Skip to main content

Table 4 AUCs for clinician predictions based on clinician-identified patient risk factors for readmission

From: Clinicians can independently predict 30-day hospital readmissions as well as the LACE index

Patient identified as having:

AUC for Clinician Predictions (95% CI)

Residents

Attendings

Nurses

Poor understanding

0.516 (0.355, 0.675) n = 79

0.713 (0.590, 0.836) n = 88

0.778 (0.614, 0.941) n = 44

Poor adherence

0.700 (0.532, 0.868) n = 52

0.610 (0.456, 0.765) n = 68

0.498 (0.280, 0.717) n = 40

Poor social support

0.669 (0.457, 0.881) n = 53

0.631 (0.432, 0.831) n = 52

0.593 (0.337, 0.849) n = 24

Severe disease

0.629 (0.542, 0.717) n = 206

0.649 (0.569, 0.728) n = 220

0.572 (0.434, 0.710) n = 93

High-risk therapy

0.577 (0.416, 0.738) n = 74

0.594 (0.461, 0.727) n = 87

0.616 (0.405, 0.826) n = 45

Medically complex

0.594 (0.464, 0.724) n = 88

0.601 (0.459, 0.742) n = 75

0.543 (0.363, 0.723) n = 52

Organ transplant

0.493 (0.309, 0.677) n = 48

0.661 (0.504, 0.818) n = 49

0.574 (0.314, 0.834) n = 30

Previous admission

0.660 (0.490, 0.829) n = 43

0.510 (0.266, 0.754) n = 26

0.547 (0.263, 0.831) n = 22