Select Abstracts from Elements of Success 2004: An International Conference on Adherence to Antiretroviral Therapy

Abstract

Background: While the notion of a “dose” is a basic construct in treatment research and practice, neither the parameters used to characterize a missed dose, nor what should be done to correct a missed dose and maintain adherence are widely shared by patients, nor agreed upon between clinicians and patients. This study examines clinician and patient understandings of the parameters of a missed dose and of what should be done when a dose is missed. Methods: Sixty African Americans taking [highly active antiretroviral therapy (HAART)] (20 women, 40 men) and 20 of their physicians and nurses, all of whom are enrolled in a longitudinal study on HAART adherence, were interviewed. Measures included: self-definition of a medication dose, missed dose, and corrective actions; adherence (CPCRA three-day pilltaking recall, three-month self assessment, Medical Outcomes Study adherence scale [Kravitz et al, 1993] and a visual analogue), and biomarkers of adherence (viral load and CD4 counts). Open-ended interviews and standardized techniques were used. Results: We found consistent systematic definitions of a missed dose and of dose management. Three definitions marking different sets of parameters for a missed dose were identified: medication not taken at the exactly same time every day (15 percent of clinicians, 20 percent of patients); medication taken outside a threeto four-hour window each day (75 percent of clinicians, 60 percent of patients); medication not taken some time within the waking day (10 percent of clinicians, 20 percent of patients). Clinicians and patients agreed that one should not double up on doses to make up for a missed dose but expressed a range of acceptable means to address a missed dose. Conclusions: Efforts to refine the measurement of HAART adherence need to be aware of the threats to validity posed by the range of working definitions of a missed dose. 2 – Adaptive Poisson regression analysis of MEMS adherence data