Discordance Between Biomarker-Confirmed Antiretroviral Therapy and Self-Reported HIV Status Among People Living with HIV in Zambia and South Africa: A Secondary Analysis of HPTN 071 (PopART)

Background: Misclassification of HIV status in population-based surveys remains a critical barrier to accurate surveillance and program evaluation. Self-reported HIV status may diverge from objective measures, particularly among individuals receiving antiretroviral therapy (ART). We used biomarker-confirmed antiretroviral (ARV) drug detection to assess the prevalence and correlates of discordance between self-reported HIV status and biologic evidence of HIV treatment among people living with HIV (PLHIV) in Zambia and South Africa. Methods: We conducted a secondary analysis of the HPTN 071 (PopART) cluster-randomized trial. At the 24-month survey visit, participants underwent HIV testing and laboratory assessment for ARV drugs in plasma. We defined discordant self-report (hereafter "non-disclosure") as reporting HIV-negative or unknown status among individuals with ARV drugs detected. We estimated the prevalence of non-disclosure, compared prevalence by study arm, and used modified Poisson regression to identify associated factors. We also examined whether non-disclosure was associated with viral suppression (<400 copies/mL). Results: Among 3,240 PLHIV with ARV drugs detected, 552 (17.0%) did not report an HIV-positive status--indicating that nearly one in six individuals on ART were misclassified by self-report. Non-disclosure did not differ between intervention and control arms (adjusted relative risk [aRR]: 1.03; 95% CI: 0.67-1.58). Non-disclosure was more common among younger individuals (age 18-24 years: aRR 2.30; 95% CI: 1.66-3.19), men (aRR: 1.39; 95% CI: 1.07-1.79), and those in formal employment (aRR: 1.42; 95% CI: 1.06-1.90). Individuals reporting condomless sex at last encounter were also more likely not to disclose (aRR: 1.59; 95% CI: 1.31-1.92). Viral suppression was high overall (93.7%) and did not differ by disclosure status (aRR: 1.06; 95% CI: 0.74-1.52). Conclusion: A substantial proportion of PLHIV receiving ART did not report a known HIV-positive status, highlighting important discordance between biomarker evidence and self-reported data. Despite high levels of viral suppression, these individuals remain "hidden" from routine surveillance, with implications for estimating HIV diagnosis and treatment coverage. Strategies that incorporate objective measures alongside self-report, and that address social and structural barriers to disclosure, are essential to improve the accuracy of HIV surveillance and guide effective public health responses.