Population-level viral suppression among pregnant and postpartum women in a universal test and treat trial

Abstract

Objective(s): We sought to determine whether universal 'test and treat' (UTT) can achieve gains in viral suppression beyond universal antiretroviral treatment (ART) eligibility during pregnancy and postpartum, among women living with HIV.

Design: A community cluster randomized trial.

Methods: The SEARCH UTT trial compared an intervention of annual population testing and universal ART with a control of baseline population testing with ART by country standard, including ART eligibility for all pregnant/postpartum women, in 32 communities in Kenya and Uganda. When testing, women were asked about current pregnancy and live births over the prior year and, if HIV-infected, had their viral load measured. Between arms, we compared population-level viral suppression (HIV RNA <500 copies/ml) among all pregnant/postpartum HIV-infected women at study close (year 3). We also compared year-3 population-level viral suppression and predictors of viral suppression among all 15 to 45-year-old women by arm.

Results: At baseline, 92 and 93% of 15 to 45-year-old women tested for HIV: HIV prevalence was 12.6 and 12.3%, in intervention and control communities, respectively. Among HIV-infected women self-reporting pregnancy/live birth, prevalence of viral suppression was 42 and 44% at baseline, and 81 and 76% (P = 0.02) at year 3, respectively. Among all 15 to 45-year-old HIV-infected women, year-3 population-level viral suppression was higher in intervention (77%) versus control (68%; P < 0.001). Pregnancy/live birth was a predictor of year-3 viral suppression in control (P = 0.016) but not intervention (P = 0.43). Younger age was a risk factor for nonsuppression in both arms.

Conclusion: The SEARCH intervention resulted in higher population viral suppression among pregnant/postpartum women than a control of baseline universal testing with ART eligibility for pregnant/postpartum women.

Conflict of interest statement

Conflicts of Interest: The authors have no competing interests.

Figures

Figure 1:

Estimates of population-level HIV viral suppression among women aged 15-45 years and reporting a current pregnancy or live birth over the prior year by study year in the SEARCH trial. Adjusted for incomplete measurement of HIV status and viral load measurement with targeted maximum likelihood estimation (TMLE); 95% confidence intervals indicated with black vertical lines. Absolute numbers given in the corresponding table.

Figure 2:

Adjusted predictors of HIV viral suppression (HIV RNA