Research Article

Predictive Models for Knee Pain in Middle-Aged and Elderly Individuals Based on Machine Learning Methods

Table 2

Multivariable logistic regression.

Odds ratio95% CI

Age1.011.00-1.020.105
Gender (female)1.281.06-1.550.011
Race
 Non-Hispanic whiteRef
 Non-Hispanic black1.150.91-1.450.249
 Mexican American1.020.81-1.280.889
 Others0.910.65-1.260.563
Education
 Below high schoolRef
 High school1.060.86-1.310.588
 Above high school0.950.78-1.150.598
Hypertension (yes)1.110.95-1.310.198
Diabetes (yes)0.910.73-1.120.376
Pain elsewhere (yes)4.643.98-5.43< 0.001
Moderate activity (yes)1.040.89-1.220.617
Vigorous activity (yes)0.900.72-1.120.338
Smoking (yes)1.120.96-1.310.158
Drinking (yes)0.860.70-1.050.152
BMI (kg/m2)1.051.02-1.08< 0.001
Waist circumference1.000.99-1.020.471
Albumin (g/L)1.010.98-1.040.584
Phosphorus (mg/dL)0.990.85-1.150.850
Total calcium (mg/dL)1.010.82-1.240.927
Triglycerides (mg/dL)1.001.00-1.000.470
Cholesterol (mg/dL)1.001.00-1.000.674
Vitamin D (nmol/L)1.000.99-1.000.438
eGFR (ml/min/)1.001.00-1.010.457

BMI: body mass index; eGFR: estimated glomerular filtration rate.