Skip to main content

Table 2 OR estimates and 95% confidence intervals from Poisson and logistic regression models of physician visits for the Australian Health Survey data

From: Odds ratios from logistic, geometric, Poisson, and negative binomial regression models

  Poisson Logistic Relative widthd
Covariate Estimate Interval Estimate Interval  
Female 1.28 (1.11, 1.47) 1.30 (1.11, 1.53) 85.85%
Agea 1.67 (1.10, 2.52) 1.71 (1.06, 2.74) 84.71%
Incomeb 0.82 (0.66, 1.01) 0.95 (0.75, 1.20) 76.49%
Private insurancec 1.17 (0.98, 1.39) 1.30 (1.07, 1.59) 78.51%
Free government insurance (low income)c 0.55 (0.36, 0.85) 0.50 (0.30, 0.84) 89.64%
Free government insurance (old age, disability, veteran)c 1.20 (0.95, 1.53) 1.53 (1.16, 2.01) 68.28%
Number of illnesses in past two weeks 1.30 (1.24, 1.36) 1.32 (1.25, 1.39) 85.36%
Number of days of reduced activity in past two weeks 1.24 (1.21, 1.26) 1.17 (1.14, 1.20) 78.17%
General health questionnaire score 1.05 (1.02, 1.08) 1.06 (1.03, 1.10) 83.68%
Has chronic condition that limits activity 1.17 (0.98, 1.41) 1.19 (0.96, 1.48) 84.20%
  1. a Age in years divided by 100 b Annual income in tens of thousands of dollars c Reference category: government Medibank insurance d Poisson compared to logistic model