Abstract 5005: Development and Testing of a Risk Prediction Algorithm for Incident Hypertension in Women with Currently Normal Blood Pressure
Background: We sought to determine whether a hypertension risk prediction model based on patient characteristics and blood biomarkers might improve upon risk prediction based on current blood pressure alone.
Methods: A prospective cohort of 14,822 normotensive women were followed over 8 years for the development of hypertension. Among a randomly selected two-thirds sample (“development cohort”; N=9,427), hypertension prediction models were developed using 52 potential predictors and compared to a model based on blood pressure alone. Each prediction model was validated in the remaining one-third sample (“validation cohort”; N=5,395).
Results: In the development cohort, the best prediction model for incident hypertension included age, blood pressure, ethnicity, body mass index, total grain intake, apolipoprotein B, lipoprotein(a), and C-reactive protein. While this model was superior to a model based on blood pressure alone. it was only marginally better than a simplified model including age, blood pressure, ethnicity, and body mass index. In the validation cohort, the best model was no longer calibrated. The simplified model, however, demonstrated adequate calibration in the validation cohort (Hosmer-Lemeshow P-value 0.14), with a c-index similar to that of the best model (0.703 vs 0.705), and when compared to the model based on blood pressure alone (Table⇓), reclassified 1499 participants to hypertension risk categories that were closer to the observed risk in all but one instance.
Conclusion: In this prospective cohort of initially normotensive women, a model based on readily available clinical information predicted incident hypertension better than a model based on blood pressure alone and, unlike more complex models using blood biomarkers, remained calibrated in the validation cohort.