The relationship between self-reported adherence and virological outcomes among adolescents on antiretroviral therapy in Ghana

BACKGROUND: Adolescents living with HIV demonstrate suboptimal adherence to antiretroviral therapy compared with adults. This significantly impedes progress toward achieving the UNAIDS 95-95-95 targets. Understanding the relationship between self-reported adherence and virological outcomes among this population is essential for guiding treatment monitoring strategies and intervention development in resource-limited settings. METHODS: A cross-sectional study was conducted among 515 adolescents aged 10 to 19 years living with HIV and receiving ART at seven health facilities in the Ashanti Region of Ghana between August and November 2023. Self-reported adherence was measured using the adapted AIDS Clinical Trials Group 4-day recall questionnaire combined with items from the LifeWindows Information-Motivation-Behavioral Skills adherence questionnaire. Viral load data were extracted from medical records. Spearman's rank correlation was used to examine the relationship between adherence rates and viral load outcomes. RESULTS: The mean adherence rate was 92% (SD = 18.93). Males comprised 49.5% of participants and females 50.5%, with a mean age of 15.44 years (SD = 2.62). Most participants (67.9%) lived with parents, and 98% had low to middle economic status. The mean number of missed doses showed an ascending pattern across the four-day recall period: 0.056 (Day 1), 0.075 (Day 2), 0.077 (Day 3), and 0.093 (Day 4). Spearman's rank correlation revealed a weak negative correlation between adherence rate and viral load (rs = -0.052, p = 0.235, 95% CI: -0.141 to 0.037), which was not statistically significant. Among participants, 323 (62.7%) achieved viral suppression, while 192 (37.3%) did not. CONCLUSIONS: Despite high self-reported adherence levels, the weak and non-significant correlation with virological outcomes raises important validity concerns regarding self-reported adherence as a proxy measure for viral suppression among adolescents in this setting. The temporal mismatch between four-day adherence recall and longer-term viral load measurements may partly explain this discordance. Findings underscore the need for objective adherence measures, routine viral load monitoring, and adolescent-specific interventions that address psychosocial barriers affecting both adherence behavior and its biological impact.