Evaluation of the Relationship between Medication Adherence and Tacrolimus Coefficient of Variation

Abstract

Introduction: High tacrolimus intrapatient variability is associated with poor outcomes following transplantation. A commonly hypothesized cause of this variability is medication non-adherence, but this has not been conclusively demonstrated. Research Question: The purpose of this study was to evaluate the relationship between medication adherence and tacrolimus intrapatient variability. Design: This was a retrospective cohort study of kidney transplant recipients. Adherence was assessed at the 12-month clinic visit as a composite of patient self-report, pharmacist assessment, and lab monitoring frequency. Tacrolimus intrapatient variability was calculated as the coefficient of variation (CV). Linear regression and receiver operating curve (ROC) analysis were used to assess the relationship between adherence and CV. Results: Nonadherence was identified in 37.5% of patients. The median CV was 27.1% for adherent patients and 29.8% for non-adherent patients (P = 0.051). In the multivariable analysis, the only significant predictor of CV was the incidence of dose changes (P = 0.002). ROC analysis demonstrated poor discriminant power with an AUC of 0.597. Discussion: The results fail to support a clinically meaningful relationship between medication adherence and tacrolimus CV. There is very little evidence at this time that adherence is the primary contributing factor to tacrolimus intrapatient variability and, by extension, that CV can be used as a surrogate for adherence.