Skip to main content

Table 2 Negative binomial regression – main coefficients

From: Where did civil servants go? the effect of an increase in public co-payments on double insured patients

  Coefficient (SE) IRR (95 % CI)
2012 -0.167 (0.030)*** 0.924 (0.868 – 0.983)**
ADSE - 2.666 (0.098)*** 0.076 (0.062 – 0.092)***
2012 ADSE -0.176 (0.038)*** 0.839 (0.779 – 0.903)***
alpha 0.004 (0.001)***
Pseudo R2 0.3504
  1. Note: These are the results of the negative binomial regression. The dispersion parameter, alpha, is significantly greater than zero based on a likelihood-ratio chi-square test. The number of observations is 104 (number of ED visits/week). Column 2 presents the average marginal effects transformed into semi-elasticities: these give an approximation of how much ED demand is expected to increase or decrease for a unit change in the independent variable. Robust standard errors are shown in parentheses. Column 3 presents the multiplicative effects in IRR. Statistical significance is denoted as follows: **p < 5 %.***p < 1 %