Share

Geospatial clusters of recent HIV-1 infection, Malawi, October 2019 - March 2020

Title
Geospatial clusters of recent HIV-1 infection, Malawi, October 2019 - March 2020
Presenter
Carson Telford
Authors
C. Telford * (1), Z. Tessema (1), M. Msukwa (2,3), M. Arons (1), J. Theu (2), F.F. Bangara (2), A. Ernst (4), G. O'Malley (5), R. Bhatkoti (1), T. Dobbs (1), V. Shanmugam (1), A. Kabaghe (6), H. Dale (1), N. Wadonda-Kabondo (6), S. Gugsa (5), G. Bello (2,7), B. Parekh (1), R.W. Shiraishi (1), A. Jahn (3), J. Eaton (8), E. Kim (6), J.L. Tobias (1), K.G. Curran (1), D. Payne (6), A.F. Auld (6).
Institutions
(1) Centers for Disease Control, DGHT, Atlanta, United States, (2) International Training and Education Center for Health (I-TECH), Department of Global Health, University of Washington,, Lilongwe, Malawi, (3) International Training and Education Center for Health (I-TECH), Department of Global Health, University of Washington, Lilongwe, Malawi, (4) Global Strategic Information, Institute for Global Health Sciences, University of California, San Francisco, United States, (5) International Training and Education Center for Health (I-TECH), Department of Global Health, University of Washington, Seattle, United States, (6) Centers for Disease Control, DGHT, Lilongwe, Malawi, (7) Ministry of Health, Department of HIV AIDS, Lilongwe, Malawi, (8) Imperial College London, MRC Centre for Global Infectious Disease Analysis, London, United Kingdom

BACKGROUND: In Malawi, recent HIV-1 infection surveillance was implemented within HIV testing services (HTS). The surveillance system utilizes a point of care (POC) test to detect persons recently infected with HIV, to guide public health response in targeting areas with ongoing transmission.
METHODS: Newly diagnosed clients at 103 facilities, from October 2019 through March 2020, were included. We conducted an analysis of recent infections per facility using a spatial scan statistic in a discrete Poisson model with SaTScanTM to identify clusters of facilities with significantly elevated rates of recent HIV infections. Rates were calculated as the number of recent HIV infections per 100,000 HIV testers at-risk and relative risk (RR) measures were calculated based on observed and expected counts of recent infections inside and outside of each cluster. Using a denominator of recent infections plus total negative HIV tests per facility to simulate an at-risk population, clusters were ranked by probability of occurrence based on log-likelihood, adjusting for sex and age, not allowing cluster overlap, at 'ยค20-kilometers.
RESULTS: Of 9,168 new HIV diagnoses, 304 (3.3%) were recent. Spatial analysis detected (Figure 1) three clusters where the observed rate of recent infections was higher than the expected rate: one in District A among four facilities, with a recent infection rate of 589 (RR: 3.1, p<0.001); the second, in B and C Districts, among four facilities with a rate of 374 (RR: 2.0, p<.02); and the third, in District A, one facility, with a rate of 825 (RR: 4.2, p=0.03).

Figure 1. Geospatial clusters of recent HIV-1 infection among health facilities implementing recent HIV infection surveillance in five districts ' Malawi, October 2019 - March 2020
CONCLUSIONS: These findings describe significantly high rates of recent HIV infection among three geospatial clusters of health facilities in Malawi. Public health responders should investigate facilities within geospatial clusters to identify risk factors for infection, inform interventions to improve HIV program services and interrupt transmission to achieve epidemic control.