Adherence to antiretroviral therapy in HIV-positive adolescents in Uganda assessed by multiple methods: a prospective cohort study.

Background: The effectiveness of traditional adherence measurements used in adolescent populations is difficult to assess. Antiretroviral (ARV) adherence research among adolescents living with HIV in resource-constrained countries is particularly challenging and little evidence is available.

Objectives: The primary objective of this study was to determine the feasibility of a large-scale, long-term study using electronic adherence monitoring in Uganda. The secondary objective was to compare accuracy of pill count (PC) and self-report (SR) adherence with electronic medication vials (eCAPs™).

Methods: Adolescents receiving ARV therapy at the Joint Clinical Research Centre in Kampala, Uganda, were recruited. ARVs were dispensed in eCAPs™ for 1 year. Person-pill-days (PPDs) [1 day where adherence was measured for one medication in one patient] were calculated and a weighted paired t-test was used to compare the levels of adherence among subjects for three different adherence measurement methods.

Results: Fifteen patients were included: 40% were female, mean age was 14 years, mean baseline CD4+ cell count was 244 cells/μL, and average treatment duration was 9 months at study entry. Overall, 4721 PPDs were observed. Some eCAPs™ required replacement during the study resulting in some data loss. Consent rate was high (94%) but was slow due to age limit cut-points. Overall adherence for SR was 99%, PC was 97% and eCAP™ was 88% (p<0.05 for all comparisons). 93%, 67% and 23% of patients had an adherence of greater than 95% as measured by SR, PC and eCAP™ methods, respectively.

Conclusions: A large-scale adherence study in Uganda would be feasible using a more robust electronic monitoring system. Adherence measurements produced by PCs and self-reporting methods appear to overestimate adherence measured electronically.