Nonadherence and unsuppressed viral load across adolescence among US youth with perinatally acquired HIV

Abstract

Objective: To identify factors associated with nonadherence and unsuppressed viral load across adolescence among youth with perinatally acquired HIV.

Design: Longitudinal study at 15 US clinical sites.

Methods: Self-reported antiretroviral medication nonadherence (any missed dose, past week) and unsuppressed viral load (HIV RNA > 400 copies/ml) were assessed annually. Individual, caregiver, social, and structural factors associated with nonadherence and unsuppressed viral load were identified by age (years): 8-11 (preadolescence), 12-14 (early adolescence), 15-17 (middle adolescence), and 18-22 (late adolescence/young adulthood), utilizing multivariable generalized linear mixed effects models.

Results: During a median 3.3-year follow-up, 381 youth with perinatally acquired HIV contributed viral load measurements and 379 completed 1190 adherence evaluations. From preadolescence to late adolescence/young adulthood, prevalence of nonadherence increased from 31 to 50% (P < 0.001); prevalence of unsuppressed viral load increased from 16 to 40% (P < 0.001). In adjusted analyses, in pre, middle, and late adolescence/young adulthood, perceived antiretroviral side effects were associated with nonadherence. Additional factors associated with nonadherence included: in preadolescence, using a buddy system (as an adherence reminder); in early adolescence, identifying as black, using buddy system; in middle adolescence, CD4% less than 15%, unmarried caregiver, indirect exposure to violence, stigma/fear of inadvertent disclosure, stressful life events. Associations with unsuppressed viral load included: in early adolescence, youth unawareness of HIV status, lower income; in middle adolescence, perceived antiretroviral side effects, lower income; in late adolescence/young adulthood, distressing physical symptoms, and perceived antiretroviral side effects.

Conclusion: Prevalence of nonadherence and unsuppressed viral load increased with age. Associated factors varied across adolescence. Recognition of age-specific factors is important when considering strategies to support adherence.

Conflict of interest statement

CONFLICTS OF INTEREST: None of the authors declares conflicts of interest.

Figures

Figure 1:

Prevalence of Nonadherence and Unsuppressed Viral Load by Age in PHACS AMP. N=1190 self- and/or caregiver-reported adherence evaluations of 379 youth; N=1257 viral load measurements from 381 youth PHACS AMP=Pediatric HIV/AIDS Cohort Study Adolescent Master Protocol; CI=Confidence Interval; VL=Viral load

Figure 2:

Adjusted Predicted Probabilities (with 95% Confidence Intervals) of Self-reported Nonadherence among Youth with vs. without Exposure to Violence in PHACS AMP. PHACS AMP=Pediatric HIV/AIDS Cohort Study Adolescent Master Protocol Model adjusted for sex, race, caregiver living arrangement, and household annual income. P-value was for testing the interaction between witnessed or experienced violence and age at self-reported nonadherence evaluation from the adjusted models. Predicted probabilities were among black males who lived in a household with annual income >$10000–40000 and whose caregiver was living with a partner. The data table under the plot presents the number of participants with a specific characteristic and the number(%) with nonadherence (by self or caregiver report) among participants with the specific characteristic, within each group from unadjusted analysis.