Skip to main content
. 2020 May 23;9(5):1587. doi: 10.3390/jcm9051587

Figure 1.

Figure 1

(a) Area under the ROC curve for the prediction model, and (b) predicted versus observed low birthweight probability in the sample. This analysis includes neonates born at term (n = 379). The calibration plot created using “givitiCalibrationBelt” in R programming. Linear predictors for estimated risk of low birthweight = 1/(1 + exp − (−2.54 + 1.593 × age(<20) + 1.516 × BMI (<18.5) + 1.213 × hemoglobin (<11) + 1.225 * height (<155) + 0.606 × prim-gravid + 1.475 × comorbidity. ROC = receiver operating characteristic.