Skip to main content

Table 3 Stepwise Logistic Regression Analysis with Odds for Predictors of Long Gun Use Among Firearm Suicides

From: Prevalence of long gun use in Maryland firearm suicides

Variable

Model 1

Model 2

Model 3

OR

95% CI

p

OR

95% CI

p

OR

95% CI

p

NCHS Rurality

 Large metro (1)

ref

–

–

ref

–

–

ref

–

–

 Large fringe metro (2)

1.87**

1.35–2.59

< .001

1.61*

1.15–2.25

.005

1.49*

1.06–2.09

.021

 Medium metro (3)

2.60**

1.76–3.84

< .001

2.20**

1.47–3.30

< .001

2.03**

1.34–3.07

< .001

 Small metro (4)

2.75**

1.79–4.20

< .001

2.00*

1.29–3.12

.002

1.89*

1.21–2.95

.005

 Micropolitan (5)

3.88**

2.27–6.64

< .001

3.27**

1.88–5.70

< .001

3.00**

1.70–5.30

< .001

 Noncore (6)

5.28**

3.17–8.79

< .001

4.22**

2.49–7.15

< .001

3.74**

2.19–6.40

< .001

 Male

   

2.68**

2.00–3.59

< .001

2.93**

2.16–3.97

< .001

 White

   

2.54**

2.01–3.21

< .001

2.56**

2.02–3.26

< .001

 Age (y)

   

.98**

.98–.99

< .001

.98**

.98–.99

< .001

 Alcohol > .08%

      

.98

.82–1.17

.834

 Hunting Season

      

1.18

.79–1.77

.423

  1. NCHS Rurality National Center for Health Statistics Urban-Rural Classification Scheme for Counties. Hunting Season includes annual weeks 49–50. N’s range from 3675 to 3931 due to occasional missing data
  2. CI confidence interval, OR odds ratio calculated by multivariate logistic regression
  3. * p < .05 ** p < .001