Abstract P373: Predictors of Statin Use for Primary Cardiovascular Disease Prevention in Community Health Centers
Background: Guidelines emphasize using cardiovascular (CVD) risk to guide statin prescribing. However, drivers of statin use for primary prevention are unknown. We aimed to identify predictors of statin use in a clinical trial of a CVD risk communication intervention.
Methods: We used randomized trial data from 10 community health centers in Illinois and Arizona that utilized telephone and mailed outreach with individualized CVD risk messages to participants without CVD or diabetes and a 10-year Framingham coronary heart disease (CHD) risk ≥10%. Statin discussion was assessed at 6 months by blinded chart review and statin prescription was assessed at 12 months through the electronic health record. We used logistic regressions to determine predictors of statin discussion and prescription.
Results: We analyzed 646 trial participants: 89% male, mean age 59.8 ± 9.5 years. Statin discussion occurred in 19% (125/646) and statin prescription occurred in 12% (78/646) of participants. 10-year CHD risk was not associated with statin discussion (OR 1.10 per 1 SD increase, 95% CI 0.91-1.33) but was associated with statin prescription (OR 1.40 per 1 SD increase, 95% CI 1.13-1.74) in univariate models. Total cholesterol and antihypertensive medication use were independently associated with statin discussion after adjusting for traditional CVD risk factors, OR 1.48 per 1 SD increase (95% CI 1.18-1.86) and OR 3.80 (95% CI 2.43-5.95), respectively. Total cholesterol and anti-hypertensive medication use were also independently associated with statin prescription, OR 1.96 per 1 SD increase (95% CI 1.51-2.55) and OR 4.03 (95% CI 2.39-6.79), respectively. Other major drivers of risk (i.e. age and smoking) were not associated with statin prescribing in univariate and multivariate models (Table).
Conclusions: Total cholesterol and antihypertensive medication use were principle drivers of statin use whereas other determinants of CVD risk were not. Further studies are needed to support risk-based frameworks to guide statin utilization.
Author Disclosures: K.N. Karmali: B. Research Grant; Significant; NIH-T32. S.D. Persell: A. Employment; Significant; Northwestern University, Northwestern Medical Group. B. Research Grant; Significant; NIH, AHRQ, HRSA. C. Other Research Support; Modest; Pfizer, Inc. G. Consultant/Advisory Board; Modest; American Board of Internal Medicine.
- © 2015 by American Heart Association, Inc.