Testing strategies to detect acute and prevalent HIV infection in adult outpatients seeking healthcare for symptoms compatible with acute HIV infection in Kenya: a cost-effectiveness analysis

Detection of acute and prevalent HIV infection using point-of-care nucleic acid amplification testing (POC-NAAT) among outpatients with symptoms compatible with acute HIV is critical to HIV prevention, but it is not clear if it is cost-effective compared to existing HIV testing strategies. We developed and parametrised a decision tree to compare the cost-effectiveness of: (1) provider-initiated testing and counseling (PITC) using rapid tests, the standard of care; (2) scaled-up PITC (SU-PITC) in which all patients were tested with rapid tests unless they opted out; and (3) opt-out testing and counseling using POC-NAAT, which detects both acute and prevalent infection. The model-based analysis used data from the Tambua Mapema Plus randomised controlled trial of a POC-NAAT intervention in Kenya, supplemented with results from a stochastic, agent-based network model of HIV-1 transmission and data from the published literature. The analysis was conducted from the perspective of the Kenyan government using a primary outcome of cost per DALY averted over a 10-year time horizon. After analysing the decision-analytic model, the average per patient cost of POC-NAAT was $214.9 compared with $173.6 for SU-PITC and $47.3 for PITC. The mean DALYs accumulated per patient for POC-NAAT were 0.160 compared with 0.176 for SU-PITC and 0.214 for PITC. In the incremental analysis, SU-PITC was eliminated due to extended dominance and the ICER comparing POC-NAAT to PITC was $3,098 per DALY averted. The ICER was sensitive to disability weights for HIV/AIDs and the costs of antiretroviral therapy. The conclusion was that POC-NAAT offered to adult outpatients in Kenya who present for care with symptoms compatible with AHI is cost-effective and should be considered for inclusion as the standard of HIV testing in this population.