Development and validation of the Reynolds Risk Score for the assessment of global cardiovascular risk in women

Global cardiovascular prediction algorithms include Framingham’s and PROCAM’s algorithms, as well as the SCORE charts. In women, using such predictive tools has relied so far on traditional risk factors. In order to develop cardiovascular risk algorithms for women based on a large panel of both traditional and novel risk factors, 35 factors were assessed among 24558 healthy US women aged ≥45 years who were followed up for a median of 10.2 years for incident cardiovascular events. Data among a random two thirds (derivation cohort) were used to develop new risk algorithms that were then tested in the remaining one third of women (validation cohort) by comparing observed against predicted outcomes. In the derivation cohort, minimization of the Bayes Information Criterion was used to develop the best-fitting parsimonious prediction models. In the validation cohort, whilst comparing predicted vs actual 10-year cardiovascular event rates, the new algorithms were compared with models based on covariates included in the Adult Treatment Panel III risk score. As a result, in the derivation cohort, a best-fitting model (model A) and a clinically simplified model (model B, the Reynolds Risk Score) had lower Bayes Information Criterion scores than models based on covariates used in the Adult Treatment Panel III. All measures of fit, discrimination and calibration were improved in the validation cohort when using either model A or model B. In this context, model A reclassified 40 to 50% of women at intermediate risk according to Adult Treatment Panel III into higher- or lower-risk categories. Similar effects were achieved for a clinically simplified model B, encompassing (1) age, (2) systolic blood pressure, (3) haemoglobin A(1c) if diabetic, (4) smoking, (5) total and (6) HDL-C, (7) hsCRP, and (8) parental history of myocardial infarction <60 years. By contrast, neither new algorithm provided substantive information about women at very low risk based on Adult Treatment Panel III score.

Abstract