PCV97 IS CARDIOVASCULAR MEDICATION USAGE A MARKER FOR CARDIOVASCULAR RISK? POTENTIAL IMPLICATION FOR SMART-EDIT FORMULARY MANAGEMENT STRATEGIES

Abstract

In the current statin landscape, a need has emerged for simple and less data-intensive tools to identify patients at risk for cardiovascular events and to appropriately identify patients for higher-efficacy statin use. To examine the feasibility of a pharmacy-based algorithm as a marker of cardiovascular risk to guide appropriate statin selection at the point of service. A retrospective analysis using an integrated database of over 70 pooled health plans was conducted for the period 2003 to mid-2010. Patients on at least one cardiovascular medication of interest (N= 4,276,418) and ≥18 years old were matched by age and gender to control subjects (not on cardiovascular medications [N= 3,046,021]). Cardiovascular medications included were antihypertensive, antidiabetic, antianginal, antiarrhythmic, anticoagulant, non-statin lipid modifier, and antiplatelet drugs. Subjects' cardiovascular risk status (high/moderate/low) was assessed according to the current National Cholesterol Education Program guidelines, from records of diagnoses and procedures. Relative risks were calculated for the likelihood of being at increased risk (high or moderate risk status) as the ratio of proportions of patients on cardiovascular medication versus controls who were at increased risk. The study population had a mean age of 53 years. Overall, a patient receiving any cardiovascular medication was 2.4 times more likely to be at increased risk (95%CI 2.39,2.40) and was 2.7 times more likely to experience a cardiovascular event (95%CI 2.73,2.74) than a subject not receiving any cardiovascular medication. Additionally, it was observed that 27% of the total patients in the health plans who were on cardiovascular medications accounted for 71.5% of all cardiovascular events in that group. A pharmacy algorithm based on usage of various cardiovascular medications can be used to identify increased cardiovascular risk. These initial findings demonstrate the potential of using widely available cardiovascular prescription data in designing statin formulary management strategies.