Research Article

Predicting Increased Blood Pressure Using Machine Learning

Table 3

Logistic regression results of women’s training set.

CoefficientsS.EWald Pseudo AUC

Model 1
 Intercept17.61512.2061.440.14900.0190.548
 BMI−0.06510.0519−1.250.2095
Model 2
 Intercept20.58417.1491.200.23000.0140.529
 WC−0.02360.0223−1.060.2886
Model 3
 Intercept18.70122.6870.820.40980.0060.542
 HC−0.01600.0224−0.720.4737
Model 4
 Intercept25.47029.0260.880.38020.0080.539
 WHR−0.03030.0382−0.790.4278
Model 5
 Intercept17.49717.7390.990.32400.0190.549
 BMI−0.06590.0973−0.680.4987
 WC0.00040.04180.010.9927
Model 6
 Intercept0.595825.2510.240.81350.0220.572
 BMI−0.10300.0891−1.160.2477
 HC0.02030.03860.530.5993
Model 7
 Intercept26.31929.1050.900.36580.0200.541
 BMI−0.05790.0562−1.030.3025
 WHR−0.01370.0414−0.330.7414
Model 8
 Intercept16.87922.8290.740.45970.0140.534
 WC−0.03110.0378−0.820.4103
 HC0.00930.03810.250.8063
Model 9
 Intercept23.66029.1570.810.41710.0140.529
 WC−0.02110.0293−0.720.4711
 WHR−0.00660.0503−0.130.8961
Model 10
 Intercept38.63935.4641.090.27590.0130.529
 HC−0.01460.0225−0.650.5151
 WHR−0.02820.0383−0.730.4626
Model 11
 Intercept0.693825.6810.270.78700.0230.566
 BMI−0.09040.1069−0.850.3976
 WC−0.00970.0454−0.210.8317
 HC0.02370.04190.570.5711
Model 12
 Intercept−235.90929.7366−0.790.4270.0230.539
 WC−0.35480.3821−0.930.353
 HC0.25710.2935 0.880.381
 WHR0.33030.3875 0.850.394
Model 13
 Intercept28.20229.7860.950.34370.0210.558
 BMI−0.08560.1070−0.800.4239
 WC0.01700.05590.300.7609
 WHR−0.02490.0554−0.450.6538
Model 14
 Intercept10.60346.9010.230.82110.0220.566
 BMI−0.09620.1061−0.910.3646
 HC0.01810.04260.430.6699
 WHR−0.00540.0458−0.120.9064
Model 15
 Intercept−22.377729.6673−0.750.45070.0300.57
 BMI−0.08330.1078−0.770.4399
 WC−0.30780.3851−0.800.4242
 HC0.24950.29260.850.3939
 WHR0.30250.38760.780.4351

Notice that the endpoint for women is systolic blood pressure greater than 120 mmHg (prehypertension), while for men is greater than 140 mmHg (hypertension).