1. Home>
2. Specialties>
3. Cardiology>
4. Summary and Comment

A New Model for Predicting Cardiovascular Risk in Women?

A model derived from Women’s Health Study data emphasizes traditional risk factors plus two factors not incorporated in traditional risk models.

Research on estimation of cardiovascular risk in women is fairly thin, suggesting that existing risk-prediction models (e.g., Framingham and its ATP-III version) may not accurately represent women. Using data from 24,558 initially healthy participants in the Women’s Health Study (95% white; baseline age 45; median follow-up, 10.2 years), researchers developed a risk algorithm for predicting incident cardiovascular events: MI, ischemic stroke, coronary revascularization, and cardiovascular death.

The researchers considered 35 variables for the prediction model, which was derived from a randomly selected 16,400 participants. Then, they compared predicted 10-year event rates with actual rates in the remaining 8158 participants (validation cohort). A robust, best-fit model and a clinically simplified model (the Reynolds Risk Score [RRS]) were developed.

The simplified RRS incorporated age, systolic blood pressure, hemoglobin A_1c level (for diabetics), current smoking, and total- and HDL-cholesterol levels, as well as high-sensitivity C-reactive protein (hs-CRP) levels and parental history of early MI (before age 60). Of women with ATP-III intermediate-risk classifications, 30% of nondiabetics and 45% of diabetics were reclassified as higher- or lower-risk by the RRS. Compared with the ATP-III model, the RRS more accurately predicted actual event rates, and the best-fit model performed even better.

Comment: The simplified Reynolds Risk Score yielded incrementally better prediction of 10-year cardiovascular risk in women, compared with the ATP-III version of the Framingham Risk Score. The RRS reclassified a large proportion of ATP-III intermediate-risk women into other risk categories. The model might not accurately reflect risk in nonwhite women or in men. Still, the RRS emphasizes the value of family history and the potential value of hs-CRP measurements in more carefully stratifying "intermediate-risk" women. For now, family history is a sensible complement to Framingham risk assessment; CRP testing should be reserved for selected patients, as defined by CDC–AHA recommendations (Journal Watch Cardiology May 9 2003).

— JoAnne M. Foody, MD

Published in Journal Watch Cardiology February 13, 2007

Citation(s):

Ridker PM et al. Development and validation of improved algorithms for the assessment of global cardiovascular risk in women: The Reynolds Risk Score. JAMA 2007 Feb 14; 297:611-9.

• Original article (Subscription may be required)
• Medline abstract (Free)

Blumenthal RS et al. Further improvements in CHD risk prediction for women. JAMA 2007 Feb 14; 297:641-3.

• Original article (Subscription may be required)
• Medline abstract (Free)