A dynamic, person-centred HIV prevention intervention resulted in substantial increases in the numbers of people covered by PrEP or PEP in rural Uganda and Kenya, researchers told the 30th Conference on Retroviruses and Opportunistic Infections (CROI 2023) in Seattle last week. Three separate randomised controlled trials conducted using village health teams, in antenatal clinics and in hospital outpatient departments each showed the benefit of this model of care.
The researchers conducted qualitative work to identify barriers to the uptake of HIV prevention and design their ‘dynamic choice prevention intervention’. The model of care was essentially the same in all three settings:
- A choice of prevention product – people could switch between oral PrEP, PEP, condoms and nothing as their needs and preferences changed.
- A choice of service location – people could receive services from a clinic, over the phone, in their home or in another community setting. Nurses and community health workers could visit people in a convenient place, including away from their home if that addressed confidentiality concerns.
- A choice of ways to take an HIV test – self-testing or a rapid test done by a healthcare worker.
- People could receive up to three months’ of PrEP at a time.
- Starter packs of PEP were provided, giving people enough tablets for the first few days, with the rest of the course provided when needed.
- Clinicians could be reached by phone seven days a week to discuss starting PEP or PrEP, give advice and answer questions.
- Staff were trained to assess barriers to starting and sticking with PEP or PrEP, and to develop personalised plans in response to those barriers.
- Staff provided referrals to services addressing reproductive health, sexually transmitted infections, trauma and gender-based violence.
In each study, the intervention was compared to the care that is routinely available – referral to other health facilities to access PEP and PrEP. Outcomes were assessed in HIV-negative people who were considered at risk of acquiring HIV (most often, because they did not know the HIV status of one of their partners). In each study, outcomes were assessed in around 400 people (half who had access to the intervention and half who received the standard of care).
The impact of the intervention was assessed by measuring biomedical HIV prevention coverage – the proportion of time that people said they were using PEP or PrEP during almost a year of follow-up. They also assessed biomedical HIV prevention coverage during periods ‘at risk’ – periods in which people reported sex with someone who was not known to be HIV negative, reported transactional sex or felt they were at risk.
The first of the three studies was conducted with community health workers going door to door in villages. Sixteen villages were randomly assigned to either receive the intervention or not, with outcomes assessed in 413 individuals. Over half were female and over a third were aged 15 to 24 years.
In the standard of care arm, PrEP or PEP was used for 0.5% of the time during follow up (0.9% of periods at risk). This compares to 28% of time in the intervention arm (36% of time at risk). Over half the participants in the intervention arm used PrEP at least once and over half used PEP at least once, with considerable fluctuations in both exposure to risk and product preferences over time.
Women are at high risk of acquiring HIV during pregnancy and after birth, so four antenatal clinics were selected as the sites of the second study. Half of the 400 HIV-negative women taking part were aged 15 to 24 years. In those receiving the standard of care, prevention coverage was 29% (and 43% during periods or risk), rising in the intervention arm to 70% (and 83% during periods of risk). PrEP was generally preferred in this setting, with all participants using it at least once, compared to 11% using PEP.
The third site was hospital outpatient departments. As in the community study, over half the participants were female and over a third were aged 15 to 24 years. In the standard of care arm, PrEP or PEP was used for 18% of the time during follow up (26% of periods at risk). This compares to 48% of the time in the intervention arm (65% of time at risk). As in the first study, the intervention worked well for both women and men.
In all three studies an increasing number of people chose self-testing as time went on. Similarly, in the last two studies, the number getting their care in community settings increased during follow-up, whereas in the village study, almost everyone received services at home throughout.
“Participant choices varied over time, emphasising the benefit of flexibility in HIV prevention services with changing needs of clients,” Dr Elijah Kakande of the Infectious Diseases Research Collaboration in Kampala concluded.
Although the dynamic model increased uptake of biomedical prevention from virtually nothing to 28% in the village study, and more than doubled it at the other two sites, not everyone took up the offer and there was some drop-off in engagement over time. The researchers hope that giving an additional option of injectable cabotegravir for PrEP may boost prevention coverage. They are planning a large community cluster randomised trial which incorporates this option and assesses the impact of the model of care on new HIV infections.
Kakande E et al. Randomized trial of community health worker delivered dynamic choice HIV prevention. Conference on Retroviruses and Opportunistic Infections, Seattle, abstract 124, 2023.
Kamya M et al. Randomized trial of dynamic choice HIV prevention in ante/postnatal care clinics. Conference on Retroviruses and Opportunistic Infections, Seattle, abstract 128, 2023.
Koss CA et al. Randomized trial of dynamic choice prevention at outpatient department in East Africa. Conference on Retroviruses and Opportunistic Infections, Seattle, abstract 975, 2023.