Skip to main content

Table 4 Multivariate logistic regression with switching as the dependent variable

From: Does the chronically ill population in the Netherlands switch their health insurer as often as the general population? Empirical evidence from a nationwide survey study

 

Model 1: Switching

Pseudo R2 = 0.03

N = 1961

Model 2: Switching with interaction terms

Pseudo R2 = 0.06

N = 1961

Switching (1 = yes. 0 = no)

 

Odds Ratio

P-value

Marginal Effect

Odds Ratios

P-value

Marginal Effect

Age

18–39

Reference

 

0.08

Reference

 

0.06

40–64

0.64

0.10

0.05

0,26

0.00*

0.05

65 years and older

0.44

0.01*

0.04

0,25

0.00*

0.04

Education

Lower

Reference

 

0.05

Reference

 

0.04

Intermediate

0.97

0.91

0.05

2,33

0.27

0.05

Higher

1.45

0.24

0.06

3,38

0.12

0.06

Perceived health condition

Very good

Reference

 

0.06

Reference

 

0.05

Good

0.78

0.36

0.04

0.66

0.22

0.04

Bad

1.02

0.95

0.06

0,85

0.76

0.05

Group

General population

Reference

 

0.06

Reference

 

0.06

Chronically ill population

0.67

0.08

0.04

0.16

0.10

0.04

Sex

Male

Reference

 

0.05

Reference

 

0.05

Female

0.82

0.37

0.05

0.93

0.83

0.05

Age * group

18–39 *

general population

–

–

–

Reference

 

0.13

40–64 *

chronically ill population

–

–

–

13.23

0.00*

0.06

65 years and older *

chronically ill population

–

–

–

7,07

0.01*

0.03

Education*group

Lower*

general population

–

–

–

Reference

–

0,03

Intermediate*

chronically ill population

–

–

–

0,38

0,25

0,04

Higher*

chronically ill population

–

–

–

0,36

0,24

0,05

Perceived health condition

* group

Very good*general population

–

–

–

Reference

 

0,07

Good*

chronically ill population

–

–

–

2,03

0,26

0,04

Bad*

chronically ill population

–

–

–

1,90

0,40

0,05

Sex*group

Male*general population

–

–

–

Reference

 

0,06

Female*Chronically ill population

–

–

–

0,91

0,84

0,04

Constant

 

0.12

0.00*

 

0,09

0.00*

 
  1. *p < 0,05