Skip to main content

Table 2 AUC for predicting asthma exacerbation over different time horizons and variable sets using different modeling methods

From: Environmental and clinical data utility in pediatric asthma exacerbation risk prediction models

Outcome

Predicted Event Rate

Overall model

Temporal factors

Clinical factors

Spatial factors

Parsimonious model

AUC

95% CI

AUC

95% CI

AUC

95% CI

AUC

95% CI

AUC

95% CI

AUC of Person-Month LASSO Models

 Exacerbation in 30 days

0.015

0.753

(0.732, 0.773)

0.600

(0.594, 0.616)

0.734

(0.713, 0.756)

0.502

(0.475, 0.535)

0.667

(0.638, 0.694)

 Exacerbation in 90 days

0.043

0.740

(0.718, 0.761)

0.576

(0.562, 0.591)

0.732

(0.709, 0.753)

0.492

(0.462, 0.521)

0.643

(0.616, 0.670)

 Exacerbation in 180 days

0.077

0.732

(0.710, 0.753)

0.547

(0.533, 0.563)

0.729

(0.706, 0.751)

0.492

(0.463, 0.520)

0.645

(0.618, 0.672)

AUC of Person-Month Random Forest Survival Models

 Exacerbation in 30 days

0.015

0.757

(0.736, 0.777)

0.608

(0.602, 0.624)

0.741

(0.720, 0.763)

0.502

(0.470, 0.535)

0.672

(0.647, 0.697)

 Exacerbation in 90 days

0.043

0.747

(0.726, 0.750)

0.582

(0.567, 0.599)

0.738

(0.717, 0.758)

0.498

(0.471, 0.524)

0.644

(0.619, 0.671)

 Exacerbation in 180 days

0.077

0.729

(0.707, 0.750)

0.557

0.539, 0.577)

0.725

(0.703, 0.746)

0.541

(0.512, 0.570)

0.648

(0.623, 0.673)

AUC of Person-Month Gradient Boosting (xgBoost) Models

 Exacerbation in 30 days

0.015

0.761

(0.742, 0.781)

0.607

(0.601, 0.623)

0.742

(0.721, 0.763)

0.501

(0.469, 0.534)

0.664

(0.637, 0.688)

 Exacerbation in 90 days

0.043

0.752

(0.730, 0.771)

0.581

(0.566, 0.597)

0.744

(0.722, 0.763)

0.488

(0.459, 0.517)

0.639

(0.613, 0.665)

 Exacerbation in 180 days

0.077

0.739

(0.717, 0.760)

0.557

(0.538, 0.576)

0.730

(0.708, 0.752)

0.503

(0.474, 0.531)

0.640

(0.614, 0.665)