Share
Title
Accuracy of indirect adherence measures among adolescent men who have sex with men (aMSM) and transgender women (aTGW) in Brazil: a longitudinal analysis
Presenter
Diana Reyna Zeballos Rivas
Authors
D.R. Zeballos Rivas1, L. Magno2, F. Soares1, J.A. Pinto Junior3, L. Amorim4, D. Greco5, A. Grangeiro6, I. Dourado1, The PrEP1519 study group
Institutions
1Universidade Federal da Bahia, Instituto de Saúde Coletiva, Salvador, Brazil, 2Universidade do Estado da Bahia, Departamento de Ciências da Vida, Salvador, Brazil, 3Universidade Federal Fluminense, Departamento de Estatística, Niterói, Brazil, 4Universidade Federal da Bahia, Instituto de Matemática e Estatística, Salvador, Brazil, 5Universidade Federal de Minas Gerais, Faculdade de Medicina, Belo Horizonte, Brazil, 6Universidade de São Paulo, Departamento de Medicina Preventiva da Faculdade de Medicina, São Paulo, Brazil

BACKGROUND: Oral PrEP has been offered free of charge in the health services of the Brazilian National Health System since 2017. In 2022, the updated Brazilian PrEP guidelines included adolescents who are at high HIV risk. Adherence evaluation is crucial for monitoring PrEP users, and indirect measures are used during clinical follow-up. We aimed to assess the accuracy of indirect PrEP adherence measures using longitudinal data and drug concentrations in dried blood spots (DBS).
METHODS: PrEP1519 is a prospective, multicenter, open-label PrEP demonstration cohort study conducted with aMSM and aTGW aged 15-19. A diagnostic accuracy analysis was performed comparing a direct measure of adherence - tenofovir diphosphate (TFV-DP) concentrations in DBS- with three indirect measures: medication possession ratio (MPR), pill count, and self-report. Generalized estimating equations (GEE) were used to assess the accuracy of each indirect measure through the computation of the Area Under The Curve (AUC) and their 95% Confidence Interval (95%CI) using the ROC (Receiver Operating Characteristics) curve analysis. We considered the repeated (longitudinal) measures for the same participants, and we used the reference of protective TFV-DP levels =800fmol/punch equivalent to 4 pills weekly or more. We also analyze the combination of different methods to assess adherence.
RESULTS: We selected a sample of 302 DBS corresponding to 188 participants. Most were aMSM (78.7%), 18–19 years old (80.3%), and non-whites (72.9%). Matched data for MPR was 294, for self-report 274, and pill count 104. AUC was 0.59 (95%CI: 0.53 – 0.66) for MPR, 0.69 (95%CI: 0.58 – 0.80) for pill count, and 0.75 (95%CI: 0.69 – 0.81) for self-report. We observed that the combination of the three methods for evaluating adherence yielded an AUC= 0.72 (95%CI: 0.62 – 0.83) and we obtained a better performance with MPR and self-report combined (AUC = 0.77; 95%CI: 0.70 – 0.83).
CONCLUSIONS: The three indirect measures discriminated participants with drug levels equivalent to four pills weekly at different time points, indicating that these are valuable methods for monitoring PrEP use. Using more than one indirect method for adherence yields better patient adherence information.

Download the e-Poster (PDF)