Predictors of Persistent HIV Viremia Among Patients in Hyper-Endemic Regions in South-Central Uganda

Photo taken in Dhaka, Bangladesh
The study authors gathered data on individual HIV viral load trajectories at the population-level following the introduction of universal test and treat in sub-Saharan Africa.

Researchers observed that a subset of patients living with HIV who were younger, mobile men remained persistently viremic during the roll-out of universal test and treat (UTT) introduction in sub-Saharan Africa. These findings were published in The Journal of Infectious Diseases.

This study analyzed data from the Rakai Community Cohort Study. Information was collected from adults living in 4 fishing communities in south-central Uganda where HIV was hyperendemic. Adults were interviewed about lifestyle, offered a free HIV test, and referred for antiretroviral therapy (ART) when appropriate. This study focused on 4 communities between 2011 and 2017, during which time participants (N=2452) who were HIV positive were assessed for viral load (VL) at 3 time points.

Participants were 56.9% women, 44.3% were aged 15 to 29 years, 63.0% were married, and 55.8% reported having no or 1 sexual partner.

A total of 1106 (45.1%) study participants were lost during follow-up. Those lost were less likely to report having a previous HIV test (P =.036), to use ART (P =.004), or to have VL suppression (<400 copies/mL) at baseline (P <.001).

New or renewed VL suppression was observed among 30.3% of participants, persistent viremia among 16.4%, and lost VL suppression among 2.9%. The average geometric VL was 6443 (interquartile range [IQR], 1804-21,184) copies/mL among those who lost suppression and 15,445 (IQR, 4537-50,892) copies/mL among those with persistent viremia (P <.001).

Durable VL suppression was observed to increase from 29.7% to 67.9% over time and persistent viremia to decrease from 20.8% to 13.3%.

During the final survey, 96.3% of participants self-reported as having HIV. Participants who had persistent viremia were less likely (81.0%) to self-report compared with those who had durable VL suppression (99.5%), new or renewed VL suppression (96.7%), or loss of VL suppression (95.5%).

Among the entire population, individuals with 2 or 3 (risk ratio [RR], 1.33; 95% CI, 1.05-1.70; P <.05) or more than 3 sexual partners (RR, 1.96; 95% CI, 1.46-2.65; P <.05) were associated with persistent viremia. However, these associations were no longer significant after adjusting for covariates.

Among women, persistent viremia was associated with those who had never been married (adjusted RR [aRR], 2.46; 95% CI, 1.39-4.36; P <.05).

Among men, persistent viremia was associated with being a recent (aRR, 2.09; 95% CI, 1.35-3.23; P <.05) or established (aRR, 2.03; 95% CI, 1.30-3.17; P <.05) migrant, being age 15 to 29 years (aRR, 1.94; 95% CI, 1.18-3.19; P <.05) or 30 to 39 years (aRR, 1.67; 95% CI, 1.04-2.68; P <.05), and never being married (aRR, 1.53; 95% CI, 1.02-2.31; P <.05).

This study was limited by the self-reported design of many variables, likely including some recall or reporting biases.

These data indicated a subset of the population of HIV-positive individuals had persistent viremia in a region where HIV was hyperendemic. Persistent viremia was associated with sex, marital status, age, and migration status.


Grabowski MK, Patel EU, Nakigozi G, et al. Prevalence and predictors of persistent HIV viremia and viral rebound following universal test and treat: a population-based study. J Infect Dis. 2021;jiab021. doi:10.1093/infdis/jiab021