Skip to main content

Table 2 Odds ratios and rate ratios of the logistic and poisson components of the tZIP Model

From: A hybrid machine learning framework to improve prediction of all-cause rehospitalization among elderly patients in Hong Kong

Probability of Inactive Rehospitalisation Status (p(t))

Odds Ratio

Coefficient (SE)

z Value

P Value

Sig.

Intercept

2.385

0.869 (0.043)

20.058

< 2E-16

***

Length of Stay

0.988

−0.012 (0.003)

−4.272

1.93E-05

***

Acute Admission (Yes)

0.627

−0.467 (0.041)

− 11.306

< 2E-16

***

Charlson Comorbidity Index

0.971

−0.029 (0.011)

−2.628

8.59E-03

**

Visits to Emergency Room during Previous 6 Months

0.627

−0.467 (0.018)

−25.955

< 2E-16

***

Probability of Active Rehospitalisation Rate (λ(t))

Rate Ratio

Coefficient (SE)

z Value

P Value

Sig.

Intercept

1.061

0.059 (0.016)

3.679

2.34E-04

***

Length of Stay

0.995

−0.005 (0.001)

−5.892

3.82E-09

***

Acute Admission (Yes)

0.812

−0.208 (0.015)

− 13.706

< 2E-16

***

Charlson Comorbidity Index

1.054

0.053 (0.003)

15.216

< 2E-16

***

Visits to Emergency Room during Previous 6 Months

1.161

0.149 (0.003)

48.291

< 2E-16

***