How do we treat the world? Experts discuss moving to universal HIV treatment

This article is more than 9 years old. Click here for more recent articles on this topic

What practical steps does the global healthcare community need to take in order to expand HIV treatment so that it can reach everyone who is diagnosed? And how do we expand testing so that as many HIV-positive people as possible are diagnosed, on treatment, and virally suppressed?

These were the themes of the strand of discussion concerning what used to be called 'Treatment as Prevention' at the recent 2015 IAPAC Controlling the HIV Epidemic with Antiretrovirals summit in Paris.  See this report for the strand of discussion about pre-exposure prophylaxis (PrEP).


As several presenters commented, we should be moving beyond talking about 'Treatment as Prevention' as the majority of recently-published HIV guidelines in the world now recommend treatment for everyone on diagnosis. 'Treatment as Prevention' may be seen to imply prioritising treatment of people with HIV with high CD4 counts purely for public health reasons. Now that the START trial has provided evidence that antiretroviral therapy (ART) benefits the health of people even with high CD4 counts, there is no longer an ethical distinction between treating people for medical reasons and treating them for public health reasons. The discussion then becomes how to equitably maximise access to treatment for all.

UNAIDS adopted its 90/90/90 target last year (90% of people with HIV diagnosed, 90% of them linked to care, 90% of them on ART and virally suppressed, i.e. 72.9% of all the HIV-positive people in the world virally suppressed and therefore non-infectious) both in anticipation of a move towards guideline harmonisation around universal treatment, and to urge its practical adoption.


90-90-90 target

A target set by the Joint United Nations Programme on HIV/AIDS (UNAIDS) for 90% of people with HIV to be diagnosed, 90% of diagnosed people to be taking treatment, and 90% of people on treatment to have an undetectable viral load. 

virological suppression

Halting of the function or replication of a virus. In HIV, optimal viral suppression is measured as the reduction of viral load (HIV RNA) to undetectable levels and is the goal of antiretroviral therapy.


Social attitudes that suggest that having a particular illness or being in a particular situation is something to be ashamed of. Stigma can be questioned and challenged.

treatment as prevention (TasP)

A public health strategy involving the prompt provision of antiretroviral treatment in people with HIV in order to reduce their risk of transmitting the virus to others through sex.

retention in care

A patient’s regular and ongoing engagement with medical care at a health care facility. 

At the December launch of its Fast Track initiative, UNAIDS stressed that because people with HIV are living longer, the decline in global HIV incidence seen from the mid-90s onwards, where new infections peaked at 3.5 million a year, to today, when 2.1 million people are infected a year, would reverse itself from 2015 and that 2.5 million a year would be infected a year in 2030. The world’s HIV-positive population would then be 41.5 million, up from 35 million today, at a time when we have all the tools to throw this process into reverse and effectively end HIV as a global epidemic.

No time to lose

At the IAPAC summit, Douglas Shaffer, Chief Medical Officer of the Office of the US Global AIDS Coordinator, said that we have “about a five-year window” to reverse the projected increases in HIV prevalence.

He offered up Uganda, an early-indicator country in Africa ever since the start of the epidemic, as an example of what could begin to happen from now on. In most African countries there has been a tight correlation in the 2000-2014 period between falls in AIDS-related deaths and falls in new infections. For instance, a recent fall of 30% in AIDS deaths in the Democratic Republic of the Congo has been accompanied by a 33% fall in new infections. In neighbouring Rwanda, an 80% fall in AIDS deaths has been accompanied by a 60% fall in new diagnoses. Even South Africa, with its huge HIV problem, saw a 40% fall in new diagnoses in this period.

But in Uganda, a 65% fall in AIDS deaths has been accompanied by a 40% increase in HIV infections. Uganda is accompanied by countries that tackled their epidemics much later such as Nigeria, which saw a 35% rise in diagnoses last year.

Shaffer put this in the context of the global standstill in HIV funding: the joint contribution to HIV of the US PEPFAR (President’s Executive Plan for AIDS Relief) programme and the Global Fund for AIDS, Tuberculosis and Malaria was $6.8 billion in 2010 and was $6.5 billion in 2015.

However, he said that despite this, the number of people on ART funded by PEPFAR alone had doubled during the same period from 3.2 million to 7.7 million and that expanding treatment did not have to depend on ever-expanding funding: “Putting the right people on the right treatment in the right way in the right places” was what mattered, he said.

The “right treatment” had to include new ideas such as offering PrEP to high-risk HIV-negative people in the same communities and pioneering “immediate test and treat” programmes that aim to maximise retention by giving people their first ART the day they are tested. The “right places” meant that resources had to be targeted at the countries not just with the highest burden of need, but with the greatest unmet needs.

Shaffer and several other speakers emphasised that choices would have to be made about what could be funded efficiently. A study in Tanzania, for instance, Shaffer said, had shown that many programmes funded under the budget line entitled 'Health Systems Strengthening' were already funded from non-HIV sources to do the same work and that it was not always necessary for HIV to carry the can for improvements in healthcare infrastructure and process, especially in countries with expanding economies.

Jonathan Mermin of the HIV, hepatitis TB and STD prevention section of the US Centers for Disease Control and Prevention (CDC) agreed, emphasising that the task of HIV prevention became ever larger as successes in ART meant that the number of people with HIV in the world would continue to rise for some time to come, from an estimated 29 million in 2000 to 37 million in 2014.

He criticised the adoption of an undiscriminating “combination prevention” approach to HIV: the combinations had to be of the measures that worked best in different populations and situations. “Not all prevention interventions are effective and not all effective interventions are equal,” he said.

He said there was evidence that simply putting more people forward for treatment was leading to global declines in HIV infections, quoting Andrew Hill’s paper presented last year at the Melbourne International AIDS Conference, which showed a strong correlation per country between the relationship between the proportion of people on treatment and the relative increase in new diagnoses. This shows, for instance, that in Botswana and Thailand, where 60% and 55% of people with HIV are on ART, new diagnoses last year formed respectively 4% and 2% of the HIV-positive population; in contrast in Nigeria and Indonesia, where respectively 15% and 5% of HIV-positive people are on ART, the increase in new diagnoses formed 8% and 12.5% of the positive population.

Doing more for the same money

He added, however, that whereas many countries reported decreases in diagnoses in other populations, countries almost universally reported increases in diagnoses in men who have sex with men (MSM); for instance in the US from 2008-2010 annual HIV incidence decreased 15% among heterosexuals, 21% among African-American women and 22% among people who inject drugs – but increased 12% in MSM.

For populations in which increased treatment coverage did not result in incidence reductions, it was necessary to add PrEP into the mix, and Mermin quoted a number of models showing that covering proportions of the high-incidence population would generally lead to a similar reduction of HIV incidence in that population; so if PrEP was offered to 30% of the highest-incidence people in Peru, incidence would fall by that figure; 20% PrEP coverage in New South Wales, Australia, would lead to a 21% decrease in incidence; even a modest 4.4% PrEP coverage target in South Africa would lead to a 3.6% fall in incidence.

Mermin added that innovative testing programmes might be among the cheapest and most cost-effective ways to increase the proportion of people diagnosed and on ART. In one mass-testing drive in Kenya recently, for instance, 47,000 people were tested in a week-long community testing programme; in Uganda, a door-to-door testing programme tested 98% of those offered a test and found an HIV prevalence of 5.8% in the previously undiagnosed; while in a household-testing programme of 7000 people in Nairobi, 96% of people took up the offer of a test – only 15% had ever tested before. Prevalence here was 13%.

Kate Callahan of the Clinton Health Access Initiative also drew attention to the role of reduced costs and efficiencies of scale in extending treatment to all. She noted that a doubling of the number of people on ART between 2014 and 2020 would be required for the global 90/90/90 target to be achieved but noted that between 2008 and 2014 the number of people on ART had been tripled while global funding had only been increased by 40%. This, she said, was because the cost of antiretroviral drugs (ARVs) had gone down by 80% during the same period and even non-ARV costs had reduced by 40%.

She said that 90/90/90 should be achievable at the current level of HIV funding. Currently the annual global HIV budget is about $19 billion. Supplying ART to people with CD4 counts below 500 cells/mm3 (in accordance with the 2013 WHO guidelines) would cost $8.9 billion. Scaling up to 90/90/90 would cost an additional $1.4 billion in drug costs with perhaps another $1.7 billion on expanded testing costs. This should, if the money is spent efficiently, leave $7 billion to finance other programmes such as PrEP, voluntary medical male circumcision and programmes for orphans and vulnerable children.

Achieving such efficiencies would be dependent, she added, on increasing the efficiency of every dollar spent on HIV; this meant that moving to near-global ART coverage would in fact be better served by moving towards targeted testing programmes instead of attempting broad-brush universal testing.

She showed, as an example, the ‘yield’ in HIV positive tests seen in Zimbabwe in children being tested for HIV. In common with adult HIV testing programmes, this found that testing only in children from specific populations with specific conditions led to much higher diagnosis rates per dollar.

If all children were tested, 2.1% tested HIV-positive, which reflects current national prevalence in under-15s. If children whose mothers had HIV and were in prevention of mother-to-child-transmission programmes were tested, the positivity rate was 6.1%. If children who were in-patients in hospital were tested, 15.5% had HIV; while in children diagnosed with malnutrition, 28% had HIV. Finally, if children were diagnosed with TB, 69% turned out to have HIV. Callahan recommended testing all children diagnosed with malnutrition in non-famine situations as a cost-effective way to test this group.

Progress towards full treatment

A number of countries report approaching the UNAIDS target, at least in some programmes. For instance, Joseph Makhema of the joint Botswana-Harvard HIV programme reported that baseline door-to-door testing surveys of people in a community prevention and treatment trial showed that 94% of people who fit the country ART criteria of CD4 counts on diagnosis over 350 cells/mm3 (350-500 cells/mm3 with a viral load below 10,000 on diagnosis) were already on ART and that 86% of those diagnosed that year were already on ART. The best estimate for Botswana’s approach to 90/90/90 is currently 82/86/95 or 67% of all people with HIV virally suppressed: higher than countries like the UK and equal to Europe’s highest performer, Switzerland.

Botswana is a small country with a relatively rich economy and larger countries may face bigger barriers to full access. Marcelo de Freitas of the Brazilian CDC said that since his country adopted access to ART for all, the proportion of people starting ART who had CD4 counts over 500 cells/mm3 was now 41% of the total, up from 10% in 2011. And the CD4 count at initiation had increased from 270 to 442 cells/mm3. Seventy-eight per cent of people on ART were virally suppressed, leading to an estimated 40% of people with HIV virally suppressed.

David Burns of the US National Institutes of Health showed a review of ‘treatment cascades’ that showed that Brazil, with its 40%, stood about midway between countries like Botswana, Australia (at 62%); was somewhat above the sub-Saharan average of 32% and the current, though increasing, US one of 30%, and way above countries like Ukraine (17%) and Russia (9%).

Rwanda is another African success story with 60% of people with HIV virally suppressed. However Sabin Nsanzimana of the country’s HIV prevention division emphasised that even at this kind of level of coverage, there were still roughly 10,000 people being infected for every 5,000 HIV-positive people who died and thus prevalence was still going up.

Velephiu Okello from the Swaziland health ministry reported on that country’s step-by-step introduction of intensified HIV testing and treatment programmes in paired clinics – see this report on this so-called MaxART project. ART uptake in those who tested HIV positive in the programme was 84% as of last June and the 12 month retention in care rate was 88%. Seventy per cent of those offered ART chose to start it the day they tested HIV positive.

Reducing the stigma of testing and treatment

Okello reported that one of the strongest predictors of not taking HIV was self-reported shame at having HIV. François Dabis and Deenan Pillay of the ANRS12249 study in KwaZulu Natal in South Africa also reported that seemingly slight barriers such as having to walk more than one kilometre to the clinic also had a strong impact on starting and continuing with ART. This study also found that people with high education levels were actually less likely to start and continue with ART, especially men, possibly because of higher mobility.

Finally, Nattaya Phanuphak of Thailand reported on programmes aimed at reducing the stigma of HIV testing and of taking PrEP.

She said she was a strong believer in social media interventions, in a world where 3.7 billion of the world’s 7.35 billion people has a mobile phone, 3.2 billion is on the internet and 2.2 billion belongs to a social media site.

She introduced the Thai site for MSM in Thailand, which is being exported to Taiwan too. However such an explicit site had a problem with more conservative Muslim countries and thus the site was devised for Indonesia and Malaysia where the message is less explicitly (though still subtly) directed at MSM.

There have been pilot mobile testing initiatives in Indonesia. In one, the stigma of HIV testing (and of being seen to be gay) was overcome by offering STI and HIV testing as an ‘extra’ alongside tests for body fat, BMI and muscle mass, designed to appeal to the vanity of bathhouse clients, both MSM and heterosexual. In the pilot 103 men took tests in two days of whom 51% disclosed being MSM in confidential settings.

In Thailand, community-based testing in parks, bathhouses and gay venues as well as clinics has found an overall prevalence of 18%, 65% of whom started ART. A quarter of those tested were in fact transgender women – a group often seen as hard to reach. Among them HIV prevalence was actually lower at 10% but ART uptake higher, at 86%.

Thailand has also introduced a subsidised PrEP programme where MSM in need of PrEP contribute a standard charge of 30 baht per month (less than a US dollar) for a prescription.

IAPAC’s Ben Young gave one of the most optimistic forecasts for the achievement of 90/90/90. He commented that life expectancy in people with HIV was rising everywhere – in the US the total life expectancy was 49 in people with CD4 counts over 350 cells/mm3 in 2000 and was already 69 by 2006; in South Africans with CD4 counts over 200 cells/mm3 it was now 67 years at age 25 and 75 years at age 55. Better ARVs continued to be developed, and there was encouragingly a fall in HIV drug resistance, often an indicator of treatment failure, and almost no resistance reported to the integrase inhibitor class (see this report).

An estimated 2.4 million people’s lives will have been saved in South Africa alone by 2020 with the introduction of ART but implementing 90/90/90 fully would save 3.25 million.

If funders and health providers protested at the cost, he commented, “We need to say that the consensus on the benefit of treatment is clear; the impact is already happening; the risks are acceptable and the consequences of not doing it serious; and we can do it at reasonable cost.”