Background
The SEARCH dynamic choice HIV prevention (DCP) intervention, including structured choice of biomedical prevention modality (oral PrEP, PEP, Cab-LA) with option to switch over time, increased time covered by biomedical prevention and reduced HIV incidence among men and women in randomized trials in Kenya and Uganda. We use HIV Synthesis, an existing individual-based HIV simulation model, to project long-term impact and cost-effectiveness of this approach in Africa.
Methods
By sampling parameter values at the start of each model run we created 500 “setting-scenarios” reflecting uncertainty in assumptions and a range of characteristics representing those observed across east and southern Africa. For each setting-scenario, 3 policies over 10 and 50 years beginning 2024 were considered: (i) status quo, (ii) DCP with oral PrEP, PEP and condoms; (iii) DCP also including Cab-LA. We used empirical data from the DCP trials to inform the intervention’s impact on PrEP/PEP initiation and persistence, and hence HIV incidence, as well as costs. Cost-effectiveness analysis was from a healthcare perspective (using cost-effectiveness threshold US$500 per DALY averted). We calculated net DALYs averted for DCP with and without Cab-LA, relative to status quo, to reflect predicted policy impact on overall population burden of disease, and incremental cost effectiveness ratio (ICER).
Results
The DCP interventions are predicted to increase the proportion of people with a current PrEP/PEP indication who take PrEP/PEP and hence reduce HIV incidence relative to status quo over 10 years; median (90% range) across setting-scenarios HIV incidence rate ratios of 0.90 (0.64 – 1.19) and 0.66 (0.42 – 0.96) for DCP without and with Cab-LA, respectively. In the context of 10 million adults, both DCP policies lead to DALYs being averted (mean -12,500 and -35,000 for DCP without and with Cab-LA, respectively) but also to greater costs (+$8.4m and +$13.6m for DCP without and with Cab-LA, respectively). Over 50 years, DCP without Cab-LA is not predicted to be cost-effective (+4200 mean net DALYs; ICER if Cab-LA inclusion is not an option = $672/DALY averted. DCP with Cab-LA incurs the lowest mean net DALYs (-7700 per year over 50 years; ICER compared with status quo = $389), and is the cost-effective policy.
Conclusions
Offering structured PrEP/PEP choice, including Cab-LA, could reduce HIV incidence by one-third over 10 years and is likely cost-effective across settings in east and southern Africa.