Population Uptake and Effectiveness of Test-and-Treat Antiretroviral Therapy Guidelines for Preventing the Global Spread of HIV

An Ecological Cross-National Analysis

A Mendez-Lopez; M McKee; D Stuckler; R Granich; S Gupta; T Noori; JC Semenza


HIV Medicine. 2019;20(8):501-512. 

In This Article

Abstract and Introduction


Objectives: Although the benefits of adopting test-and-treat antiretroviral therapy (ART) guidelines that recommend initiation of ART regardless of CD4 cell counts have been demonstrated at the individual level, there is uncertainty about how this translates to the population level. Here, we explored whether adopting ART guidelines recommending earlier treatment initiation improves population ART access and viral suppression and reduces overall disease transmission.

Methods: Data on ART initiation guidelines and treatment coverage, viral suppression, and HIV incidence from 37 European and Central Asian countries were collected from the European Centre for Disease Prevention and Control and the Global HIV Policy Watch and HIV 90-90-90 Watch databases. We used multivariate linear regression models to quantify the association of ART initiation guidelines with population ART access, viral suppression, and HIV incidence, adjusting for potential confounding factors.

Results: Test-and-treat policies were associated with 15.2 percentage points (pp) [95% confidence interval (CI) 0.8–29.6 pp; P = 0.039] greater treatment coverage (proportion of HIV-positive people on ART) compared with countries with ART initiation at CD4 cell counts ≤ 350 cells/μL. The presence of test-and-treat policies was associated with 15.8 pp (95% CI 2.4–29.1 pp; P = 0.023) higher viral suppression rates (people on ART virally suppressed) compared with countries with treatment initiation at CD4 counts ≤ 350 cells/μL. ART initiation at CD4 counts ≤ 500 cells/μL did not significantly improve ART coverage compared to initiation at CD4 counts ≤ 350 cells/μL but achieved similar degrees of viral suppression as test-and-treat.

Conclusions: Test-and-treat was found to be associated with substantial improvements in population-level access to ART and viral suppression, further strengthening evidence that rapid initiation of treatment will help curb the spread of HIV.


In 2015, the World Health Organization (WHO) and the European AIDS Clinical Society called for universal test-and-treat programmes, with initiation of antiretroviral therapy (ART) immediately upon diagnosis of HIV infection, as a means to reduce rates of HIV-related illness and mortality and onward transmission.[1–3] The rationale for reducing onward transmission derived primarily from evidence that early treatment reduced the risk of mother-to-child transmission and in serodiscordant couples.[4–16] The HIV Prevention Trials Network (HPTN052) trial had found that ART initiation at CD4 counts of between 350 and 550 cells/μL led to a reduction of 96% in HIV transmission compared to delaying ART initiation until the CD4 count was ≤ 250 cells/μL.[12,16] This was consistent with earlier observational studies and supported by systematic reviews.[11,13,14,17] Yet, the argument that this would lead to population-level benefits was controversial. Some argued that expanding ART might create a false sense of security among those affected, perversely encouraging greater rates of unsafe sex,[18–20] which has been contested.[21–24] Others highlighted constraints to scaling up treatment as a result of limited resources, especially in low-income settings,[25–27] uncertainty about the use of data from clinical trials that showed 'modest benefits',[28] nonreplicability at the community level,[29,30] and the risk of increasing rates of adverse effects caused by ART and resistance.[26,31]

In Europe, a key argument centred on whether findings in couples could be generalized to the wider population, especially as the incidence was lower than in other parts of the world and, in many European countries, was declining. This reflected the limited evidence at the population level, with studies producing mixed findings but often suggesting that population-level benefits may be more modest than those found in trials at the individual level. A number of ecological studies have been carried out, but mostly in single communities. An association between greater ART coverage and lower viral loads and transmission has been reported in diverse settings, including British Columbia in Canada,[32–34] San Francisco in the USA,[35] KwaZulu-Natal in South Africa,[17,36] and Taiwan.[37] One cross-national study found that expanding ART coverage in the 30 highest AIDS mortality burden countries correlated with reduced mortality rates from HIV-related causes.[38] However, a recent review argued that findings from existing population-level studies were mixed, with one study reporting decreasing risk per contact among those on ART being counteracted by more unsafe sexual episodes.[18] Another review found that test-and-treat appeared to be less effective at the population level than anticipated from modelling studies.[29,30]

Here, we take advantage of a unique opportunity to test the impact of the expansion of test-and-treat policies in 37 European and Central Asian countries. Several countries pre-empted the adoption of the test-and-treat guidelines in WHO's 2015 recommendations, while others have yet to change (see Table S1). These marked differences in timing enabled us to test the hypothesis that expanding test-and-treat guidelines increases population access to ART coverage and, in so doing, improves viral suppression and reduces HIV incidence (as described in Figure 1).

Figure 1.

Conceptual framework of the relationship between antiretroviral therapy (ART) initiation policies, ART coverage and viral suppression.