Biomedical prevention: rhetoric and realityadmin
Biomedical prevention: rhetoric and reality
HIV Australia | Vol. 13 No. 2 | July 2015
SUSAN KIPPAX unpacks the real world implications of treatment as prevention research, noting an important distinction between ‘efficacy’ and ‘effectiveness’.
Over the last few years, a number of what are typically termed ‘biomedical’ HIV prevention technologies have been trialled and many found to be efficacious. These include ‘treatment as prevention’, pre-exposure prophylaxis (PrEP) and microbicides: all of which are based on antiretroviral therapeutic drugs used in the treatment of HIV.
They are termed ‘biomedical’ prevention to distinguish them from so-called ‘behavioural’ prevention, such as condom use or reduction in number of partners. Although, as most if not all prevention involves changes in behaviours or social practices, the distinction is not a very useful one.
This article focuses on one of these ‘biomedical’ preventions – ‘treatment as prevention’ or TasP – and its efficacy and its effectiveness.
The prevention method, TasP, is efficacious in as much as antiretroviral therapy (ART) lowers the viral load of people with HIV, thereby reducing the risk of transmission to sexual (or drug injection) partners.
Whether TasP is effective, as distinct from efficacious, is more complex and depends on a number of factors as described below. Efficacy and effectiveness are defined as follows:
Efficacy is defined as the ‘improvement in health outcome achieved in individuals in a research setting, in expert hands, under idealcircumstances … ”.1
Effectiveness on the other hand is defined as: ‘ … the impact an intervention achieves in the real world, under resource constraints, in entire populations, or in specified subgroups of a population. It is the improvement in health outcome … ’2 and includes long term and far-reaching population effects.
Effectiveness is dependent not only on the efficacy of the prevention technology but also on the responses of people.
Effectiveness, which is the outcome of sustained adoption of efficacious technologies by populations, is a social, political, and economic matter.3
Efficacy of TasP
There is little doubt of the efficacy of TasP – at least under certain conditions.
Following on from the ground-breaking analysis of Vernazza, et al. (2008),4 Cohen, et al. (2011)5 in their HPTN 052 study – a randomised controlled trial – demonstrated that for stable cohabiting couples (mostly heterosexual) in serodiscordant relationships, the efficacy of TasP was 96%.
ART does indeed result in lowered transmission risk for the HIV-negative individual members of cohabiting discordant couples.
However, it is important to note that people who acquired HIV during the HPTN 052 study from someone other than their cohabiting regular partner were excluded from the calculation of efficacy.
In total, 38 people seroconverted during the trial: 28 of these acquired HIV from their cohabiting partners (with one in the immediate ART group and 27 in delayed ART group), and 10 became HIV-positive as a result of sexual activity outside of their primary relationship.6
Therefore, the HPTN 052 trial demonstrated efficacy within the context of stable, monogamous, cohabiting relationships.
The findings of the HPTN 052 trial taken together with earlier mathematical modelling7 were hailed by many, including Hilary Clinton at the 2012 International AIDS Conference in Washington, as heralding ‘the end of the epidemic’.
In other words, the claim being made by many was (and is) that if a large proportion of people living with HIV were on treatment then HIV acquisition would decline and HIV would eventually disappear.
There was an immediate call to roll-out TasP, with The Lanceteditorial (May 21, 2011) endorsing TasP as a population strategy.8
Indeed, the editorial in The Lancet went so far as to state that funding agencies such as President’s Emergency Plan For AIDS Relief and the Global Fund to Fight AIDS, Tuberculosis and Malaria ‘need to reassess their prevention portfolios and consider diverting funds from programmes with poor evidence (such as behavioural change communication) to treatment for prevention’.9
Effectiveness of TasP
What evidence is there for TasP’s impact in the real world in terms of lowering HIV incidence, under resource constraints, in entire populations, or in specified subgroups of a population?
In some places and some contexts, the results demonstrate an impact – in other places and contexts not. Some studies have shown a decline in HIV incidence over time in association with widespread uptake of ART.
These include: a study in British Columbia, Canada among people who inject drugs, which demonstrated a decline in HIV acquisition;10 a study among homosexually active men in San Francisco;11 and two more recent studies focused on cohabiting couples – one in KwaZulu-Natal12 in a very large population-based prospective cohort study, and a study in China.13
However, there has been some caution expressed about interpreting the results of the Canadian and San Franciscan studies as evidence for the success of TasP.
The timing of the 2010 Canadian study is pertinent for interpreting their results: there had recently been a large intervention among people who inject drugs that some believe may account for the decline in HIV-transmission. More importantly, no risk compensation is likely among people who inject drugs.
Unlike condoms, which are not necessary for sexual engagement, not only are needles and syringes required to inject, it is in people’s interest to use clean needles: fewer abscesses, less bruising, better injecting experience, less risk of hepatitis C transmission, etc.
With regard to the 2010 San Francisco study, Garnett, et al. note that the ‘positive result’ may be due to other factors and that one would ‘expect a delay between incident infection and an infectious case being diagnosed and reported’.14
The two more recent studies, cited above, conducted in China15and in Hlabisa, KwaZulu-Natal,16 demonstrated a significant and strong relationship between antiretroviral and reduction in HIV acquisition.
However, it is important to note that although Jia et al.’s 2013 study confirms the effectiveness of TasP with reference to serodiscordant couples (their study based on 38,862 serodiscordant couples in China showed a 26% reduction in HIV-transmission under real world conditions), they note that ‘protection was only significant in the first year’.17
They are unsure of the reasons for this and comment that the long-term durability of the protectiveness of treatment needs to be confirmed in additional studies.
In Hlabisa, which comprises around 60,000 people who were participants in the open population cohort, about 80% consented to be tested for HIV. Antiretroviral coverage rose from 10% to 30–40% of all HIV-infected individuals over a six-year period.18
Given this proportion of people on antiretroviral therapy is not high when compared with that in Australia and many countries in Western Europe, why the strong and very positive finding?
Barnighausen19 has suggested that in settings such as rural KwaZulu-Natal where increased access to therapy has transformed the community life with many people living with HIV being able to go back to work and lead normal lives, renewed hope has meant that they take more care now regarding HIV transmission.
Indeed, Tanser et al. reported a significant increase in condom use among regular partners in the years between 2005 and 2011, which was independent of the strong relationship between antiretroviral coverage and the lowered risk of HIV acquisition.20
This study clearly demonstrates evidence for the effectiveness of TasP in this setting as well as the absence of any behavioural disinhibition or risk compensation or alternatively the presence of what might be called ‘risk reduction enhancement’.
It is possible, as Wilson21 has suggested, that ART may have less impact in reducing new HIV infections in countries or among populations where very high levels of ART coverage have been reached, as in Europe and Australia – a sort of ceiling effect.
In contrast to the above, in many communities and regions as the number of people on antiretroviral therapy has increased, there has been no concomitant decline in HIV acquisition, for example, in gay male populations in Australia,22 New Zealand,23 in France24 and Switzerland25 and the UK and North America.26 27
In these countries, where there are very high rates of antiretroviral uptake and adherence and where population viral load has been shown to have declined, there has been no decline in HIV acquisition.
Indeed in many of these countries, there have been increases in HIV acquisition, and unlike in Hlabisa, a decline in condom use.
So while the modelling of Granich, et al. (2009)28 indicates that TasP could be effective and could under certain model assumptions radically reduce HIV transmission at the population level, other models – using different assumptions – do not (for example, Wilson, et al., 2008;29 Mei, et al., 2011;30 Garnett, et al., 2012;31Kretzschmar, et al., 201232).
As Garnett, et al. discuss, modelling shows that the extent of a reduction in HIV transmission depends primarily on whether optimistic or pessimistic assumptions are made about the programmatic use of antiretrovirals and state that ‘[o]nly the most extremely optimistic scenarios predict that treatment alone can halt the HIV pandemic, and even these assume that treatment enables reductions in sexual risk behaviour’.33
Wilson sums up saying that the examination of data from TasP ‘natural experiments’ suggests there are limitations to reductions in population HIV incidence.37
The limitations include the following: risk compensation; difficulties in linking people living with HIV to treatment and retaining them in clinical care; the increasing pool of potential transmitters produced by successful antiretroviral therapy; and high rates of frequent testing required by those undiagnosed.
The interplay between serostatus discordance, risk behaviours and detectability of viral load differs depending on the social and political contexts.
In Sydney, Australia, a study has demonstrated that despite around 70% of HIV-infected men being on antiretroviral treatment and a high proportion of these men having undetectable viral load, the per-contact probability of HIV transmission due to unprotected anal intercourse is similar to estimates reported from income rich world in the preantiretroviral treatment era.38
In Australia, there is evidence from behavioural surveillance that unprotected anal intercourse with casual sexual partners is increasing. Zablotska, et al., 2009;39 Bavinton, et al., 201340 have also shown risk compensation among gay men: with gay men with undetectable viral load (and their partners) more likely to engage in unprotected anal intercourse than those with detectable viral load.
To these considerations of the reach and limits of TasP, we would add that if TasP is to involve anchoring HIV prevention efforts in the clinic there is a real risk that it will function to further privatise HIV, making HIV (whether we are talking about treatment or prevention) a matter for individuals or couples, and not for broad public discussion, debate and action.
If the early successful response in Australia taught us anything, it is that a public involvement and a collective response are central to reducing HIV incidence: a solely clinic-based response can function to position people as patients or information recipients rather than engage them in collective attempts to devise and develop apt responses to transmission.
There are few in biomedicine or indeed in social science who doubt the efficacy of TasP.
For many in serodiscordant relationships, TasP provides a genuine alternative to condom use – especially for cohabiting couples where there is the opportunity for a frank exchange of information and negotiation of risk.
However, it is clear that TasP is not always a successful population strategy: TasP is effective at the population level only under certain conditions and only in certain contexts – not exactly a ‘magic bullet’.
Cohen, et al., (2012) acknowledge the differences and note that: ‘Implementation of ART as prevention faces substantial challenges, including logistical limitations, potential challenges in risk-taking behaviors, and cost’.41
However, they seem to think that there is a ‘best strategy’ for the use of antiretrovirals, which has not yet been developed.42
Social scientists argue that there is unlikely to be a ‘best strategy’ because the reasons for the differences between the studies lie in the social and cultural differences in the populations under study and their differing responses.
The effectiveness of any HIV prevention strategy is the contingentoutcome of the collective activity of a diverse range of actors both human and non-human, including the mode of prevention being adopted, scientific practices, clinical services, cultural, political and social environments, and the norms, values, and discourses that animate human behaviour or practice.43
There is no ‘best strategy’: any particular prevention strategy is likely to change over time, particularly as there are new developments in the field of HIV and in communities’ understandings of and access to treatment and different means of prevention.
3 Kippax, S. (2003). Sexual health interventions are unsuitable for experimental evaluation. In Stephenson, J., Bonnell, C., Imrie, J. (eds.), Effective sexual health interventions. Oxford University Press, Oxford, 17–34.
4 Vernazza, P., Hirschel, B., Bernasconi, E. (2008). Les personnes seropositives suivant un TAR efficace ne transmettant pas le VIH pat voie sexuelle. Bulletin de Medicins Suisses, 89, 5.
5 Cohen, M., Chen, Y., McCauley, M., Gamble, T., Hosseinipour, M., Kumarasamy, N., et. al. (2011). Prevention of HIV-1 infection with early antiretroviral therapy. New England Journal of Medicine, 365, 493–505. doi: dx.doi.org/10.1056/NEJMoa1105243
7 Granich, R., Gilks, C., Dye, C., De Cock, K., Williams, B. (2009). Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model. The Lancet, 373, (9657), 48–57. doi: dx.doi.org/10.1016/S0140-6736(08)61697-9
9 ibid. 1719.
10 Montaner, J., Lima, V., Barrios, R., Yip, B., Wood, E., Kerr, T., et al. (2010). Association of highly active antiretroviral therapy coverage, population viral load, and yearly new HIV diagnoses in British Columbia, Canada: a population-based study. The Lancet, 376, (9740), 532–539. doi: dx.doi.org/10.1016/S0140-6736(10)60936-1
11 Das, M., Chy, P., Santos, G., Scheer, S., Vintinghoff, E., McFarland, W., et al. (2010). Decreases in community viral load are accompanied by reductions in new HIV infections in San Francisco. PLOS ONE, 5(6). doi: dx.doi.org/10.1371/journal.pone.0011068
12 Tanser, F., Barnighausen, Y., Grapsa, E., Zaidi, J., Newell, M. (2013). High coverage of ART associated with decline in risk of HIV acquisition in rural KwaZulu-Natal, South Africa. Science, 329 (6122), 966–971. doi: dx.doi.org/10.1126/science.1228160
13 Jia, Z., Ruan, Y., Li, Q., Xie, P., Li, P., Wang, X., et al. (2013). Antiretroviral therapy to prevent HIV transmission in serodiscordant couples in China (2003–11): a national observational cohort study. The Lancet, 382, (9899), 1195–1203. doi: dx.doi.org/10.1016/S0140-6736(12)61898-4
14 Garnett, G., Becker, S., Bertozzi, S. (2012). Treatment as prevention: translating efficacy trials to population effectiveness. Current Opinion in HIV & AIDS, 7, 157–163. 160
15 Jia, Z., et al. (2013). op. cit.
16 Tanser, F., et al. (2013). op. cit.
17 Jia, Z., et al. (2013), op. cit. 6
18 Tanser, F., et al. (2013). op. cit.
19 Personal communication, Till Bärnighausen, Associate Professor of Global Health, Harvard School of Public Health, 2013.
20 Tanser, F., et al. (2013). op. cit.
21 Wilson, D. (2012). HIV treatment as prevention: natural experiments highlight limits of antiretroviral treatment as HIV prevention. PLOS Medicine, 9(7), e1001231. doi: dx.doi.org/10.1371/journal.pmed.1001231
23 Saxton, P., Dickson, N., McAllister, S., Sharples, K., Hughes, A. (2011). Increase in HIV diagnoses among men who have sex with men in New Zealand from a stable low period. Sexual Health, 8, 311–318. doi: dx.doi.org/10.1071/SH10087
24 As reported in Wilson, D. (2012). op cit.
25 van Sighem, A., Vidondo, B., Glass, T., Bucher, H., Vernazza, P., Gebhardt, M., et al. (2012). Resurgence of HIV infection among men who have sex with men in Switzerland: mathematical modeling study. PLOS ONE, 7(9), e44819.
26 van Griensven, F., de Lind van Wijngaarden, J., Baral, S., Grulich, A. (2009). The global epidemic of HIV infection among men who have sex with men. Current Opinion in HIV and AIDS, 4(4), 300–307. doi: dx.doi.org/10.1097/COH.0b013e32832c3bb3
27 Sullivan, P., Hamouda, O., Delpech, V., Geduld, J., Prejean, J., Semaille, C., et al. (2009). Reemergence of the HIV epidemic among men who have sex with men in North America, Western Europe, and Australia, 1996–2005. Annals of Epidemiology, 19(6), 423–431. doi: dx.doi.org/10.1016/j.annepidem.2009.03.004
28 Granich, R., et al. (2009). op. cit.
29 Wilson, D., Law, M., Grulich, A., Cooper, D., Kaldor, J. (2008). Relation between HIV viral load and infectiousness: a model-based analysis. The Lancet, 372(9635), 314–320. doi: dx.doi.org/10.1016/S0140-6736(08)61115-0
30 Mei, S., Quax, R., Van de Vijver, D., Zhu, Y., Sloot, P. (2011). Increasing risk behaviour can outweigh the benefits of antiretroviral drug treatment on the HIV incidence among men-having-sex-with-men in Amsterdam. BMC Infectious Diseases, 11, 118. doi: dx.doi.org/10.1186/1471-2334-11-118
31 Garnett, G., et al. (2012). op cit.
32 Kretzschmar, M., Schim van der Loeff, M., Coutinho, R. (2012). Elimination of HIV by test and treat: a phantom of wishful thinking? AIDS, 26(2), 247–8 [letter]. doi: dx.doi.org/10.1097/QAD.0b013e32834e1592
33 Garnett, G., et al. (2012). op cit. 162.
34 Mei, S., et al. (2011). op. cit.
35 Holtgrave, D., Maulsby, C., Wehrmeyer, L., Hall, H. (2012). Behavioural factors in assessing impact of HIV treatment as prevention. AIDS and Behavior, 16(5),1085– 1091. doi: dx.doi.org/10.1007/s10461-012-0186-1
36 Wilson, D. (2012). op. cit.
38 Jin, F., Jansson. J., Law, M., Prestage. G., Zablotska, I., Imrie, J., et al. (2010). Per-contact probability of HIV transmission in homosexual men in Sydney in the era of HAART. AIDS, 24(6), 907–913. doi: doi.dx.org/10.1097/QAD.0b013e3283372d90
39 Zablotska, I., Crawford, J., Imrie, J., Prestage, G., Jin, F., Grulich A., et al. (2009). Increases in unprotected anal intercourse with sero-discordant casual partners among HIV-negative gay men in Sydney. AIDS and Behavior, 13(4), 638–644. doi: dx.doi.org/10.1007/s10461-008-9506-x
40 Bavinton, B. (2013). The Opposites Attract study: HIV treatment as prevention among gay male serodiscordant relationships, paper presented at The Inaugural Kirby Institute Symposium 2013, UNSW Australia, Sydney, 27 June.
41 Cohen, M., Muessig, K., Kumi Smith, M., Powers, K., Kashuba, A. (2012). Antiviral agents and HIV prevention: controversies, conflicts and consensus. Editorial Review, AIDS, 26, 1585–1598.
42 ibid. 1595.
Professor Susan Kippax is a social researcher of international standing with over twenty years’ experience.
She has an extensive track record in sexuality and illicit drug use research, and has managed numerous programs in the social aspects of the prevention and care of HIV, hepatitis C, and sexually transmissible infections (STIs).
This includes serving as Director of the National Centre in HIV Social Research (NCHSR) from 1994–2008.
Through her research, teaching and policy advisory roles, Professor Kippax has played a central role in the framing of Australia’s response to these blood borne viruses and infections.