Photo: Peter Mogere/PHRD Thika

“DSD in action” interview: Integrating HIV self-testing into PrEP in Kenya

In Kenya, PrEP is dispensed mostly through HIV clinics within public health facilities at quarterly visits. Barriers include stigma associated with accessing prevention services at HIV clinics, long wait and travel times, overburdened healthcare providers, stock outs, inadequate information about HIV prevention options, and limited clinic hours. DSD for PrEP refills and continuation could empower people to take charge of their own care. However, these models rely on a robust supply chain of PrEP drugs, which has been a challenge in many settings.

We spoke with IAS Governing Council member Dr Kenneth Ngure, Associate Professor at the School of Public Health, Jomo Kenyatta University of Agriculture and Technology, Kenya, and Dr Katrina Ortblad, Assistant Professor at the Fred Hutchinson Cancer Research Center, United States.

At CROI 2022, you presented the results of a randomized integrated PrEP and HIV self-testing (HIVST) model you implemented in Kenya. Can you describe the service delivery model in detail?

We conducted a randomized non-inferiority implementation trial that tested six-monthly PrEP dispensing supported with interim HIVST. Participants received either six-month PrEP dispensing with clinic visits at the same frequency supported by HIVST conducted at home after three months, or three-month PrEP dispensing with clinic visits at the same frequency and clinic-based HIV testing. The screening and initiation took place at public clinics, whereas PrEP refills and routine clinical follow up took place at the research clinic. The clinical follow-up service package included rapid diagnostic testing, counselling, syndromic STI testing, creatinine testing, screening for symptoms of PrEP side-effects and/or early HIV acquisition.

What are the key findings from this trial?

We found that six-month PrEP dispensing supported by interim at-home HIV self-testing at three months halved the number of PrEP clinic visits while maintaining equivalent HIV testing, PrEP refilling and PrEP adherence at six months compared with standard-of-care PrEP delivery with quarterly clinic visits.

Looking at client outcomes at sub-population level, did you see any significant differences?

Among women not part of HIV serodifferent couples, this model of six-month PrEP dispensing with interim HIVST significantly increased PrEP adherence, measured objectively with blood sampling, compared with standard-of-care PrEP delivery.

What did you learn about clients’ and healthcare workers’ perspectives and satisfaction with the model?

We collected serial qualitative data from clients on their perceived acceptability of this HIVST-supported PrEP continuation model (at enrolment) and experiences with this model (at months six and 12). We assessed concepts related to clients’ convenience, self-efficacy, stigma, costs (time and economic), sexual partner perceptions and willingness to continue the delivery model post-study and recommend it to others. We also collected qualitative data from health providers to capture their perceptions of this novel delivery model and its components, including how it reduced workload and assisted clients’ PrEP access and use. Responses from clients and providers were mostly positive.

What are the next steps for implementation and scale up?

Our team is excited about the findings from this trial and has plans to evaluate scale up this model to public facilities in Kenya and the region. Our findings should motivate new DSD models of PrEP refilling supported with HIVST that can help move PrEP continuation outside healthcare facilities. Our team is finalizing the costs and cost savings associated with this new HIVST-supported PrEP delivery model, which will inform future policy decisions.

We will share the findings from our study with the Kenya Ministry of Health’s PrEP Technical Working Group and, based on our earlier discussions during the conceptualization and conduct of the study, we are positive that the ministry will find these findings informative for future PrEP programming.