Table 4

Regression coefficients of CAR models
Geographical unit Maximal % of variation
Smallest Municipality Census consolidated subdvision Agriculture CLSC Census division Watershed
Variable (n=1119) (n=1063) (n=903) (n=319) (n=155) (n=97) (n=71)
Intercept 34.32 (1.83)* 35.98 (2.11)* 35.12 (2.37)* 39.54 (2.80)* 36.38 (3.97)* 23.35 (8.79)* 16.33 (6.34)*
Ruminant density per km2 (ref.=none)
>20 8.72 (2.04)* 7.17 (2.14)* 10.76 (2.57)* 4.47 (3.21) 4.75 (5.20) 17.37 (10.55) 16.00 (13.10) 50
≤20 5.44 (1.63)* 3.25 (1.76) 4.68 (2.21)* 3.21 (2.40) 2.96 (3.41) 15.05 (8.98) 8.88 (5.77) 16
Poultry density per km2 (ref.=none)
>250 3.44 (1.52)* 2.06 (1.53) 3.84 (1.50)* −1.32 (2.75) 1.24 (3.82) 4.17 (4.69) 19.02 (7.55)* 453
≤250 −0.28 (1.62) −1.05 (1.60) 1.68 (1.48) −4.62 (2.48) −1.15 (3.28) 1.36 (4.50) 18.38 (4.59)*
Slaughterhouse (ref.=none)
Large poultry 65.90 (6.75)* 59.24 (6.47)* 27.81 (5.91)* 24.68 (7.33)* 20.42 (5.91)* 15.99 (6.84)* 1.96 (7.14) 312
Others −1.18 (2.63) 0.94 (2.53) −1.40 (2.30) −2.51 (3.33) 1.72 (2.96) 1.52 (3.41) −5.74 (4.58)
Diploma in % (ref.=50-75)
<50 −3.80 (1.42)* −3.76 (1.41)* −4.25 (1.34)* −4.71 (3.09) −18.16 (6.49)* −19.68 (9.06)* −6.37 (4.38) 423
>75 −0.20 (1.54) −1.98 (1.64) −2.04 (1.78) −1.72 (2.50) −4.58 (2.56) −6.81 (4.13) 5.24 (6.00)
Population density per km2 (ref.=6-400)
≤6 −2.25 (1.47) −2.48 (1.49) −1.28 (1.38) −6.78 (4.09) n/ab n/a 3.64 (5.66)
>400 3.58 (1.50)* 4.59 (1.52)* 0.36 (1.53) 5.69 (2.16)* 0.91 (2.76) 2.93 (3.21) 0.68 (3.75) 59
Precipitation in mm (ref.=2.9-3.1)
<2.9 −3.91 (1.86)* −3.60 (2.40) −4.56 (2.07)* −9.59 (3.49)* −14.09 (4.52)* −15.96 (5.39)* −7.92 (4.51) 308
>3.1 1.33 (1.57) 1.09 (1.99) 1.19 (1.81) 3.18 (2.94) 8.89 (3.55)* 5.80 (4.31) 5.61 (5.35)
Temperature in °C (ref.=3.1-6.9)
<3.1 1.08 (1.65) −3.45 (2.18) −1.55 (1.88) 1.67 (3.08) 3.14 (3.39) 7.03 (3.90) 10.25 (4.05)*
>6.9 −9.50 (1.92)* −4.00 (2.57) −5.32 (2.61)* −10.42 (3.42)* −6.47 (3.87) −6.43 (5.06) −17.30 (8.73)* 225
Lambdac 0.58* 0.8* 0.68* 0.45* 0.59 0.56* 0.31*
Model r2 0.21 0.27 0.26 0.24 0.48 0.37 0.52

Regression coefficients (standard errors) of conditional autoregressive models predicting the annual incidence of campylobacteriosis per 100,000 people in Quebec, Canada, 1996–2006.

a Maximal percentage of variation in statistically significant estimates of regression coefficient based on different geographical units.

b Not applicable (absence of areas with those values for the specific unit set).

c Spatial error parameter adjusting for spatial correlation.

*p<0.05.

Arsenault et al.

Arsenault et al. BMC Infectious Diseases 2012 12:318   doi:10.1186/1471-2334-12-318

Open Data