Table 2.

Performance of DAP model and baseline models within 365 days after discharge on 10-fold internal validation for cohorts from NHIP and PHS.

Models	ROC-AUC	P-value	PR-AUC	P-value
NHIPa
LR	0.60 ($\pm$0.011)	.000	0.32 ($\pm$0.009)	.000
RF	0.73 ($\pm$0.011)	.000	0.51 ($\pm$0.015)	.000
XGB	0.72 ($\pm$0.012)	.000	0.46 ($\pm$0.026)	.000
DAP	0.80 ($\pm$0.010)	ref	0.61 ($\pm$0.018)	ref
PHS
LR	0.60 ($\pm$0.072)	.000	0.41 ($\pm$0.101)	.000
RF	0.60 ($\pm$0.057)	.000	0.34 ($\pm$0.056)	.000
XGB	0.65 ($\pm$0.072)	.000	0.41 ($\pm$0.072)	.000
DAP	0.91 ($\pm$0.028)	ref	0.80 ($\pm$0.067)	ref

^aThe records from PHS were removed.

The numbers in parentheses are the standard deviation. The bolded part indicates the best performance of the corresponding data under the respective metric.

Evaluation metrics included ROC-AUC and PR-AUC. We conducted a t-test to compare the differences in results between the two groups generated from the 10-fold data.

Abbreviations: DAP, depression and anxiety prediction; LR, logistic regression; NHIP, Nanjing Health Information Platform; PHS, primary healthcare services; RF, random forest; XGB, extreme gradient boosting.