Table 3.

Prediction performance of LASSO, FRS, and ASCVD models, n = 612.

	AIC	BIC	AUC (95% CI)	Cut-point	Sensitivity	95% CI	Specificity	95% CI	+LR	95% CI	–LR	95% CI	+PV	95% CI	–PV	95% CI
LASSO model	386.5443	430.6682	0.780 (0.731–0.829)	>0.2205	78.49	68.8–86.3	67.62	62.1–72.8	2.42	2.1–2.8	0.32	0.2–0.5	41.7	34.3–49.4	91.4	87.1–94.7
FRS model	624.3103	633.0672	0.654 (0.605–0.702)	>11.1	71.22	62.9–78.6	57.78	53.1–62.4	1.69	1.5–1.9	0.5	0.4–0.7	34.3	28.8–40.0	86.7	82.3–90.3
ASCVD model	633.0280	641.7848	0.661 (0.612–0.709)	>5.5	69.78	61.4–77.3	57.78	53.1–62.4	1.65	1.4–1.9	0.52	0.4–0.7	33.8	28.3–39.6	86.1	81.7–89.8

LASSO, least absolute shrinkage and selection operator; FRS, Framingham risk score; ASCVD, atherosclerotic cardiovascular disease; AIC, Akaike information criterion; BIC, Bayesian information criterion; AUC, area under the ROC curve; CI, confidence interval; LR, likelihood ratio; PV, predictive value.