Uncovering ART adherence inconsistencies: An assessment of sustained adherence among adolescents in South Africa

Abstract

Introduction: Antiretroviral treatment (ART) adherence rates are lower among adolescents living with HIV (ALHIV) than among adults and children, but more evidence is needed on long-term sustained ART adherence among ALHIV. This study assesses rates of sustained ART adherence in a cohort of adolescents in South Africa.

Methods: A prospective cohort of adolescents (10-19 years) living with HIV (baseline sample N = 1 046, 55% female, mean age 13.6) in the Eastern Cape Province in South Africa were interviewed at baseline (2014-15) and followed-up twice (2015-16, 2017-18). All adolescents ever initiated on treatment in 52 government health facilities were traced (with 90% uptake, 94% retention at Wave 2, and 97% retention at Wave 3, 3.4% mortality) and their clinic records were extracted where available. We investigate sustained ART adherence among adolescents interviewed at all three waves of data collection (N = 933). To quantify adherence at each study wave, we used self-reported past-week adherence (including weekdays and weekends). Self-reported adherence was validated using HIV-1 RNA viral load (>50 copies/mL cut-off) reported in clinic records, in a random-intercept logistic regression.

Results and discussion: At baseline, approximately 66% (N = 615) of adolescents reported past-week ART adherence, and of these 45.3% reported adherence at both baseline and follow-up. Only 37.1% of the sample reported sustained past-week ART adherence over the three waves of the study. Most adolescents (N = 587, 62.9%) report inconsistent adherence across time (including 6.4% disengaged from care). Older (P = 0.007) and adolescents with horizontally acquired HIV (P = 0.002) were more likely to report inconsistent adherence across time. Controlling for socio-demographic characteristics, past-week adherence was associated with non-detectable viral load (aOR 1.72, 95%CI 1.14-2.59, P = 0.009). Overall, of the adolescents with viral load measurements at study Wave 1 and Wave 2, 50.6% maintained undetectable viral load for the preceding year.

Conclusions: Adolescents living with HIV reported very low rates of sustained ART adherence. Adherence reported at a single time may mask high rates of variability in adherence over time. These findings highlight the urgent need for enhanced and effective interventions to assist ALHIV with ART adherence through the challenging years of adolescence.

Keywords: ARV; adherence; adolescents; cohort studies; viral suppression.

Figure 1

Past‐week cumulative sustained adherence rates among adolescents living with HIV retained across the three waves (N=933). Notes: Proportions were computed based on the same denominator (N=933) participants interviewed at all waves. 65.9% of participants who reported past‐week adherence at Wave 1, 45.3% of participants reported past‐week adherence at both Wave 1 and Wave 2 as a proportion on (N=933). Lastly, we computed the number of participants who reported past‐week adherence at all three waves (37.1%) as a fraction of (N=933) which is the proportion defined as sustained adherence.