OBJECTIVES: To evaluate three measures of compliance by applying established methodologies for each to one population. METHODS: Prescription claims data for 1164 patients receiving a chronic lipid-lowering medication (a combination product versus a statin) in 2002-2003 were obtained; MPR's, PDC's, and ELPT's were calculated. Because each method has different inclusion criteria, the measures were analyzed for three different subsets of the overall population. A second set of analyses was conducted on a fourth subset of patients (those included in all three of the other analyses) to evaluate the impact of inclusion criteria on the endpoints. Logistic regression analyses were conducted for each measure to determine the probability of being categorized as ‘compliant’ by established definitions. RESULTS: The results differed depending on the methodology used, suggesting each measure evaluates a different aspect of overall compliance. MPR analysis demonstrated higher overall compliance rates with the combination product versus statins (p=0.001). PDC analysis demonstrated lower compliance rates with the combination product versus statins (2nd quarter p=0.003, 4th quarter p=0.011). ELPT analysis also demonstrated lower compliance rates for the combination product versus statins (6th month p=0.022, 9th month p=0.008). CONCLUSIONS: Each measure evaluates a different aspect of overall compliance. MPR methodology, which measures the degree of prescription-filling during treatment and does not factor in duration of treatment, may be considered a surrogate measure for adherence. ELPT's are a pure measure of persistence, useful even in sub-optimally adherent patients. PDCs are a composite measure of adherence and persistence where the degree of prescription-filling is reported quarterly in an attempt to describe patient behavior longitudinally.