Skip to main content

Table 1 Example 1 Model Results

From: Reference effect measures for quantifying, comparing and visualizing variation from random and fixed effects in non-normal multilevel models, with applications to site variation in medical procedure use and outcomes

 

Estimate

SE

P-value

OR

95% CI for OR

(Intercept)

−2.292

0.321

<  0.001

0.101

(0.054, 0.190)

Patient Risk

Alcohol Use

−0.358

0.061

<  0.001

0.699

(0.620, 0.788)

Liver Disease

−0.309

0.100

0.002

0.734

(0.604, 0.892)

Age (10 Years)

−0.267

0.017

<  0.001

0.766

(0.741, 0.791)

Drug Abuse

−0.202

0.098

0.039

0.817

(0.675, 0.990)

…

     

White

0.342

0.055

<  0.001

1.407

(1.263, 1.568)

Prior AC

0.368

0.070

<  0.001

1.445

(1.261, 1.656)

CHF

0.409

0.039

<  0.001

1.505

(1.395, 1.623)

Hospital Risk

Rural Prop.

−0.014

0.024

0.569

0.986

(0.941, 1.034)

…

     

EP Onsite

0.084

0.127

0.509

1.088

(0.848, 1.395)

Hospital SD

0.511

0.040

   
  1. Standard logistic regression output for initiation of rhythm control treatment (y/n) for 29,759 AF patients at 124 VA hospitals during the period October 1, 2001 to September 30, 2012. Some covariates included in the model have been omitted from the table and figures to simplify presentation
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us