HIV treatment eligibility expansion and timely antiretroviral treatment initiation following enrollment in HIV care: A metaregression analysis of programmatic data from 22 countries

Abstract

BackgroundThe effect of antiretroviral treatment (ART) eligibility expansions on patient outcomes, including rates of timely ART initiation among those enrolling in care, has not been assessed on a large scale. In addition, it is not known whether ART eligibility expansions may lead to “crowding out” of sicker patients.Methods and findingsWe examined changes in timely ART initiation (within 6 months) at the original site of HIV care enrollment after ART eligibility expansions among 284,740 adult ART-naïve patients at 171 International Epidemiology Databases to Evaluate AIDS (IeDEA) network sites in 22 countries where national policies expanding ART eligibility were introduced between 2007 and 2015. Half of the sites included in this analysis were from Southern Africa, one-third were from East Africa, and the remainder were from the Asia-Pacific, Central Africa, North America, and South and Central America regions. The median age of patients enrolling in care at contributing sites was 33.5 years, and the median percentage of female patients at these clinics was 62.5%. We assessed the 6-month cumulative incidence of timely ART initiation (CI-ART) before and after major expansions of ART eligibility (i.e., expansion to treat persons with CD4 ≤ 350 cells/μL [145 sites in 22 countries] and CD4 ≤ 500 cells/μL [152 sites in 15 countries]). Random effects metaregression models were used to estimate absolute changes in CI-ART at each site before and after guideline expansion. The crude pooled estimate of change in CI-ART was 4.3 percentage points (95% confidence interval [CI] 2.6 to 6.1) after ART eligibility expansion to CD4 ≤ 350, from a baseline median CI-ART of 53%; and 15.9 percentage points (pp) (95% CI 14.3 to 17.4) after ART eligibility expansion to CD4 ≤ 500, from a baseline median CI-ART of 57%. The largest increases in CI-ART were observed among those newly eligible for treatment (18.2 pp after expansion to CD4 ≤ 350 and 47.4 pp after expansion to CD4 ≤ 500), with no change or small increases among those eligible under prior guidelines (CD4 ≤ 350: −0.6 pp, 95% CI −2.0 to 0.7 pp; CD4 ≤ 500: 4.9 pp, 95% CI 3.3 to 6.5 pp). For ART eligibility expansion to CD4 ≤ 500, changes in CI-ART were largest among younger patients (16–24 years: 21.5 pp, 95% CI 18.9 to 24.2 pp). Key limitations include the lack of a counterfactual and difficulty accounting for secular outcome trends, due to universal exposure to guideline changes in each country.ConclusionsThese findings underscore the potential of ART eligibility expansion to improve the timeliness of ART initiation globally, particularly for young adults.

Early outcomes after implementation of treat all in Rwanda: an interrupted time series study

“…Similar results were observed even when excluding patients who were not eligible for ART prior to Treat All implementation, indicating that the change was not solely due to the new guideline and its implementation at the sites, but rather in improved programme capacity to rapidly initiate ART. These data are consistent with successful implementation of earlier ART guidelines in Rwanda leading to rapid increases in median CD4 count at ART initiation, as well as global data showing rapid increases in ART uptake as treatment thresholds increased .…”

“…Among patients enrolling in care after Treat All implementation, we did not find an association between baseline CD4 count and 30‐day ART initiation. Prior studies, including the Treatment as Prevention trial, have demonstrated that treatment expansion does not result in delayed ART initiation among sicker patients . To our knowledge, our study is the first to confirm these findings within a routine implementation of a Treat All paradigm.…”

Research priorities to inform “Treat All” policy implementation for people living with HIV in sub‐Saharan Africa: a consensus statement from the International epidemiology Databases to Evaluate AIDS (IeDEA)

“…In addition, in some settings, routine HIV viral load monitoring is infrequent, and pre‐ART CD4 count monitoring is declining in frequency with Treat All implementation , which limits opportunities to evaluate individual and public health impacts of HIV programming. Four key metrics for assessing the timeliness of continuum milestones include: (1) median CD4 count at diagnosis , care enrollment , and ART initiation ; (2) time between diagnosis, enrollment ; (3) time between enrollment and ART initiation ; and (4) time to first HIV viral suppression and sustained HIV viral suppression. Research approaches : Metrics for this research priority could be generated from routinely collected, patient‐level, programmatic data. Where CD4 count and viral load data are not readily available for a large enough proportion of clinics and patients, it may be possible to produce estimates from a systematic sample of sites.…”

“…at higher CD4 counts) and linkage to care are needed. Although more real‐world data on timely ART uptake under Treat All implementation are needed, early evidence on ‘Treat All’ in SSA and evidence from previous HIV treatment guideline expansions suggest that if people are eligible for treatment when they link to care, they will start ART rapidly with early retention in care and viral suppression following ART initiation .…”

Temporal changes in ART initiation in adults with high CD4 counts in Latin America: a cohort study

“…Temporal trends in time to cART initiation in Latin America in asymptomatic patients with high CD4 cell counts have not been evaluated previously, though several studies have explored time to cART initiation overall, also showing trends towards earlier cART initiation over time . Of note, most studies done looking at time to cART initiation look at CD4 count at the time of cART initiation, rather than median time (in days or years).…”