Skip to main content

Table 2 Initial exploratory model (multivariable logistic regression) with outcome as odds ratio of admissions

From: Using health administrative data to model associations and predict hospital admissions and length of stay for people with eating disorders

 

Initial Exploratory Model

Variable

Odds Ratio

SE

Z

p

95% CI (Lower)

95% CI (Upper)

Year (Ref: 2018)

-

-

-

0.089*

-

-

2011

0.49

0.44

-0.80

0.425

0.08

2.87

2012

0.26

0.16

-2.23

0.026

0.08

0.85

2013

0.44

0.35

-1.04

0.299

0.09

2.09

2014

0.89

0.68

-0.15

0.882

0.20

4.01

2015

0.79

0.56

-0.33

0.738

0.20

3.15

2016

1.16

0.96

0.18

0.857

0.23

5.85

2017

0.81

0.58

-0.30

0.767

0.20

3.27

2019

0.34

0.19

-1.93

0.054

0.11

1.02

2020

0.18

0.12

-2.67

0.008

0.05

0.64

Eating Disorder Diagnoses (Ref: Anorexia Nervosa)

-

-

-

0.065*

-

-

Bulimia Nervosa

0.31

0.18

-2.05

0.04

0.10

0.95

Other

0.58

0.20

-1.62

0.105

0.30

1.12

Age

1.04

0.02

2.43

0.015

1.01

1.08

Facility (Ref: Not RPA)

      

RPA

0.73

0.35

-0.65

0.513

0.28

1.89

Triage Category

0.37

0.11

-3.39

0.001

0.21

0.66

Referral Source (Ref: Self/Family/Friends)

      

Other

2.03

0.74

1.96

0.050

1.00

4.13

Mode of Arrival (Ref: Not Ambulance)

      

Ambulance

2.28

0.88

2.15

0.032

1.07

4.86

Gender (Ref: Male)

      

Female

0.60

0.42

-0.73

0.465

0.15

2.36

Marital Status (Ref: Not Married/De facto)

      

Married/De facto

1.01

0.51

0.02

0.984

0.38

2.71

Index of Relative Socioeconomic Disadvantage

1.17

0.14

1.34

0.179

0.93

1.48

  1. *Wald Test for significance of multinomial variables