Linkage to HIV care still challenging for young men, Botswana study finds

Almost one in five people newly diagnosed with HIV in a study employing active methods to link people to care and treatment had still not linked to care a year after diagnosis, a large study in Botswana has found. The problem was most acute in young men and those in employment, leading the investigators to suggest that despite high rates of linkage to care in the rest of the population, alternative treatment delivery models may be needed for these groups.

The findings were presented last month at the Conference on Retroviruses and Opportunistic Infections (CROI 2017) in Seattle.

Achieving linkage to care of people diagnosed with HIV has emerged as a major obstacle to achieving the UNAIDS 90-90-90 goals, which call for 90% of people diagnosed with HIV to be on treatment and 90% of people on treatment to have undetectable viral load by 2020.

A recently published study conducted in South Africa showed good retention in care but very slow linkage to care among people diagnosed with HIV in a large population study in KwaZulu-Natal. Active approaches that link people to care, whether through personal or telephone follow-up, peer support, accompaniment to health facilities, or even antiretroviral therapy (ART) initiation at home, are being tested in sub-Saharan Africa in order to identify what can be done to improve linkage to care.

The Botswana Combination Prevention Project is a community-randomised trial comparing the impact of standard of care or an aggressive combination prevention package on HIV incidence in selected communities in Botswana. The intervention includes home-based or mobile testing and point-of-care CD4 test, next-day clinic appointments for people who test positive. Text message appointment reminders and active tracing by phone or home visits are used to ensure retention in care.

The investigators of the Botswana Combination Prevention Project reported on linkage to care and ART initiation among people in intervention communities in two poster presentations at CROI 2017.

The first round of HIV testing campaigns began in September 2013. The study has randomised 30 communities to receive the intervention package or the standard of care, and the intervention has been rolled out in three phases, each to five pairs of villages. An intensive local HIV testing campaign lasting six to eight weeks and linkage to care is followed by an analysis of linkage to care, treatment initiation and retention in care during the subsequent year.

In 2016 results of a survey of 20% of households in 24 of the villages in the study were presented at CROI. The results presented at CROI 2017 consist of the linkage to care outcomes of the 15 communities receiving the combination prevention package, after the first round of testing campaigns and one-year follow-up, and the treatment initiation outcomes of all those in the intervention communities who tested positive and who were not on treatment at the time they were enrolled into the study.

Linkage to care

The study population consisted of 40,628 people tested for HIV, of whom 24% were HIV positive. Testing identified 2593 people not already on ART, of whom 1592 were previously undiagnosed.

Of those who knew their HIV status already, 85% linked to care within 90 days and 87% within a year, with no substantial difference between men and women overall. Of those who were newly diagnosed, 81% linked to care within 90 days and 82% within a year – in other words, almost one in five of the newly diagnosed remained out of care.

Again, there was no substantial difference between men and women overall. However, analysis by age and gender showed that young men aged 16-24 were less likely to link to care (72% vs 83% for women aged 16-24, and 84% for men aged 25-64). Men aged 16-24 who were employed were least likely to link to care, and inability to miss work was commonly cited as a reason for not having linked to care.

The study also found that future lack of linkage to care might be predictable at the time of HIV diagnosis from one simple question: only 45% of those who refused follow-up services subsequently linked to care, compared to 88% of those who agreed to follow-up services.

The researchers said that tracing of people who had not linked to care was necessary for 32% of those who eventually linked to care. On average, these people required 2.2 home visits or telephone calls before they attended the clinic.

Alternatives to standard clinic delivery might be needed to achieve further improvement in the uptake of ART, the researchers concluded.

Fast-track ART initiation

The researchers also looked at ART initiation for those diagnosed with HIV, before and after the introduction of fast-track ART initiation in June 2016. Botswana moved to a policy of universal treatment in June 2016, together with fast-track initiation. Prior to June 2016, people eligible for ART were required to attend their clinic for two or three counselling visits and for baseline blood tests.

Round One identified 2354 people with HIV not on ART, of whom 34% had CD4 cell counts below 200 cells/mm³. Prior to the introduction of the universal test and treat (UTT) policy, 1120 people had CD4 cell counts below 350 cells/mm³ and qualified for treatment. A further 239 people were classified as eligible for treatment, so that overall, 77% of eligible people had initiated treatment prior to the introduction of UTT. It took a median of 35 days for people to start treatment prior to the introduction of fast-track treatment.

After the introduction of UTT, 1426 people eligible for treatment attended a clinic up to November 2016, and 96% subsequently started treatment, 76% on the same day as their clinic visit, 90% within one week and 97% within one month. Almost a quarter of those who started treatment in this period had a CD4 cell count below 200 cells/mm³. Of 803 people who started fast-track treatment and who had a viral load test three months after starting treatment, 89% had a viral load below 400 copies/ml.

Baseline testing for creatinine clearance was carried out for fast-track patients, in order to identify people with impaired kidney function who might be harmed by tenofovir treatment. Waiting for test results might delay treatment initiation, so people started treatment at the same time as having blood drawn for creatinine testing.

Three per cent of people (42) had a creatinine clearance below the recommended level (< 60mls/min). Thirteen needed urgent treatment owing to clinical symptoms and were started on non-tenofovir-containing regimens. Of the remaining 29 who started tenofovir, 12 experienced improvement in creatinine clearance, ten switched to an alternative regimen, five awaited repeat test results and two died.

The investigators concluded that fast-track initiation is feasible in Botswana, resulting in high rates of treatment initiation and big reductions in the time taken to start treatment once people became eligible.