Share

Impact of a community-wide HIV test and treat intervention on population-level tuberculosis transmission in rural Uganda

Title
Impact of a community-wide HIV test and treat intervention on population-level tuberculosis transmission in rural Uganda
Presenter
Carina Marquez
Authors
C. Marquez * (1), M. Atukunda (2), J. Nugent (3), E. Charlebois (1), G. Chamie (1), F. Mwangwa (2), E. Ssemmondo (2), J. Kironde (2), J. Kabami (2), A. Owaraganise (2), E. Kakande (2), R. Abbott (1), B. Ssekaynzi (2), J. Ayieko (4), T. Ruel (1), D. Kwariisima (2), M. Kamya (2,5), M. Petersen (6), D. Havlir (1), L. Balzer (3)
Institutions
(1) University of California, San Francisco, San Francisco, United States, (2) Infectious Diseases Research Collaboration, Kampala, Uganda, (3) University of Massachusetts- Amherst, Amherst, United States, (4) Kenya Medical Research Institute, Nairobi, Kenya, (5) Makerere University School of Medicine, Kampala, Uganda, (6) University of California, Berkeley, Berkeley, United States

BACKGROUND: HIV and TB are linked epidemics. We tested whether a community-wide universal HIV testing and treat strategy would reduce population-level TB transmission in the SEARCH trial.
METHODS: SEARCH was 32-community cluster-randomized trial conducted in rural Uganda and Kenya from 2013-2017 (NCT:01864603). Intervention communities received population-level HIV test and treat; control communities received enhanced standard-of-care. We measured incident TB infection in a nested cohort of children and adults ages (>5 years) in in 4 intervention and 5 control communities (8 and 10 parishes, respectively) in Eastern Uganda. All people in the cohort had a negative tuberculin skin test (TST) at baseline (induration <10mm or <5 mm among PWH); the cohort was enriched with PWH. Incident TB infection (primary outcome) was defined as TST conversion from negative to positive one year after baseline TST (2015-2017). Incident TB was compared between arms, adjusted for sampling and participation, and parish-level drivers of incident TB infection and accounting for clustering using Targeted Maximum Likelihood Estimation.
RESULTS: The TST negative cohort was comprised of 3,242 persons in the intervention and 4,125 persons in the control arms; 2,922 (39.7%) were children (5-11 years) and 4,445 (60.3%) youth and adults (ages 12+ years). Within the cohort 53.3% were women, and PWH comprised 16.6% of adults, 1.1% of children. One-year cumulative incidence of TB infections was 16% in the intervention and 22% in the control; the population-level intervention reduced risk of incident TB infection by 27% (aRR of 0.73; 95% CI: 0.58-0.93, one-sided p=0.007). The effect was largest among children aged 5-11 years (Figure 1).
CONCLUSIONS: A universal HIV test and treat intervention reduced incident TB infection, a marker of population-level TB transmission. Investments in community-level HIV interventions have direct impacts on HIV and broader population-level benefits, including reductions in TB.

Figure 1: