Four steps necessary: the conditions for making treatment-as-prevention work

For antiretrovirals to produce the hoped-for effect in preventing HIV on a population level, four processes have to follow on sequentially from each other, and in each case these processes need to achieve the best possible results.

They are:

1. Testing

A high enough proportion of the population, and especially of high-risk populations, need to be tested to diagnose the maximum number of previously undiagnosed people. Testing also needs to be frequent enough in populations with very high HIV incidence for recent infections to be diagnosed before the infected person has a chance in infect someone else.

For instance, a study by the US Centers for Disease Control1 found that 44% of gay men with HIV in 21 cities were unaware of their infection, despite the fact that the majority of men surveyed (58%) had tested within the past year. The study found that HIV prevalence in those who had tested within the year was less than half (13%) of those whose last test was more than a year ago (32%). This may show that frequent testing does help to protect against infection, or it may mean that those with health-seeking behaviours are less likely to become infected.

On the other hand, of those who were unaware of their infection, more had actually tested within the last year (51%) than had last tested more than a year ago (31%). This may indicate that amongst the highest-risk men, HIV incidence is so high that men are getting infected before the behavioural and medical benefits of knowing their status can reduce their risk.

2. Linkage to care

Medical systems have to be in place that ensure that the lowest possible number of people drop out of care after receiving a positive test result (or fail to collect their result altogether). This includes measures such as point-of-care, while-you-wait testing, the integration of testing and treatment facilities, and in many countries abolishing complex or burdensome financial requirements, such as co-payments and having sufficient health insurance.

For instance, in the study above, while HIV prevalence was no higher in those who had no health insurance or had not visited a healthcare provider in the last year, having undiagnosed HIV was much more common in those who did not have health insurance (57% vs. 37%), and vastly more common in those who had not visited a healthcare provider in the last year (81% vs. 37%).

In the British Columbia study referred to above,2 a second increase in the proportion of people with HIV on antiretrovirals – and an associated decrease in diagnoses – was only achieved by actively seeking out, testing and treating injecting drug users.

3. Proportion on therapy

A high proportion of HIV patients have to be on antiretroviral therapy if treatment is to have an effect on onward transmission. Drug access and cost will clearly influence how many patients are on therapy in many countries. In well-resourced countries, guidelines on starting therapy are likely to have the most important influence; raising the threshold CD4 count from 200 to 350 cells/mm³ or more will increase the proportion of patients, out of the total number with HIV, who have undetectable viral loads.

This crucially depends on how many people are already diagnosed, and at what stage of disease. In the UK the proportion of heterosexual men who are diagnosed with CD4 counts over 350 cells/mm³ (the current BHIVA threshold for starting treatment) is only 33% and, therefore, in the absence of considerably expanded and accessed testing, even putting every diagnosed heterosexual man on ARVs would only expand the number by one-third. In the case of gay men, however, 57% are diagnosed with a CD4 count of over 350 cells/mm³, so a treatment-as-prevention programme that resulted in nearly every diagnosed gay man taking ARVs would have to more than double the number on therapy.3

Maximising the proportion on therapy may also be a matter of intensifying the follow-up of existing patients: in the British Columbia study, part of the increase in the proportion of people on ARVs was also achieved by re-contacting people who were on treatment interruptions and persuading them to restart therapy.

4. Proportion failing therapy

Finally, even if the majority of patients are on therapy, the effect of treatment on prevention will clearly be compromised if there is a high level of therapeutic failure.4

For instance, one study of serodiscordant married couples in Henan Province, China, produced a result that was out of line with other studies that have shown reduced infection rates where the spouse was taking ARVs. In this study, the infection rate in the HIV-negative partner was 4.8% a year when the spouse was on therapy and 3.2% when they were not: this was not a significant difference.

However, the authors also noted that a previous study of drug resistance in Henan found that only one-third of people with HIV were adherent to their treatment after six months of therapy, and that by this time, from a baseline of 14% with drug-resistant HIV, no less than 63% had drug resistance.5 Suboptimal treatment regimens and levels of support, in short, produce prevention failure as well as treatment failure.

Summing up

So for treatment-as-prevention to work, there need to be unprecedentedly high rates of success in all steps from increasing HIV testing to increasing the proportion of people on ARVs who have an undetectable viral load.

To give the San Francisco study above as an example:6

  • 85.5% of gay men with HIV were diagnosed.

  • Of these, 80% were linked to care.

  • Of these, 90% (72% of those diagnosed) were taking antiretrovirals.

  • Of these, 75% had an undetectable viral load.

This means that roughly 46% of gay men with HIV in San Francisco, diagnosed and otherwise, were on treatment with an undetectable viral load.

Furthermore, only gay men were included in the study, which means that at least 10% of people with HIV in San Francisco were omitted. This 10% are likely to test less frequently and may have lower rates of diagnosis. The 46% are therefore likely to be a maximum.

A similar calculation for the UK yields a slightly higher figure because while the UK has lower rates of diagnosis than San Francisco, it has higher rates of people taking antiretrovirals and of viral suppression:

  • As of the end of 2008, an estimated 83,000 people in the UK were living with HIV.3
  • Of these, 65,319 (78.7%) were diagnosed.
  • Of these, 50,292 (77%) were seen for care in 2009.7
  • Of these, at least 83% (41,742 – 2007 figures) had a viral load under 50 copies/ml.8

Thus, in 2009, just over 50% of the people with HIV in the UK were on treatment and had an undetectable viral load.

While the above calculations suggest that several factors need to be optimised for a treatment-as-prevention programme to work well, there are a number of different strategies possible to make a large and immediate improvement. Which one should be chosen may depend on which ‘link in the chain’ is weakest. These strategies include, or may be a mix of:

  • HIV testing recommended for all patients in all healthcare settings (a strategy recommended by the Centers for Disease Control in the US, though not as yet enacted).9 This could be combined with treatment offered at any CD4 count, as recommended by half the panel responsible for the most recent US HIV treatment guidelines.10
  • Increased testing with linkage to care and treatment based on clinical need (the strategy recommended by BHIVA and BASHH in their testing guidelines).11

  • Increased treatment uptake by follow-up of hard-to-reach groups and drop-outs from clinic (e.g. the ‘seek and treat’ strategy adopted in British Columbia).

  • Changing from deferred to early treatment (e.g. the ‘treat everyone diagnosed’ strategy adopted in San Francisco).


  1. Smith A et al. Prevalence and awareness of HIV infection among men who have sex with men — 21 cities, United States, 2008. MMWR Morb Mortal Wkly Rep 59: 1201-1207, 2010
  2. Montaner J et al. Association of expanded HAART coverage with a decrease in new HIV diagnoses, particularly among injection drug users in British Columbia, Canada. 17th Conference on Retroviruses and Opportunistic Infections, San Francisco, abstract 88LB, 2010
  3. Health Protection Agency HIV in the United Kingdom: 2009 Report. See, 2009
  4. Wany L et al. HIV transmission among serodiscordant couples: a retrospective study of former plasma donors in Henan, China. J Acquir Immune Defic Syndr, 55: 232-38, 2010
  5. Li JY et al. Prevalence and evolution of drug resistance HIV-1 variants in Henan, China. Cell Res. 5(11-12):843-9, 2005
  6. Das-Douglas M et al. Decreases in community viral load are associated with a reduction in new HIV diagnoses in San Francisco. 17th Conference on Retroviruses and Opportunistic Infections, San Francisco, abstract 33, 2010
  7. Health Protection Agency SOPHID data. HPA, 2009
  8. Bansi L et al. Trends over calendar time in antiretroviral treatment success and failure in HIV clinic populations. HIV Medicine 11: 432-38, 2010
  9. Branson B et al. Revised Recommendations for HIV Testing of Adults, Adolescents, and Pregnant Women in Health-Care Settings. MMWR 55 (RR14): 1-17, 2006
  10. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services, 1 december: 1-161, available online at [Date accessed 30 March 2011], 2009
  11. BHIVA, BASHH and FSRH Guidelines for the management of the sexual and reproductive health of people living with HIV infection. HIV Medicine 9: 681-720, 2008
This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.
Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.