Table 3.

Results of the readmissions prediction model.

Model		AUROCa	Sensitivity	Specificity	PPVb	NPVc
LACEd baseline
	Frequente	0.8288	0.7021	0.7840	0.7466	0.7438
	Standardf	0.8302	0.7341	0.7606	0.6649	0.8156
	Geriatricsg	0.8254	0.7479	0.7328	0.6713	0.7994
	Combinedh	0.8397	0.7303	0.7757	0.6696	0.8221
LACE baseline+sociali
	Frequent	0.8573	0.7598	0.7573	0.7394	0.7767
	Standard	0.8621	0.7661	0.7796	0.6922	0.8375
	Geriatrics	0.8686	0.7749	0.7832	0.7228	0.8267
	Combined	0.8707	0.7763	0.7825	0.6896	0.8490
LACE baseline+physicianj+social
	Frequent	0.8952	0.8232	0.8136	0.8001	0.8354
	Standard	0.9001	0.8224	0.8235	0.7509	0.8776
	Geriatrics	0.9100	0.8318	0.8331	0.7843	0.8716
	Combined	0.9069	0.8254	0.8318	0.7534	0.8845

^aAUROC: area under the receiving operating characteristic curve.

^bPPV: positive predictive value.

^cNPV: negative predictive value.

^dLACE: Length of stay, Acuity of the admission, Comorbidity of the patient, and Emergency department use.

^eHold-out sample of patients identified by the hospital as frequent readmission patients.

^fHold-out sample of a typical hospital patient.

^gSubset of patients in the “Standard” sample who are 65 years or older.

^hCombination of “Frequent” and “Standard” hold-out samples.

ⁱSocial represents the text-mined notes that medical social workers and case managers provided.

^jPhysician represents the text-mined notes provided by physicians.