- Email: [email protected]
Role of concurrency in generalised HIV epidemics
Role of concurrency in generalised HIV epidemics It is heartening that Frank Tanser and colleagues (July 16, p 247)1 support direct messaging about sexual partner reduction for HIV prevention. However, it is not surprising that they did not see a role for concurrent or overlapping sexual partnerships in their study from South Africa. Their approach is very insensitive: it lacks individual exposure data for women, uses only a single initial point-in-time measurement of men’s behaviour to construct a remote secondary geography-bound aggregate surrogate for women’s risk, and has low response rates. But why rely on such indirect analysis? Evidence from the intensive community cohort study In Rakai, Uganda (webappendix),2 provides direct clear evidence for the ongoing importance of concurrent partners in a mature epidemic. Within couples where both partners’ status was known, roughly five in eight infections occurred when neither partner was previously infected (concordant negative). This occurrence necessarily implicates concurrency, since the newly infected partner must have had an outside partner in addition to the marital partner (barring nonsexual transmission, which is likely to be very uncommon). Additionally, concurrency is important for transmission even within discordant couples—for whom evidence indicates that roughly 25% of infections come from an outside partner.3 This contemporaneous snapshot does not elucidate the earlier role of concurrency in generating these observed infections nor subsequent infections resulting from them. But concurrency is clearly actively involved on an ongoing basis. Ironically the very next Article in the same issue of The Lancet by Kimberly Powers and colleagues4 from Malawi, strongly supports a key role for sexual concurrency. They attribute 38·4% of infections to sex with an acutely www.thelancet.com Vol 378 November 26, 2011
infected individual. For an acutely infected person to infect another, given the short time involved, necessarily implicates substantial sexual partner overlap. Moreover, concurrency’s role is not limited to serial acute infections, since concurrency-generated infections progressing to later stages of infection can then also lead to further transmission. Finally, it does matter whether concurrency is addressed in programmes. It can support rather than dilute plain messaging on partner limitation. Addressing concurrency adds salience both at the individual level and for affecting social norms. Raising awareness about the risk from sexual networks, and the heightened risk when one’s partner (or partners) has other partners, seem to resonate. And helping people to see that having multiple partners puts one’s family at risk builds on strong African social norms of looking after one’s family.5 Is such messaging controversial? I declare that I have no conflicts of interest.
James D Shelton [email protected] US Agency for International Development, Bureau for Global Health, Washington, DC 20523, USA 1
2
3
4
5
Tanser F, Barnighausen T, Hund L, Garnett GP, McGrath N, Newell M. Effect of concurrent sexual partnerships on the rate of new HIV infections in a high-prevalence, rural South African population: a cohort study. Lancet 2011; 378: 247–55. Gray R, Ssempiija V, Shelton JD, et al. The contribution of HIV-discordant relationships to new HIV infections in Rakai, Uganda. AIDS 2011; 25: 863–65. Celum C, Wald A, Lingappa JR, et al. Acyclovir and transmission of HIV-1 from persons infected with HIV-1 and HSV-2. N Engl J Med 2010; 362: 427–39. Powers K, Ghani AZ, Miller WC, et al. The role of acute and early infection in the spread of HIV and implications for transmission prevention strategies in Lilongwe, Malawi: a modelling study. Lancet 2011; 378: 256–68. Swidler A, Watkins SC. Ties of dependence: AIDS and transactional sex in rural Malawi. Stud Fam Plan 2007; 38: 147–62.
Despite its claims, the study by Frank Tanser and colleagues1 is not “a formal test of the concurrency hypothesis”. The analysis does not tell us whether a woman in this community
is more likely to become HIV positive if her partner(s) have concurrent partners, because Tanser and colleagues did not measure this; they measured the future risk a woman faces from the past characteristics of men who live within a few km radius of her. The model assumes that women’s sexual relationships are most likely to be with their immediate neighbours, become increasingly less likely with men living somewhat farther away, and are non-existent with men who live beyond 2·5–3·8 km. These assumptions are neither credible nor supported by the empirical evidence in this study. Furthermore, the estimated coefficient on neighbourhood men’s lifetime partners does not change after controlling for HIV prevalence. This either means that neighbourhood men’s previous behaviour affects a woman’s future incidence even in a neighbourhood with no HIV-infected men, or it means that there is something wrong with the model, the data, or both. The data, like the model, have serious problems, the most important of which is the very low response rates. Tanser and colleagues report that the HIV testing response rate was 63%, but factoring in the low contact rate, only 43% of eligible respondents were tested in the first wave, falling to only 34% by the third wave.2 Refusal to test was linked to higher HIV risk and prevalence.3,4 Non-response was high for the behavioural measures too: 25% of women did not complete the sexual behaviour questions, and 37% of men refused to participate in the 2004 behavioural survey. One does not have to turn to such convoluted, indirect analytical strategies to test the concurrency hypothesis. Direct empirical evidence comes from nine published studies,5–13 spanning 20 years and multiple sites in sub-Saharan Africa, none of which is cited by Tanser and colleagues. In these studies, half to three-quarters of observed HIV incidence in stable couples can be definitively attributed to concurrent sexual partners. It is
Reuters
Correspondence
See Online for webappendix
Submissions should be made via our electronic submission system at http://ees.elsevier.com/ thelancet/
1843
Correspondence
not acceptable to state that “the effect of concurrent partnerships on HIV incidence has not been appropriately tested in a sub-Saharan African setting”.1 The tone of the academic debate about the effect of concurrent partnerships on HIV risk is puzzling. Science and policy would be best served by moving beyond the longstanding partner reduction messages to test specific concurrency reduction interventions the old-fashioned way, with a multisite randomised controlled trial. Such studies are in development, and, contrary to Tanser and colleagues’ assertions, they are quite feasible.
9
We declare that we have no conflicts of interest.
In their population-based cohort study, Frank Tanser and colleagues1 conclude that they find “no evidence to suggest that concurrent partnerships are an important driver of HIV incidence in this typical high-prevalence rural African population”. We believe that this conclusion is unjustified since they have used an inappropriate method to assess this association. Tanser and colleagues assessed whether each woman’s risk of HIV is higher if there is a higher proportion of men within a 3 km radius who have concurrent partners (with those living closer having a bigger effect on the Gaussian weighting of the average community score). However, this method explicitly discards information about sexual networks. Detection of network-level effects depends on measurement of networks. As an example, Cristakis and colleagues2 showed, using a network approach, that weight gain in one person in the Framingham cohort was associated with substantial weight gain in his or her friends, siblings, spouse, and neighbours. If, instead of measuring the effect of weight gain in one’s actual contacts, they had substituted an average weight gain in all people in a weighted 3 km radius of each person, this effect would in all likelihood have been lost. We agree that prevention efforts need to target both concurrency and
*Martina Morris, Helen Epstein [email protected] University of Washington, Seattle, WA 98125, USA (MM); and 424 West 144th Street, New York, NY, USA (HE) 1
2
3
4
5
6
7
8
1844
Tanser F, Barnighausen T, Hund L, Garnett GP, McGrath N, Newell M-L. Effect of concurrent sexual partnerships on rate of new HIV infections in a high-prevalence, rural South African population: a cohort study. Lancet 2011; 378: 247–55. Tanser F, Hosegood V, Barnighausen T, et al. Cohort profile: Africa Centre Demographic Information System (ACDIS) and population-based HIV survey. Int J Epidemiol 2008; 37: 956–62. Nyirenda M, Zaba B, Baernighausen T, Hosegood V, Newell M-L. Adjusting HIV prevalence for survey non-response using mortality rates: an application of the method using surveillance data from rural South Africa. PLoS One 2010; 5: e12370. Welz T, Hosegood V, Jaffar S, Batzing-Feigenbaum J, Herbst K, Newell M-L. Continued very high prevalence of HIV infection in rural KwaZulu-Natal, South Africa: a population-based longitudinal study. AIDS 2007; 21: 1467–72. Serwadda D, Gray RH, Wawer MJ, et al. The social dynamics of HIV transmission as reflected through discordant couples in rural Uganda. AIDS 1995; 9: 745–50. Carpenter LM, Kamali A, Ruberantwari A, Malamba SS, Whitworth JAG. Rates of HIV-1 transmission within marriage in rural Uganda in relation to the HIV sero-status of the partners. AIDS 1999; 13: 1083–89. Hugonnet S, Mosha F, Todd J, et al. Incidence of HIV infection in stable sexual partnerships: a retrospective cohort study of 1802 couples in Mwanza Region, Tanzania. J Acquir Immune Defic Syndr 2002; 30: 73–80. Senkoro KP, Boerma JT, Klokke AH, et al. HIV incidence and HIV-associated mortality in a cohort of factory workers and their spouses in Tanzania, 1991 through 1996. J Acquir Immune Defic Syndr 2000; 23: 194–202.
10
11
12
13
Gray R, Ssempijaa V, Shelton J, Serwadda D, Nalugoda F, Wawer MJ. The contribution of HIV-discordant relationships to new HIV infections. AIDS 2011; 25: 1343–44. Mermin JJ, Musinguzi J, Opio A, et al. Risk factors for recent HIV infection in Uganda. JAMA 2008; 300: 540–49. Celum C, Wald A, Lingappa JR, et al. Acyclovir and transmission of HIV-1 from persons infected with HIV-1 and HSV-2. N Engl J Med 2010; 362: 427–39. Manigart O, Kraft C, Makombe N, et al. Co-and superinfection of partners in a cohort of couples previously infected by genotypically different viruses in Kigali. Retrovirology 2009; 6 (suppl 3): P384. Trask SA, Derdeyn CA, Fideli U, et al. Molecular epidemiology of human immunodeficiency virus type 1 transmission in a heterosexual cohort of discordant couples in Zambia. J Virol 2002; 76: 397–405.
number of sexual partners.3 The most striking sexual behaviour in Tanser and colleagues’ study is, however, not the mean lifetime number of sexual partners. This number is not significantly different from that found in low-HIV-prevalence countries—a finding that has been documented before.4 Rather, and in common with other regions with generalised HIV epidemics,5 it is the extraordinarily high proportion of sexually active men who report having two or more ongoing sexual relationships at the time of the survey (29%). We declare that we have no conflicts of interest.
*Chris Kenyon, Bob Colebunders [email protected] Groote Schuur Hospital, Observatory, 7935 Cape Town, South Africa (CK); and Antwerp Institute of Tropical Medicine, Antwerp, Belgium (BC) 1
2
3 4
5
Tanser F, Barnighausen T, Hund L, Garnett GP, McGrath N, Newell M-L. Effect of concurrent sexual partnerships on rate of new HIV infections in a high-prevalence, rural South African population: a cohort study. Lancet 2011; 378: 247–55. Cristakis N, Fowler J. The spread of obesity in a large social network over 32 Years. N Engl J Med 2007; 357: 370–79. Padian N, Manian S. The concurrency debate: time to put it to rest. Lancet 2011; 378: 203–04. Wellings K, Collumbien M, Slaymaker E, et al. Sexual behaviour in context: a global perspective. Lancet 2006; 368: 1706–28. Morris M, Epstein H, Wawer M. Timing Is everything: international variations in historical sexual partnership concurrency and HIV prevalence. PLoS One 2010; 5: e14092.
Frank Tanser and colleagues1 argue that concurrent partnerships are not an important driver of HIV incidence in KwaZulu Natal, South Africa. They acknowledge that their results suffer from possible attenuation bias because they measure the effects of a community-level exposure (concurrency in men) on an individual disease outcome. However, they do not mention attenuation stemming from measurement error in their main exposure variables, instead assuming that, unlike women, men report their concurrent partnerships accurately. This assumption is wrong. Even if UNAIDS-recommended survey www.thelancet.com Vol 378 November 26, 2011
infected individual. For an acutely infected person to infect another, given the short time involved, necessarily implicates substantial sexual partner overlap. Moreover, concurrency’s role is not limited to serial acute infections, since concurrency-generated infections progressing to later stages of infection can then also lead to further transmission. Finally, it does matter whether concurrency is addressed in programmes. It can support rather than dilute plain messaging on partner limitation. Addressing concurrency adds salience both at the individual level and for affecting social norms. Raising awareness about the risk from sexual networks, and the heightened risk when one’s partner (or partners) has other partners, seem to resonate. And helping people to see that having multiple partners puts one’s family at risk builds on strong African social norms of looking after one’s family.5 Is such messaging controversial? I declare that I have no conflicts of interest.
James D Shelton [email protected] US Agency for International Development, Bureau for Global Health, Washington, DC 20523, USA 1
2
3
4
5
Tanser F, Barnighausen T, Hund L, Garnett GP, McGrath N, Newell M. Effect of concurrent sexual partnerships on the rate of new HIV infections in a high-prevalence, rural South African population: a cohort study. Lancet 2011; 378: 247–55. Gray R, Ssempiija V, Shelton JD, et al. The contribution of HIV-discordant relationships to new HIV infections in Rakai, Uganda. AIDS 2011; 25: 863–65. Celum C, Wald A, Lingappa JR, et al. Acyclovir and transmission of HIV-1 from persons infected with HIV-1 and HSV-2. N Engl J Med 2010; 362: 427–39. Powers K, Ghani AZ, Miller WC, et al. The role of acute and early infection in the spread of HIV and implications for transmission prevention strategies in Lilongwe, Malawi: a modelling study. Lancet 2011; 378: 256–68. Swidler A, Watkins SC. Ties of dependence: AIDS and transactional sex in rural Malawi. Stud Fam Plan 2007; 38: 147–62.
Despite its claims, the study by Frank Tanser and colleagues1 is not “a formal test of the concurrency hypothesis”. The analysis does not tell us whether a woman in this community
is more likely to become HIV positive if her partner(s) have concurrent partners, because Tanser and colleagues did not measure this; they measured the future risk a woman faces from the past characteristics of men who live within a few km radius of her. The model assumes that women’s sexual relationships are most likely to be with their immediate neighbours, become increasingly less likely with men living somewhat farther away, and are non-existent with men who live beyond 2·5–3·8 km. These assumptions are neither credible nor supported by the empirical evidence in this study. Furthermore, the estimated coefficient on neighbourhood men’s lifetime partners does not change after controlling for HIV prevalence. This either means that neighbourhood men’s previous behaviour affects a woman’s future incidence even in a neighbourhood with no HIV-infected men, or it means that there is something wrong with the model, the data, or both. The data, like the model, have serious problems, the most important of which is the very low response rates. Tanser and colleagues report that the HIV testing response rate was 63%, but factoring in the low contact rate, only 43% of eligible respondents were tested in the first wave, falling to only 34% by the third wave.2 Refusal to test was linked to higher HIV risk and prevalence.3,4 Non-response was high for the behavioural measures too: 25% of women did not complete the sexual behaviour questions, and 37% of men refused to participate in the 2004 behavioural survey. One does not have to turn to such convoluted, indirect analytical strategies to test the concurrency hypothesis. Direct empirical evidence comes from nine published studies,5–13 spanning 20 years and multiple sites in sub-Saharan Africa, none of which is cited by Tanser and colleagues. In these studies, half to three-quarters of observed HIV incidence in stable couples can be definitively attributed to concurrent sexual partners. It is
Reuters
Correspondence
See Online for webappendix
Submissions should be made via our electronic submission system at http://ees.elsevier.com/ thelancet/
1843
Correspondence
not acceptable to state that “the effect of concurrent partnerships on HIV incidence has not been appropriately tested in a sub-Saharan African setting”.1 The tone of the academic debate about the effect of concurrent partnerships on HIV risk is puzzling. Science and policy would be best served by moving beyond the longstanding partner reduction messages to test specific concurrency reduction interventions the old-fashioned way, with a multisite randomised controlled trial. Such studies are in development, and, contrary to Tanser and colleagues’ assertions, they are quite feasible.
9
We declare that we have no conflicts of interest.
In their population-based cohort study, Frank Tanser and colleagues1 conclude that they find “no evidence to suggest that concurrent partnerships are an important driver of HIV incidence in this typical high-prevalence rural African population”. We believe that this conclusion is unjustified since they have used an inappropriate method to assess this association. Tanser and colleagues assessed whether each woman’s risk of HIV is higher if there is a higher proportion of men within a 3 km radius who have concurrent partners (with those living closer having a bigger effect on the Gaussian weighting of the average community score). However, this method explicitly discards information about sexual networks. Detection of network-level effects depends on measurement of networks. As an example, Cristakis and colleagues2 showed, using a network approach, that weight gain in one person in the Framingham cohort was associated with substantial weight gain in his or her friends, siblings, spouse, and neighbours. If, instead of measuring the effect of weight gain in one’s actual contacts, they had substituted an average weight gain in all people in a weighted 3 km radius of each person, this effect would in all likelihood have been lost. We agree that prevention efforts need to target both concurrency and
*Martina Morris, Helen Epstein [email protected] University of Washington, Seattle, WA 98125, USA (MM); and 424 West 144th Street, New York, NY, USA (HE) 1
2
3
4
5
6
7
8
1844
Tanser F, Barnighausen T, Hund L, Garnett GP, McGrath N, Newell M-L. Effect of concurrent sexual partnerships on rate of new HIV infections in a high-prevalence, rural South African population: a cohort study. Lancet 2011; 378: 247–55. Tanser F, Hosegood V, Barnighausen T, et al. Cohort profile: Africa Centre Demographic Information System (ACDIS) and population-based HIV survey. Int J Epidemiol 2008; 37: 956–62. Nyirenda M, Zaba B, Baernighausen T, Hosegood V, Newell M-L. Adjusting HIV prevalence for survey non-response using mortality rates: an application of the method using surveillance data from rural South Africa. PLoS One 2010; 5: e12370. Welz T, Hosegood V, Jaffar S, Batzing-Feigenbaum J, Herbst K, Newell M-L. Continued very high prevalence of HIV infection in rural KwaZulu-Natal, South Africa: a population-based longitudinal study. AIDS 2007; 21: 1467–72. Serwadda D, Gray RH, Wawer MJ, et al. The social dynamics of HIV transmission as reflected through discordant couples in rural Uganda. AIDS 1995; 9: 745–50. Carpenter LM, Kamali A, Ruberantwari A, Malamba SS, Whitworth JAG. Rates of HIV-1 transmission within marriage in rural Uganda in relation to the HIV sero-status of the partners. AIDS 1999; 13: 1083–89. Hugonnet S, Mosha F, Todd J, et al. Incidence of HIV infection in stable sexual partnerships: a retrospective cohort study of 1802 couples in Mwanza Region, Tanzania. J Acquir Immune Defic Syndr 2002; 30: 73–80. Senkoro KP, Boerma JT, Klokke AH, et al. HIV incidence and HIV-associated mortality in a cohort of factory workers and their spouses in Tanzania, 1991 through 1996. J Acquir Immune Defic Syndr 2000; 23: 194–202.
10
11
12
13
Gray R, Ssempijaa V, Shelton J, Serwadda D, Nalugoda F, Wawer MJ. The contribution of HIV-discordant relationships to new HIV infections. AIDS 2011; 25: 1343–44. Mermin JJ, Musinguzi J, Opio A, et al. Risk factors for recent HIV infection in Uganda. JAMA 2008; 300: 540–49. Celum C, Wald A, Lingappa JR, et al. Acyclovir and transmission of HIV-1 from persons infected with HIV-1 and HSV-2. N Engl J Med 2010; 362: 427–39. Manigart O, Kraft C, Makombe N, et al. Co-and superinfection of partners in a cohort of couples previously infected by genotypically different viruses in Kigali. Retrovirology 2009; 6 (suppl 3): P384. Trask SA, Derdeyn CA, Fideli U, et al. Molecular epidemiology of human immunodeficiency virus type 1 transmission in a heterosexual cohort of discordant couples in Zambia. J Virol 2002; 76: 397–405.
number of sexual partners.3 The most striking sexual behaviour in Tanser and colleagues’ study is, however, not the mean lifetime number of sexual partners. This number is not significantly different from that found in low-HIV-prevalence countries—a finding that has been documented before.4 Rather, and in common with other regions with generalised HIV epidemics,5 it is the extraordinarily high proportion of sexually active men who report having two or more ongoing sexual relationships at the time of the survey (29%). We declare that we have no conflicts of interest.
*Chris Kenyon, Bob Colebunders [email protected] Groote Schuur Hospital, Observatory, 7935 Cape Town, South Africa (CK); and Antwerp Institute of Tropical Medicine, Antwerp, Belgium (BC) 1
2
3 4
5
Tanser F, Barnighausen T, Hund L, Garnett GP, McGrath N, Newell M-L. Effect of concurrent sexual partnerships on rate of new HIV infections in a high-prevalence, rural South African population: a cohort study. Lancet 2011; 378: 247–55. Cristakis N, Fowler J. The spread of obesity in a large social network over 32 Years. N Engl J Med 2007; 357: 370–79. Padian N, Manian S. The concurrency debate: time to put it to rest. Lancet 2011; 378: 203–04. Wellings K, Collumbien M, Slaymaker E, et al. Sexual behaviour in context: a global perspective. Lancet 2006; 368: 1706–28. Morris M, Epstein H, Wawer M. Timing Is everything: international variations in historical sexual partnership concurrency and HIV prevalence. PLoS One 2010; 5: e14092.
Frank Tanser and colleagues1 argue that concurrent partnerships are not an important driver of HIV incidence in KwaZulu Natal, South Africa. They acknowledge that their results suffer from possible attenuation bias because they measure the effects of a community-level exposure (concurrency in men) on an individual disease outcome. However, they do not mention attenuation stemming from measurement error in their main exposure variables, instead assuming that, unlike women, men report their concurrent partnerships accurately. This assumption is wrong. Even if UNAIDS-recommended survey www.thelancet.com Vol 378 November 26, 2011