weak or absent associations between standard behavior metrics and HIV prevalence, whether individual or ecological, that nonbehavioral factors must be cofactors in transmission. Many of the biological differences that Pettifor [[1]Pettifor A.E. Levandowski B.A. Macphail C. et al.A Tale of two countries: Rethinking sexual risk for HIV among young people in South Africa and the United States.J Adolesc Health. 2011; 49: 237-243Abstract Full Text Full Text PDF PubMed Scopus (29) Google Scholar] and Jaspen [[2]Jaspan H. The wrong place at the wrong time: Geographic disparities in young people's HIV risk.J Adolesc Health. 2011; 49: 227-229Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar] highlight as potential cofactors of transmission such as viral subtype and circumcision cannot be invoked as explanations for differences in HIV prevalence between demographic groups within country. In Cape Town, for example, all young adults are at threat from the same viral subtypes, and black men are the most likely to be circumcised [[6]Kenyon C. Dlamini S. Boulle A. et al.A network level explanation for the differences in HIV prevalence in South Africa's racial/ethnic groups.Afr J AIDS Res. 2009; 8: 243-254Crossref PubMed Scopus (35) Google Scholar]. We believe that the most plausible reason for the large differences in HIV prevalence between countries and between demographic groups within countries is differences in sexual behavior at the network level, factors which the authors readily identify they were unable to assess in their study.There is a danger, in the excitement of the recent advances in the development of biologically based HIV prevention technologies, that we may lose sight of important behavioral determinants of HIV transmission, which could be highly amenable to intervention if afforded the same investment that is being contemplated for some of these newer technologies. Behavioral campaigns have been responsible for the dramatic declines in HIV incidence in Uganda [[9]Kirby D. Changes in sexual behaviour leading to the confirmation from multiple sources of evidence decline in the prevalence of HIV in Uganda.Sex Transm Infect. 2008; 84: ii35-ii41Crossref PubMed Scopus (80) Google Scholar]; in both Uganda and Zimbabwe, reductions in partner concurrency were instrumental in bringing down HIV incidence [9Kirby D. Changes in sexual behaviour leading to the confirmation from multiple sources of evidence decline in the prevalence of HIV in Uganda.Sex Transm Infect. 2008; 84: ii35-ii41Crossref PubMed Scopus (80) Google Scholar, 10Daniel T. Halperin D.T. Mugurungi O. et al.A surprising prevention success: Why did the HIV epidemic decline in Zimbabwe?.PLoS Med. 2011; 8: e1000414Crossref PubMed Scopus (131) Google Scholar]. Pettifor et al's recent study concludes that “the remarkable prevalence of HIV in young people in South Africa cannot be ascribed to exceptional risk taking behavior” [[1]Pettifor A.E. Levandowski B.A. Macphail C. et al.A Tale of two countries: Rethinking sexual risk for HIV among young people in South Africa and the United States.J Adolesc Health. 2011; 49: 237-243Abstract Full Text Full Text PDF PubMed Scopus (29) Google Scholar]. This and the accompanying editorial [[2]Jaspan H. The wrong place at the wrong time: Geographic disparities in young people's HIV risk.J Adolesc Health. 2011; 49: 227-229Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar] argue that behavioral change campaigns have “failed,” and that more emphasis needs to be placed on biological interventions. This conclusion is based on their comparative study of sexual behaviors in representative survey samples of 18–24-year-olds in the United States (Add Health) and South Africa (South African Youth Survey—SAYS). Their study found that all the risk factors evaluated (excluding couple age discordance) to be more prevalent in the United States—the low HIV prevalence area. We believe this conclusion is unwarranted for two reasons. First, their study ignores the literature that explores the reasons for the large differences in HIV prevalence by race within both these countries. Second, they do not evaluate any network-level risk factors. The Add Health Survey revealed that for every category of individual-level risk behavior, non-Hispanic blacks had considerably higher sexually transmitted infection rates (up to 28 times higher) compared with non-Hispanic whites [[3]Hallfors D.D. Iritani B.J. Miller W.C. Bauer D.J. Sexual and drug behavior patterns and HIV and STD racial disparities: The need for new directions.Am J Public Health. 2007; 97: 125-132Crossref PubMed Scopus (302) Google Scholar]. This does not however mean that behavioral factors do not determine these racial differences. It may just be that these differences are due to network-level factors that have not been assessed. Indeed, a further analysis of the Add Health and three other surveys from the United States showed that differences in rates of concurrency and assortative mixing explained a 2.6-fold difference by race in HIV prevalence [[4]Morris M. Kurth A.E. Hamilton D.T. et al.Concurrent partnerships and HIV prevalence disparities by race: linking science and public health practice.Am J Public Health. 2009; 99: 1023-1031Crossref PubMed Scopus (238) Google Scholar]. Our group has shown a similar phenomenon in the Cape Area Panel Survey—a representative sample of 14–22-year-olds in Cape Town, South Africa. HIV prevalence in South Africa differs by racial group by over an order of magnitude [[5]Shisana O. Rehle T. Simbayi L. et al.South African National HIV Prevalence, HIV Incidence, Behaviour and Communication Survey. Human Sciences, Research Council, Pretoria, South Africa2005Google Scholar]. We found that individual-level risk factors were unable to explain these large differences in HIV spread, but that network-level factors could [[6]Kenyon C. Dlamini S. Boulle A. et al.A network level explanation for the differences in HIV prevalence in South Africa's racial/ethnic groups.Afr J AIDS Res. 2009; 8: 243-254Crossref PubMed Scopus (35) Google Scholar]. For example, the sexually active white respondents had had the greatest number of lifetime partners, but partner concurrency was more commonly reported by the black respondents—41% of black men compared with 4% of white men stated that they had engaged in a concurrent sexual partnership. The SAYS study found that the biggest “risk factor” for HIV acquisition was race—black women, for example, had an adjusted odds ratio of 8.3 for being HIV-infected compared with white women [[7]Pettifor A.E. Rees H.V. Kleinschmidt I. et al.Young people's sexual health in South Africa: HIV prevalence and sexual behaviors from a nationally representative household survey.AIDS. 2005; 19: 1525-1534Crossref PubMed Scopus (452) Google Scholar]. However, they did not collect data on network parameters such as concurrency. In a separate study using the Cape Area Panel Survey data, we described that sexually transmitted infection symptom prevalence remained similarly associated with race when controlling for individual-level parameters only [[8]Kenyon C. Badri M. The role of concurrent sexual relationships in the spread of Sexually Transmitted Infections in young South Africans.South Afr J HIV Med. 2009; 10: 29-36Google Scholar], but not when additionally adjusting for concurrency. These findings caution against concluding, on the basis of weak or absent associations between standard behavior metrics and HIV prevalence, whether individual or ecological, that nonbehavioral factors must be cofactors in transmission. Many of the biological differences that Pettifor [[1]Pettifor A.E. Levandowski B.A. Macphail C. et al.A Tale of two countries: Rethinking sexual risk for HIV among young people in South Africa and the United States.J Adolesc Health. 2011; 49: 237-243Abstract Full Text Full Text PDF PubMed Scopus (29) Google Scholar] and Jaspen [[2]Jaspan H. The wrong place at the wrong time: Geographic disparities in young people's HIV risk.J Adolesc Health. 2011; 49: 227-229Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar] highlight as potential cofactors of transmission such as viral subtype and circumcision cannot be invoked as explanations for differences in HIV prevalence between demographic groups within country. In Cape Town, for example, all young adults are at threat from the same viral subtypes, and black men are the most likely to be circumcised [[6]Kenyon C. Dlamini S. Boulle A. et al.A network level explanation for the differences in HIV prevalence in South Africa's racial/ethnic groups.Afr J AIDS Res. 2009; 8: 243-254Crossref PubMed Scopus (35) Google Scholar]. We believe that the most plausible reason for the large differences in HIV prevalence between countries and between demographic groups within countries is differences in sexual behavior at the network level, factors which the authors readily identify they were unable to assess in their study. There is a danger, in the excitement of the recent advances in the development of biologically based HIV prevention technologies, that we may lose sight of important behavioral determinants of HIV transmission, which could be highly amenable to intervention if afforded the same investment that is being contemplated for some of these newer technologies. Behavioral campaigns have been responsible for the dramatic declines in HIV incidence in Uganda [[9]Kirby D. Changes in sexual behaviour leading to the confirmation from multiple sources of evidence decline in the prevalence of HIV in Uganda.Sex Transm Infect. 2008; 84: ii35-ii41Crossref PubMed Scopus (80) Google Scholar]; in both Uganda and Zimbabwe, reductions in partner concurrency were instrumental in bringing down HIV incidence [9Kirby D. Changes in sexual behaviour leading to the confirmation from multiple sources of evidence decline in the prevalence of HIV in Uganda.Sex Transm Infect. 2008; 84: ii35-ii41Crossref PubMed Scopus (80) Google Scholar, 10Daniel T. Halperin D.T. Mugurungi O. et al.A surprising prevention success: Why did the HIV epidemic decline in Zimbabwe?.PLoS Med. 2011; 8: e1000414Crossref PubMed Scopus (131) Google Scholar]. A Tale of Two Countries: Rethinking Sexual Risk for HIV Among Young People in South Africa and the United StatesJournal of Adolescent HealthVol. 49Issue 3PreviewTo compare the sexual behaviors of young people in South Africa (SA) and the United States (US) with the aim to better understand the potential role of sexual behavior in HIV transmission in these two countries that have strikingly different HIV epidemics. Full-Text PDF The Authors replyJournal of Adolescent HealthVol. 50Issue 2PreviewIn their letter, Potterat et al raise the question of whether the high prevalence of HIV observed among young people in South Africa is the result of parenteral HIV exposure. Although we found that young people in South Africa did not report more sexual risk behaviors than their U.S. peers, current evidence suggests that the vast majority of infections in sub-Saharan Africa, including South Africa, are the result of sexual transmission [1,2]. Several studies have documented that unclean needles contribute only a small proportion of new HIV infections in sub-Saharan Africa [3–7]. Full-Text PDF