A Stepped-Wedge, Cluster-Randomized, Multisite Study of Text Messaging Plus Peer Navigation to Improve Adherence and Viral Suppression Among Youth on Antiretroviral Therapy

Abstract

BACKGROUND: To address the need for improved virologic suppression among youth living with HIV (YLH) on antiretroviral treatment (ART), we evaluated peer navigation plus TXTXT daily text message ART reminders. SETTING: YLH aged 15-24 years on ART for at least 3 months at 6 research sites in 4 Nigerian cities. METHODS: Using a stepped-wedge design, cluster 1 was nonrandomized, whereas clusters 2 and 3 were randomized to sequences of routine care (control period) and 48 weeks of the combination intervention (intervention period). The primary end point was viral suppression (HIV-1 RNA <200 copies/mL) at week 48 of the intervention. Secondary end points included adherence measured by self-report ( 90% considered adherent). Post hoc analysis assessed virologic control at <50 copies per milliliter and <1000 copies per milliliter. Generalized estimating equations determined the difference between intervention and control periods in the intention-to-treat population. RESULTS: We enrolled 558 YLH and followed 541 over time, mean age 18 years, 53.8% female, 71.7% perinatally infected, and 38.6% virologically nonsuppressed at enrollment. For the primary end point, the intervention periods displayed a small, nonsignificant increase in viral suppression < 200 copies per milliliter [odds ratio (OR) = 1.16 (0.88-1.54), P = 0.297]. There was a significant effect of the combination intervention on virologic control <1000 copies per milliliter (OR = 1.42 [1.03-1.94], P = 0.030). Self-reported adherence also improved (OR = 2.07 [1.46-2.95], P < 0.001). CONCLUSIONS: Peer navigation plus daily text message ART reminders demonstrated limited benefit among ART-experienced, predominantly perinatally infected YLH, with no significant effect on viral suppression below 200 copies per milliliter despite improvement in self-reported adherence.