Discrimination and reclassification for 5-y all-cause mortality in the FINRISK cohort with and without circulating biomarkers in the risk prediction score.

<p>Risk discrimination was assessed for FINRISK study participants with the risk prediction scores derived in the Estonian Biobank cohort (<a href="http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.1001606#pmed-1001606-t002" target="_blank">Table 2</a>). The reference risk score included age, sex, body mass index, systolic blood pressure, fasting time, total cholesterol, HDL cholesterol, triglycerides, creatinine, smoking status, alcohol consumption, prevalent diabetes, prevalent cardiovascular disease, and prevalent cancer. The biomarker risk prediction score was extended with alpha-1-acid glycoprotein, albumin, VLDL particle size, and citrate. Complete data were available for 7,110 individuals, of which 157 died during the 5-y follow-up period. Category-based reclassification was assessed for four risk categories (≤1.25%, 1.25%–2.5%, 2.5%–5%, ≥5%) based on the reference risk score and the biomarker risk score. Reclassification tables for these groups and model calibration of the prediction scores within risk deciles are shown in <a href="http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.1001606#pmed.1001606.s003" target="_blank">Figure S3</a> and <a href="http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.1001606#pmed.1001606.s007" target="_blank">Table S2</a>.