Skip to main content
. 2024 Apr 25;5(4):427–434. doi: 10.1093/ehjdh/ztae034

Table 2.

A logistic regression model for detecting left ventricular systolic dysfunction

Units Coefficient (absolute) Coefficient (normalized) P-value Independent AUC
aVR T amplitude µV 3.69E−3 (3.35E−3 to 4.04E−3) 0.52 (0.47–0.57) 7.22E−96 0.77 (0.75–0.78)
V3 QRS duration ms 2.03E−2 (1.82E−2 to 2.23E−2) 0.51 (0.45–0.56) 2.56E−81 0.75 (0.73–0.76)
Heart rate b.p.m. 1.97E−2 (1.73E−2 to 2.21E−2) 0.38 (0.34–0.43) 1.17E−58 0.63 (0.61–0.65)
V3 QRS minimum deflection µV −4.76E−4 (−5.34E−4 to −4.18E−4) −0.34 (−0.38 to −0.30) 7.07E−58 0.70 (0.68–0.72)
QTc ms 6.73E−3 (5.38E−3 to 8.08E−3) 0.27 (0.22–0.33) 1.50E−22 0.75 (0.73–0.77)
Intercept Unitless −9.10 (−9.66 to −8.55) −2.61 (−2.68 to −2.55) 3.11E−225

Coefficients for absolute covariates and covariates normalized to mean 0 and standard deviation 1 are shown, along with units for each absolute covariate, P-values for each coefficient and AUCs for each covariate as an independent predictor.