The relationship between directly measured statin adherence, self-reported adherence measures and cholesterol levels in patients with coronary heart disease

We aimed to determine the relationship between statin adherence measured directly, and by self-report measures and serum cholesterol levels. Patients prescribed atorvastatin (N = 373) participated in a cross-sectional study 2–36 months after a coronary event. Self-reported adherence included statin adherence the past week, the 8-item Morisky medication adherence scale (MMAS-8), and the Gehi et al. adherence question. Atorvastatin was measured directly in spot blood plasma by a novel liquid chromatography tandem mass-spectrometry method discriminating adherence (0–1 doses omitted) and reduced adherence (≥2 doses omitted). Participants were unaware of the atorvastatin analyses at study participation. Mean age was 63 (SD 9) years and 8% had reduced atorvastatin adherence according to the direct method. In patients classified with reduced adherence by the direct method, 40% reported reduced statin adherence, 32% reported reduced adherence with the MMAS-8 and 22% with the Gehi question. In those adherent by the direct method, 96% also reported high statin adherence, 95% reported high adherence on the MMAS-8 whereas 94% reported high adherence on the Gehi question. Cohen's kappa agreement score with the direct method was 0.4 for self-reported statin adherence, 0.3 for the Gehi question and 0.2 for the MMAS-8. Adherence determined by the direct method, self-reported statin adherence last week, and the Gehi question was inversely related to LDL-cholesterol levels with a p-value of