Table 3.

Statistical model coefficients for the relationship between weather variables and prevalence of bacteria along the agri-food chain in Canada from 2002 through 2012

Outcome	Independent variables
	Temperature (°C)a	Variable temperature (°C)a	Maximum temperature (°C)a	Precipitation (mm)b	Variable precipitation (mm)b
Poultry
Salmonella Retail	NS	NS	NS	NS	NS
Salmonella Abattoir	NS	NS	−0.00884 (11)	0.00174 (55)	NS
Campylobacter Retail	0.0105 (9)	NS	NS	NS	NS
Campylobacter Abattoir	NS	−0.0817 (1.4)	0.0268 (3.6)	NS	NS
Beef
E. coli Retail	NS	NS	NS	0.00312 (31)	NS
Swine
Salmonella Retail	NS	NS	NS	NS	NS
Salmonella Abattoir	NS	NS	NS	NS	0.00243 (40)
Salmonella On-farm	NS	NS	−0.0146 (7)	NS	0.00424 (23)
Campylobacter Retail	0.00817 (10)	NS	NS	NS	NS
Campylobacter Abattoir	NS	−0.1365 (0.7)	0.0299 (3.5)	NS	NS
E. coli Retail	0.0173 (6)	0.026 (3.7)	NS	NS	NS

NS, No significant relationship at significance level of 5%.

Coefficients are shown, followed by the required absolute change in the weather variable in parentheses required to increase (if positive coefficient) or decrease (if negative coefficient) odds of a positive sample by 10% (keeping all other factors constant). Only categories with sufficient data for analysis are shown (see text).

^aIncrease in C temperature required to increase (if + coefficient) or decrease (if − coefficient) odds of a positive sample by 10% (keeping all other factors constant).

^bIncrease in mm of precipitation required to increase (if + coefficient) or decrease (if − coefficient) odds of a positive sample by 10% (keeping all other factors constant).