Association between HDL-C concentration and risk for a major cardiovascular event

OBJECTIVE: To determine the association between baseline HDL-C concentrations and risk of a major cardiovascular event (within 5 years) in a large US claims database. METHODS: A retrospective longitudinal analysis using claims data from the i3 Ingenix LabRx database was conducted. Patients were included if they had complete lipid panel lab results, were continuously enrolled for >or=6 months prior to and >or=12 months following the lab test (index date), and were >or=50 years of age. Cox proportional hazards analysis assessed the association between HDL-C concentrations and risk of a major cardiovascular event within 5 years of the index date, after adjusting for covariates. RESULTS: There was a statistically significant association between HDL-C and risk of a major cardiovascular event. A 0.026 mmol/L (1 mg/dL) increase in HDL-C from baseline was associated with a 1.3% decreased risk of a major cardiovascular event (e.g., a 0.13 mmol/L [5 mg/dL] increase in HDL-C above baseline concentrations was associated with a 6.5% decrease). This association became evident within 1 year of follow-up. CONCLUSIONS: HDL-C concentrations were inversely associated with the occurrence of cardiovascular events within 5 years. Women who were >or=0.26 mmol/L (10 mg/dL) below their target concentrations had cardiovascular risk similar to that of women with baseline ischemic heart disease and hypertension, and men who were >or=0.26 mmol/L (10 mg/dL) below their target concentrations had cardiovascular risk similar to that of men with baseline cardiovascular disease, diabetes, or cerebrovascular disease. Limitations inherent to claims-based analyses must be considered when interpreting these findings, such as the potential for miscoding or incomplete data, and the fact that the presence of a diagnosis code on a medical claim is not positive presence of disease. Furthermore, patients who suffered a fatal cardiovascular event at home during the study period, and therefore did not receive treatment, were not captured in this analysis.