Skip to main content

Table 1 Multivariable negative binomial regression model results estimating associations between explanatory variables and E. coli BSI ratesa,b

From: Increasing incidence and antimicrobial resistance in Escherichia coli bloodstream infections: a multinational population-based cohort study

Variable aIRR 95% CI p-value
Region     < 0.001
 Finland 1.00 Referent  
 Calgary 0.71 0.65–0.77  < 0.001
 Canberra 0.71 0.65–0.79  < 0.001
 Sherbrooke 0.86 0.77–0.96  < 0.001
 Skaraborg 1.05 0.96–1.15 0.312
 Western interior 0.73 0.65–0.82  < 0.001
Sex    
 Female 1.00 Referent  
 Male 0.74c 0.67–0.80  < 0.001
Age category    
 < 70-years-old 1.00 Referent  
 ≥ 70-years-old 9.37c 8.63–10.19  < 0.001
Interaction—sex and age   
 Male and ≥ 70 1.39c 1.23–1.57  < 0.001
  1. BSI Bloodstream infection; aIRR Adjusted incidence rate ratio; CI Confidence interval
  2. aOverdispersion parameter 0.014, 95% CI:0.008–0.025, p < 0.001
  3. bModel fit the data based on normally distributed Anscombe residuals and non-significant deviance goodness-of-fit test (p = 0.24)
  4. cExponentiated coefficients are not true aOR due to interaction term—see contrasts in Table 2