Skip to main content

Table 3 Logistic Generalised Estimating-Equation (GEE) and Alternating Logistic Regression (ALR) model of the probability of a false positive result to the IFNγ test.

From: Identification of risk factors associated with disclosure of false positive bovine tuberculosis reactors using the gamma-interferon (IFNγ) assay

Variable Label b S.E. OR Low Upper p-value1 p-value2
GEE model
Constant   -2.55 0.41      
Age (years) 0.5-1 Referent       0.046
2 -0.20 0.25 0.82 0.50 1.34 0.431
3-4 0.17 0.26 1.19 0.71 1.98 0.518
≥ 5 0.33 0.19 1.40 0.96 2.03 0.080
Region3 NE Referent       0.035
NW -0.36 0.43 0.70 0.30 1.62 0.400
SE -0.26 0.56 0.77 0.26 2.30 0.643
SW 0.80 0.41 2.23 1.00 4.98 0.051
Ρ   0.068      
ALR model
Constant   -2.53 0.40      
Age (years) 0.5-1 Referent       0.045
2 -0.19 0.25 0.83 0.51 1.35 0.445
3-4 0.16 0.27 1.18 0.70 1.99 0.542
≥ 5 0.34 0.20 1.40 0.95 2.06 0.088
Region3 NE Referent       0.026
NW -0.40 0.42 0.67 0.29 1.53 0.343
SE -0.34 0.51 0.71 0.26 1.95 0.511
SW 0.76 0.40 2.13 0.97 4.71 0.061
logOR   0.592 0.155     <0.001
  1. b coefficient, S.E Standard Error, OR Odds ratio.
  2. 1Wald test. 2 Generalised score test. 3 NE (North-East), NW (North-West), SE (South-East), SW (South-West).