Confidential social network referrals for HIV testing (CONSORT) - UNAIDS set for 2030 the ambitious 95-95-95 target: diagnosing 95% of all persons living with HIV (PWH), initiating antiretroviral therapy for 95% of those diagnosed, and achieving viral suppression for 95% of those treated. In Tanzania, a PEPFAR strategy country with an adult HIV prevalence of 4.8%, only 83% of PWH are aware of their status. More than 200,000 undiagnosed PWH need to test in order to reach the “first 95”. With the cost-effectiveness of traditional HIV testing approaches declining, we propose to evaluate a novel application of mobile health (mHealth) technology that leverages the ubiquity of mobile phones and the reach of social networks to increase rates of HIV testing, especially among higher-risk individuals. Specifically, we will evaluate the acceptability, efficacy, and cost-effectiveness of automated, confidential, SMS-based HIV testing referrals, as a means of ‘nudging’ individuals to test. We hypothesize that an automated, confidential referral system, developed and deployed in the Kilimanjaro Region of Tanzania, will be acceptable to both index clients and their referrals and that it will be cost-effective for increasing HIV testing rates. Building on existing relationships with all 25 HIV counseling and testing (HCT) providers in the study area, including 8 HIV care and treatment centers (CTCs), and using an open-source, low-code, and highly versatile mobile phone-based appointment reminder and incentive system (mParis), this study will address the following specific aims: R21 Phase: Aim 1 will conduct qualitative, formative work with (1) HCT providers, (2) HCT clients, (3) PWH who are in care at CTCs, and (4) social network contacts of HCT and CTC clients, to identify desirable provider- and client-side characteristics of an SMS-based HIV testing referral system. Aim 2 will adapt our mParis system to receive phone numbers of HIV testing referrals identified by index clients and autonomously send confidential HIV testing invitations. Aim 3 will pilot-test the Confidential Social Network Referrals for HIV Testing (CONSORT) system to collect preliminary data on the system’s acceptability, performance, and potential efficacy. R33 Phase: Aim 4 will evaluate the acceptability and efficacy of CONSORT in a pragmatic randomized controlled trial with 400 HCT clients and 200 HIV-infected CTC patients. Arm 1 participants will be offered confidential SMS referrals and physical invitation cards (“card referrals”) to extend to any of their network contacts. Arm 2 participants will be offered card referrals alone. The primary outcome will be uptake of HIV testing. Secondary outcomes include the number of new HIV diagnoses, and the risk correlation within referral networks. Aim 5 will evaluate the incremental cost-effectiveness of CONSORT+card referrals vs. card referrals alone. The study will provide estimates of the acceptability, efficacy, and cost-effectiveness of an exceptionally low-cost intervention for increasing the uptake of HIV testing. More broadly it will develop analytic and mHealth capacity in Tanzania and inform the development of mobile phone-based chain-referral interventions that can reach key social networks, are transferrable to other technologies, and can be extended beyond HIV testing.
