CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

1062 Costs of Hybrid Mobile Multi-Disease TestingWith High HIV Test Coverage, East Africa Wei Chang 1 ; Gabriel Chamie 1 ; HarshaThirumurthy 2 ;Tamara Clark 1 ; Edwin Charlebois 1 ; Maya Petersen 3 ; Moses R. Kamya 4 ; DianeV. Havlir 1 ; James G. Kahn 1 ; for the Sustainable East Africa Research for Community Health (SEARCH) Collaboration 1 Univ of California San Francisco, San Francisco, CA, USA; 2 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 3 Univ of California Berkeley, Berkeley, CA, USA; 4 Makerere Univ Coll of Hlth Scis, Kampala, Uganda Background: In 2013-14, we achieved 89% adult HIV testing coverage using a hybrid mobile HIV testing approach of 2-week multi-disease community health campaigns (CHC) followed by home-based testing (HBT) of CHC non-attendees in 32 communities in Uganda and Kenya (SEARCH: NCT01864603). To inform scalability, we sought to determine: 1) overall costs of our hybrid testing approach; and 2) costs associated with including point-of-care (POC) CD4 testing and multi-disease services (hypertension, diabetes, and malaria) - elements crucial to our hybrid approach, but absent in most community-based testing strategies. Methods: We applied micro-costing methods to quantify resources used for population-wide HIV testing using our hybrid mobile approach in 12 of 32 SEARCH Trial communities. Key intervention components are door-to-door baseline census enumeration, multi-disease CHCs, and HBT of CHC non-attendees. POC CD4 tests were done for all HIV+ participants. Data were obtained from expenditure records and study logs (e.g., costs for salaries, rapid HIV antibody and POC CD4 tests, transportation, and mobilization). Interviews with study staff and time and motion exercises were conducted to determine staff time allocation to various activities. Results: The average cost per adult (≥15 years) tested for HIV was $20.51 (range: $17.06 – $32.08 [SD = $3.84]) across communities [2014 US$], including POC CD4 at $16 per test which represents 5-13% of total testing costs. Cost per adult tested at CHCs was $13.83 vs . $31.71 via HBT. The costs per adult tested were consistent across sites, except for an island community where staff transportation costs were high. The cost per HIV+ adult identified was $231 (range: $87 – $1,245 [SD=$336]); the variability in this measure was mainly due to differences in HIV prevalence (e.g., HIV prevalence of 23.56% vs .1.62%). The marginal cost attributable to multi-disease services at CHCs was $1.16 per person for hypertension and diabetes screening, and $0.90 per person tested for malaria. Figure 1 compares costs of our hybrid mobile approach vs . mobile testing alone, HBT alone, and fixed venue based testing. Conclusions: While achieving high HIV testing coverage with added intervention components including census, POC CD4 testing, and multi-disease services, the cost of this hybrid mobile testing approach is in the range of previously reported mobile, home-based, and fixed venue based HIV testing implementations.

Poster Abstracts

1063 Targeting Serodiscordant Couples Within Home HIV Testing Campaigns: A Modeling Study Sarah T. Roberts 1 ; Aditya S. Khanna 2 ; RuanneV. Barnabas 1 ; Steven M. Goodreau 1 ; Jared M. Baeten 1 ; Connie M. Celum 1 ; Susan L. Cassels 3 1 Univ of Washington, Seattle, WA, USA; 2 Univ of Chicago, Chicago, IL, USA; 3 Univ of California Santa Barbara, Santa Barbara, CA, USA

Background: The World Health Organization recommends immediate antiretroviral therapy (ART) for HIV-infected members of HIV serodiscordant couples (SDC) to reduce the risk of HIV transmission. Despite the proven efficacy of this intervention, slow implementation and low uptake have limited its impact on population-level HIV incidence. Home HIV testing and counseling (HTC) campaigns have the potential to increase ART uptake among SDC by incorporating couples’ testing and ART referral. Alternatively, because these campaigns have achieved high rates of testing and linkage to care in the general population, there may be little gained by adding activities to target SDC. We estimated the reduction in adult HIV incidence achieved by incorporating ART for SDC into home HTC campaigns in KwaZulu Natal, South Africa (KZN) and southwestern (SW) Uganda. Methods: We constructed dynamic, stochastic, agent-based network models of HIV transmission for each setting, parameterized with HTC and linkage data from 2011-2012. We compared adult HIV incidence after 10 years under 3 scenarios: 1) baseline rates of HIV testing and linkage to care without a home HTC intervention (Current Practice), 2) home HTC campaigns delivered every 3 years, with linkage to ART for eligible persons (Home HTC), and 3) immediate ART for SDC, regardless of CD4, delivered during home HTC campaigns every 3 years, with 90% ART uptake among couples who tested (ART for SDC). In all scenarios, ART eligibility criteria for the general population was based on a CD4 threshold of <350 cells/ul (Ugandan and South African guidelines at the time of data collection). Results: ART for SDC reduced adult HIV incidence by 38% compared to Home HTC: from 1.12 (95% CI: 0.98-1.26) to 0.68 (0.54-0.82) cases per 100 person-years (py) in KZN, and from 0.56 (0.50-0.62) to 0.35 (0.30-0.39) cases per 100 py in SW Uganda (Figure); 1/4 of incident HIV infections were averted over 10 years compared to the Home HTC scenario. The proportion of virally suppressed HIV-infected persons increased by approximately 15%with ART for SDC. Conclusions: Our analyses suggest that using home HTC to identify SDC and deliver immediate ART could avert substantially more new HIV infections than home HTC alone, with a smaller number needed to treat to prevent new HIV infections. Scale-up of home HTC will not diminish the effectiveness of targeting SDC for treatment. Home HTC programs should invest resources to increase rates of couples’ testing, disclosure, and linkage to care.

456

CROI 2016