Share

Determinants of long-term HIV infections at point of HIV diagnosis in the Kingdom of Eswatini: results from the HIV-1 recent infection surveillance program

Title
Determinants of long-term HIV infections at point of HIV diagnosis in the Kingdom of Eswatini: results from the HIV-1 recent infection surveillance program
Presenter
Suzue Saito
Authors
S. Saito * (1,2), A. Pasha (1), D. Sibandze (3), L. Dube (3), E. Poirot (1), M. Arons (4), S. Simelane (5), V. Lokotfwako (3), N. Mahlalela (3), C. Ngcamphalala (5), R. Sahabo (5), M. Pasipamire (6), H. Nuwagaba-Biribonwoha (2,5)
Institutions
(1) ICAP at Columbia University, Mailman School of Public Health, New York, United States, (2) Columbia University, Department of Epidemiology, Mailman School of Public Health, New York, United States, (3) Ministry of Health, Government of the Kingdom of Eswatini, Mbabane, Eswatini, (4) CDC Atlanta, Atlanta, United States, (5) ICAP at Columbia University, Mailman School of Public Health, Mbabane, Eswatini, (6) CDC Eswatini, Mbabane, Eswatini

BACKGROUND: Delayed HIV diagnosis continues to be a major obstacle for epidemic control. We examined determinants of long-term (LT) infection among newly diagnosed people living with HIV that participated in the Eswatini HIV-1 recent infection surveillance program.
METHODS: We used data collected during July 2019"­"­'September 2020 for individuals '¥15 years newly diagnosed with HIV in routine HIV testing services who additionally received a rapid test for recent infection (RTRI). LT infection was defined as a RTRI LT result, which likely indicates infection of '¥12 months. Bivariate and stepwise backward multivariate logistic regressions were run to determine adjusted odds ratios (aORs) between a range of sociodemographic, geographic, and health risk factors and LT infection. Python 3.7 was used for the analyses.
RESULTS: Of 6,171 newly diagnosed clients from 113 testing facilities, 5,502 (89.2%) had LT infection and 669 (10.8%) had a recent infection. In adjusted analyses, male sex (aORvs.non-pregnant female: 2.00; 95% CI: 1.59 ' 2.51), clients who were >25 years of age (see age-group-specific aORs in Figure 1), residence in the Hhohho region (aORvs.Lobombo: 1.32; 95%CI: 1.03 ' 1.70), clients who do not report testing history (aORvs.tested before: 1.86; 95%CI: 1.55 ' 2.23) and clients who reported no use of a HIV self-test kit in the past 12 months (aORvs.use of HIV self-test kit): 1.44; 95%CI: 1.06 ' 1.94 ) were more likely to have LT infection. Distance traveled for HIV diagnosis, marital status, education, injection drug use, number of partners in the past 12 months and experiencing forced sex in the past 12 months were not associated with LT infection.


CONCLUSIONS: Recent infection surveillance can sharpen program response to accelerate epidemic control. Further promotion of frequent HV testing, including through expanded access to self-test kits, particularly among men and older individuals, is critical to reduce delayed HIV diagnosis'a major impediment to achieve epidemic control.