Contrasting Measures of Adherence with Simple Drug Use, Medication Switching, and Therapeutic Duplication

Abstract

Multiple measures of adherence have been reported in the research literature and it is difficult to determine which is best, as each is nuanced. Occurrences of medication switching and polypharmacy or therapeutic duplication can substantially complicate adherence calculations when adherence to a therapeutic class is sought. To contrast the Proportion of Days Covered (PDC) adherence metric with 2 variants of the Medication Possession Ratio (MPR, truncated MPR). This study was a retrospective analysis of the North Carolina Medicaid administrative claims data from July 1999 to June 2000. Data for patients with schizophrenia (ICD-9-CM code 295.xx) who were not part of a health maintenance organization, not hospitalized, and not pregnant, taking at least one antipsychotic, were aggregated for each person into person-quarters. The numerator for PDC was defined as the number of days one or more antipsychotics was available and the MPR numerator was defined as the total days' supply of antipsychotics; both were divided by the total days in each person-quarter. Adherence rates were estimated for subjects who used only one antipsychotic, switched medications, or had therapeutic duplication in the quarter. The final sample consisted of 25,200 person-quarters from 7069 individuals. For person-quarters with single antipsychotic use, adherence to antipsychotics as a class was: PDC 0.607, truncated MPR 0.640, and MPR 0.695 (p < 0.001). For person-quarters with switching, the average MPR was 0.690, truncated MPR was 0.624, and PDC was 0.562 (p < 0.001). In the presence of therapeutic duplication, the PDC was 0.669, truncated MPR was 0.774, and MPR was 1.238 (p < 0.001). The PDC provides a more conservative estimate of adherence than the MPR across all types of users; however, the differences between the 2 methods are more substantial for persons switching therapy and prescribed therapeutic duplication, where MPR may overstate true adherence. The PDC should be considered when a measure of adherence to a class of medications is sought, particularly in clinical situations in which multiple medications within a class are often used concurrently.