Skip to main content

Table 2 Associations between differences in risk factors with differences in age-adjusted mortality rates between adjacent counties

From: Explaining large mortality differences between adjacent counties: a cross-sectional study

 

Univariate models

Multivariable model

Variable

Estimate

95 % CI

DIC

Estimate

95 % CI

Predisposing factors

       

 ΔAfrican Americans (%)

0.54

−1.33

2.27

687.66

   

Access

       

 ΔPreventable hospitalization ratea

6.67

4.86

8.70

722.07

   

 ΔPopulation without health insurance (%)

3.97

2.78

5.10

692.23

   

 ΔPopulation unable to obtain medical care because of cost (%)

11.77

9.70

13.97

688.89

2.60

0.59

4.57

Use of health services

       

 ΔEmergency department visits ratea

16.96

10.45

24.14

695.67

   

 ΔHospital discharge ratea

1.87

1.43

2.34

725.89

1.03

0.66

1.38

Health behavior

       

 ΔPopulation currently smoking (%)

4.78

3.15

6.26

710.03

   

 ΔPopulation without leisure-time physical activity (%)

6.69

4.88

8.51

724.84

   

Need

       

 ΔPopulation in fair or poor health (%)

10.66

9.09

12.22

706.82

2.93

1.26

4.59

 ΔPopulation overweight or obese (%)

3.60

1.97

5.28

686.08

   

 ΔPopulation with high blood pressure (%)

11.19

9.31

13.10

686.56

4.75

2.94

6.47

 ΔPopulation with diabetes (%)

5.61

1.97

8.93

685.38

   

Enabling

       

 ΔPopulation below federal poverty level (%)

11.26

9.50

13.13

704.97

6.08

4.38

7.97

Deviance information criteria (DIC)

    

768.84

  
  1. Abbreviations: CI credible interval, DIC deviance information criteria
  2. aper 100,000 population bΔ signifies the difference between adjacent counties
  3. Bold font indicates estimates with 95 % CI that do not include the value of zero