SEARCH Human Immunodeficiency Virus (HIV) Streamlined Treatment Intervention Reduces Mortality at a Population Level in Men With Low CD4 Counts

Abstract

Background: We tested the hypothesis that patient-centered, streamlined human immunodeficiency virus (HIV) care would achieve lower mortality than the standard treatment model for persons with HIV and CD4 ≤ 350/uL in the setting of population-wide HIV testing.

Methods: In the SEARCH (Sustainable East Africa Research in Community Health) Study (NCT01864603), 32 communities in rural Uganda and Kenya were randomized to country-guided antiretroviral therapy (ART) versus streamlined ART care that included rapid ART start, visit spacing, flexible clinic hours, and welcoming environment. We assessed persons with HIV and CD4 ≤ 350/uL, ART eligible in both arms, and estimated the effect of streamlined care on ART initiation and mortality at 3 years. Comparisons between study arms used a cluster-level analysis with survival estimates from Kaplan-Meier; estimates of ART start among ART-naive persons treated death as a competing risk.

Results: Among 13 266 adults with HIV, 2973 (22.4%) had CD4 ≤ 350/uL. Of these, 33% were new diagnoses, and 10% were diagnosed but ART-naive. Men with HIV were almost twice as likely as women with HIV to have CD4 ≤ 350/uL and be untreated (15% vs 8%, respectively). Streamlined care reduced mortality by 28% versus control (risk ratio [RR] = 0.72; 95% confidence interval [CI]: .56, .93; P = .02). Despite eligibility in both arms, persons with CD4 ≤ 350/uL started ART faster under streamlined care versus control (76% vs 43% by 12 months, respectively; P < .001). Mortality was reduced substantially more among men (RR = 0.61; 95% CI: .43, .86; P = .01) than among women (RR = 0.90; 95% CI: .62, 1.32; P = .58).

Conclusions: After population-based HIV testing, streamlined care reduced population-level mortality among persons with HIV and CD4 ≤ 350/uL, particularly among men. Streamlined HIV care models may play a key role in global efforts to reduce AIDS deaths.

Keywords: advanced HIV-disease; mortality; population-based HIV testing; streamlined care.

© The Author(s) 2021. Published by Oxford University Press for the Infectious Diseases Society of America. All rights reserved. For permissions, e-mail: journals.permissions@oup.com.

Figures

Figure 1.

Participant flow chart. Abbreviation: HIV, human immunodeficiency virus.

Figure 2.

Cumulative probability of initiating ART by 1, 12, and 24 months among PWH with CD4 ≤ 350 cells/ml not on ART at baseline (N = 1283), stratified by sex (680 women, 603 men). Estimated for each community censored at outmigration and treated death as a competing risk; ART initiation between arms was compared using community-level TMLE. Error bars indicate 95% CIs for each sex-arm strata. Abbreviations: ART, antiretroviral therapy; CI, confidence interval; PWH, people with human immunodeficiency virus; TMLE, targeted maximum likelihood estimation.

Figure 3.

Difference in 3-year risk of mortality among PWH with CD4 ≤ 350 cells/mL (N = 2973), overall and stratified by sex (1384 women, 1589 men). Cumulative risk of death was estimated for each community censoring at outmigration; mortality between arms was compared using community-level TMLE. Abbreviations: CI, confidence interval; PWH, people with human immunodeficiency virus; RR, risk ratio; TMLE, targeted maximum likelihood estimation.

Figure 4.

Multivariable predictors of mortality among PWH with CD4 ≤ 350 cells/mL (N = 2973), stratified by intervention arm (N = 1572 intervention arm, 1401 control arm). Adjusted relative risks were estimated using individual-level TMLE. Abbreviations: ART, antiretroviral therapy; PWH, people with human immunodeficiency virus; TMLE, targeted maximum likelihood estimation.