Fig. 3: Test area under the curve (AUC) for the random forest (RF) classifier (same global classifier) stratified by number of previous distinct prescriptions. | Translational Psychiatry

Fig. 3: Test area under the curve (AUC) for the random forest (RF) classifier (same global classifier) stratified by number of previous distinct prescriptions.

From: Predicting treatment dropout after antidepressant initiation

Fig. 3

Stratification according to how many distinct treatments have already been tried in the past. AUC improves with the number of tried treatments in the past. Four different types of input data are considered: sociodemographic features (dem), date of prescription (date), and diagnostic/procedure codes (codes). Confidence intervals computed using 500 bootstraps.

Back to article page