查看文章

Established risk factors (blood pressure, lipids, smoking, and diabetes) are strong predictors of coronary heart disease (CHD). However, many risk factors, such as blood pressure and cholesterol, exhibit long-term variation-a combination of instrument measurement error, physiologic short-term variability, and long term changes. We examined the extent to which adjusting for long term variation, using a measurement error framework, reduces the underestimation of risk due to information dilution in a single measurement approach. Using 12,378 participants from the Atherosclerosis Risk in Communities (ARIC) Study, we compared several standard Poisson models with the corresponding regression calibration adjusted models (real in STATA 8) to predict the CHD event rate over an average of 10 years of followup. The adjusted model incorporated measurements for systolic blood pressure (SBP), total cholesterol and …