Clinicians deciding on which risk model to use should be aware that the Framingham CVD model will likely increase the number of people who qualify as moderate and high risk.
In guidelines on primary prevention of cardiovascular disease (CVD), the Framingham coronary heart disease (CHD) risk score has long been recommended for assessing 10-year risk of myocardial infarction (MI) or cardiac death. In 2008, the Framingham group developed the Framingham CVD risk score, which added new endpoints of angina, cerebrovascular disease, peripheral vascular disease, and heart failure to the CHD risk score. The Framingham CVD risk score has since been adopted and favored by the latest guidelines for cardiovascular risk assessment.
Significant Implications of CVD Risk
Little is known about the population level effects of switching from the Framingham CHD risk score to the Framingham CVD score. In the May/June 2011 Journal of Clinical Lipidology, my colleagues and I conducted a study to determine the impact of switching to the Framingham CVD score from the Framingham CHD score on the United States population. For our study, 1,020 patients from NHANES were analyzed to define the changes in risk category distribution and corresponding changes to lipid goal achievements. When the Framingham CVD risk score was used, 63% of men and 74% of women moved up by at least one risk category when compared with the Framingham CHD risk score. Additionally, the low-risk population decreased from 52% to 16%, and the high-risk population increased from 4% to 20%.
“Little is known about the effect of switching from the Framingham CHD risk score to the Framingham CVD score.”
In February 2011, the American Heart Association released recommendations on CVD prevention in women. It modified 10-year risk thresholds and recommended using the Framingham CVD risk model for routine risk assessment. In the updated guidelines, the “high-risk” threshold for a woman’s 10-year cardiovascular risk was lowered from 20% to 10%. When we used the updated thresholds and applied the Framingham CVD risk score— as opposed to the Framingham CHD risk score—we found that 44% of women aged 55 to 74 with no prior cardiac disease would now be classified as high risk for CVD.
Treatment Implications for CVD
When subjects in NHANES were reclassified based on the Framingham CVD risk score, about 30% of those previously meeting lipid goals by the Framingham CHD score no longer met lipid goals. This means that pharmacotherapy would likely need to be initiated or intensified for many more patients. Published research has consistently documented that statin therapies can reduce CHD mortality, the endpoint of the Framingham CHD, by 25% to 30%. The effects of statins on lowering the risks associated with the other endpoints—stroke, peripheral vascular disease, and heart failure—included in the Framingham CVD risk score are not as clear. More data are needed to assess the potential benefits of therapies for patients based on the additional components that are accounted for in the Framingham CVD risk score.
Increase Awareness of CVD Model
It’s important to note that the Framingham CVD risk score is a valuable tool in the fight against heart disease. Current primary prevention guidelines recommend routine assessment of cardiovascular risk using a multivariate risk model. Clinicians deciding on which risk model to use should be aware that the Framingham CVD model will likely increase the number of people who qualify as moderate and high risk. The treatment effects of statin therapy, while known to reduce cardiac death and MI, have less certain and differing treatment effects on other endpoints.
Tattersall M, Karmali K, Gangnon R, Keevil J. The population effects of the global cardiovascular risk model in United States adults: findings from the National Health and Nutrition Surveys 2005-2006. J Clin Lipidol. 2011;5:166-172.
Mosca L, Benjamin E, Berra K, et al. Effectiveness-based guidelines for the prevention of cardiovascular disease in women—2011 update: a guideline from the American Heart Association. Circulation. 2011;123:1243-1262.
Unverzagt F, McClure L, Wadley V, et al. Vascular risk factors and cognitive impairment in a stroke-free cohort. Neurology. 2011;77:1729-1736.
Defilippis A, Blaha M, Ndumele C, et al. The Association of Framingham and Reynolds Risk Scores With Incidence and Progression of Coronary Artery Calcification in MESA (Multi-Ethnic Study of Atherosclerosis). J Am Coll Cardiol. 2011;58:2076-2083.
Yoshida M, Mita T, Yamamoto R, et al. The Combination of Framingham Risk Score and Carotid Intima-Media Thickness Improves the Prediction of Cardiovascular Events in Patients With Type 2 Diabetes. Diabetes Care. 2011, Oct 25 [Epub ahead of print]. Available at http://care.diabetesjournals.org/content/early/2011/10/20/dc11-1333.abstract.
Kimokoti R, Newby PK, Gona P, et al. Stability of the Framingham Nutritional Risk Score and its component nutrients over 8 years: the Framingham Nutrition Studies. Eur J Clin Nutr. 2011, Oct 5. [Epub ahead of print]. Available at www.nature.com/ejcn/journal/vaop/ncurrent/abs/ejcn2011167a.html.
Greenland P, Alpert JS, Beller GA, et al. (2010) 2010 ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: a report of the American College of Cardiology Foundation/American Heart Association task force on practice guidelines. Circulation 122: e584-636.