Using Artificial Intelligence to Develop a Multivariate Model with a Machine Learning Model to Predict Complications in Mexican Diabetic Patients without Arterial Hypertension (National Nested Case-Control Study): Metformin and Elevated Normal Blood Pressure Are Risk Factors, and Obesity Is Protective
Table 5
Mann–Whitney test for numerical variables between complicated patients versus uncomplicated patients.
Predictor
Patients
UMW test
Complicated
Uncomplicated
Abortions
2.0 (IQR 1.0-2.0)
2.0 (IQR 2.0-2.0)
<0.001
Cesarean section
1.0 (IQR 1.0-2.0)
1.0 (IQR 1.0-1.0)
<0.001
Pregnancies
4.0 (IQR 4.0-5.0)
4.0 (IQR 4.0-4.0)
<0.001
Deliveries
4.0 (IQR 3.0-4.0)
4.0 (IQR 4.0-4.0)
<0.001
BMI
30.04 (IQR 27.03-33.33)
29.85 (IQR 26.84-33.49)
0.004
Maximum weight
93.0 (IQR 82.0-110.0)
116.5 (IQR 103.0-116.5)
<0.001
Age
59.0 (IQR 51.0-66.0)
59.0 (IQR 52.0-66.0)
0.277
Diastolic pressure
80.0 (IQR 80.0-90.0)
80.0 (IQR 70.0-80.0)
<0.001
Systolic pressure
130.0 (IQR 120.0-140.0)
126.0 (IQR 120.0-140.0)
<0.001
Pulse pressure
50.0 (IQR 40.0-60.0)
50.0 (IQR 40.0-60.0)
<0.001
Fasting glucose
279.0 (IQR 200.0-350.0)
200.0 (IQR 140.0-280.0)
<0.001
HbA1c
8.8 (IQR 7.1-10.7)
8.0 (IQR 6.8-10.0)
<0.001
Years smoking
25.0 (IQR 12.0-38.0)
38.0 (IQR 30.0-38.0)
<0.001
Complicated: total number of patients and percentage in whom physicians registered one or more of the five events of interest in their medical record. Uncomplicated: total number of patients and percentage in whom physicians registered none of the five events of interest in their medical record.