CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

1146 COST-EFFECTIVENESS OF REGULAR HIV SCREENING FOR YOUNG MEN WHO HAVE SEX WITH MEN Anne M. Neilan 1 , Alexander Bulteel 1 , Kenneth Freedberg 1 , Sybil Hosek 2 , Raphael J. Landovitz 3 , Rochelle P. Walensky 1 , Craig M. Wilson 4 , Andrea L. Ciaranello 1 1 Massachusetts General Hospital, Boston, MA, USA, 2 Cook County Health & Hospitals System, Chicago, IL, USA, 3 University of California Los Angeles, Los Angeles, CA, USA, 4 University of Alabama at Birmingham, Birmingham, AL, USA Background: Of new HIV diagnoses among US youth, 81% occur among young men who have sex with men (YMSM). Using data from Adolescent Trials Network (ATN) studies 110/113 of high-risk US YMSM ages 15-22, we examined the clinical impact, cost, and cost-effectiveness of 4 HIV screening strategies for high-risk YMSM starting at age 15. Methods: We simulated a cohort of high-risk HIV-uninfected 14-year-old MSM in the US who faced age-specific risks of HIV infection (0.91-6.41/100,000PY, peak incidence ages 15-18) based on ATN 110/113 (observational; incident infections despite PrEP adherence of 34-56%). We modeled HIV screening ($36/ test) every 3 years, annually, biannually, and quarterly beginning at age 15, each in addition to current US screening practices for YMSM (16-53% screened at least once by ages 15-22). We used published YMSM HIV care continuum data, including screen acceptance (80%), linkage-to-care/antiretroviral therapy (ART) initiation (76%), disease progression, and HIV care costs. Outcomes included CD4 count at diagnosis, the HIV care continuum (proportions HIV-diagnosed, linked to care, retained in care, and virologically suppressed), one generation of secondary HIV transmissions, life expectancy, lifetime costs, and incremental cost-effectiveness ratios (ICER) in $/year-of-life saved (YLS) from the healthcare system perspective. In sensitivity analyses, we varied HIV incidence, screening and linkage rates, and costs. Results: All screening strategies beginning at age 15 diagnosed greater proportions of lifetime infections compared to current practice alone (81-99% vs. 35%). Compared to the next most effective strategy, quarterly screening beginning at age 15 was cost-effective ($84,000/YLS) by US standards (<$100,000/YLS) (Table). Including just first-generation HIV transmissions averted, the ICER was markedly lower ($20,900/YLS). These results were most sensitive to current HIV screening practice rates and linkage-to-care/ART initiation. If HIV incidence peaked at older ages, an older starting age for HIV screening had more favorable cost-effectiveness outcomes; if absolute HIV incidence was lower, less frequent screening was more favorable. Conclusion: For high-risk US YMSM, quarterly HIV screening beginning at age 15, compared to less frequent screening beginning at age 15, would improve clinical outcomes and be cost-effective. To optimize clinical outcomes, screening should begin at or after the peak of population-specific HIV incidence.

Kenya, hypothesizing that retesting would be cost-effective when compared to initial HIV testing alone due to health benefits accrued by mother and child. Methods: We used TreeAge software to model a decision tree with the initial decision node comparing the alternative HIV testing strategies (a single antenatal HIV test early in pregnancy, or the initial antenatal HIV test plus a repeat HIV test three months later) and the successive chance nodes representing antepartum possibilities including maternal seroconversion, maternal ART uptake, fetal HIV acquisition, facility delivery, and mortality during delivery. At delivery of the infant, each branch culminates in a state- transition model that jointly tracks the mother-infant pair in one-month cycles for a ten-year horizon (Figure 1). All inputs were drawn from the literature and were varied across their range or distribution in univariate and probabilistic sensitivity analyses. Results: In the base case, the retesting strategy was cost-effective for the Kenyan setting at $1,098 per quality-adjusted life year (QALY) saved, yielding fewer infant HIV infections during pregnancy and breastfeeding (n=504 and 253, respectively), infant deaths (n=30), and maternal deaths (n=178) per 100,000 women. Results were sensitive to low cumulative incidence of HIV during pregnancy and monthly cost of maternal ART (thresholds of 1% and $45, respectively). Probabilistic sensitivity analyses confirmed the base-case analysis. Conclusion: This modeling study indicates that repeat HIV testing is likely cost- effective and results in fewer infant HIV infections. In the “test and treat era,” in which immediate ART is recommended for all HIV infected persons, retesting for HIV in pregnant women not only improves maternal health outcomes but may also contribute to the elimination of perinatal HIV transmission in Kenya.

Poster Abstracts

1148 PREFERENCES AND WILLINGNESS-TO-PAY FOR BLOOD AND ORAL-FLUID HIV SELF-TESTS IN KENYA Kristen Little 1 , Nicholas L. Wilson 2 , Patrick Alyward 1 , Hildah Essendi 3 1 Population Services International, Washington, DC, USA, 2 Reed College, Portland, OR, USA, 3 Population Services International, Nairobi, Kenya

Corrected data and updated results for this study, as presented at CROI 2018, are available in the electronic poster for abstract 1148 at http://www.croiconference.org/abstracts/search-abstracts.

1147 REPEAT HIV TESTING DURING PREGNANCY IN KENYA: AN ECONOMIC EVALUATION Anna Joy Rogers, Janet M. Turan , Ellen F. Eaton, Stephen T. Mennemeyer University of Alabama at Birmingham, Birmingham, AL, USA Background: Repeat HIV testing during late pregnancy may identify women who seroconvert after an initial negative HIV test early in pregnancy, allowing these women to adopt lifelong antiretroviral therapy (ART) for the sake of their own health as well as to prevent mother-to-child transmission of HIV. We evaluated the cost-effectiveness of repeat HIV testing during late pregnancy in

CROI 2018 444