E. coli O157 on Scottish cattle farms: Evidence of local spread and persistence using repeat cross-sectional data

Table 3 Multiple variate analysis of risk factors

Predictor	Estimate	SE	p
Farm positive in Survey 1	0.264	0.3120	0.398
Season			0.680
Spring	0.132	0.4293	0.758
Summer	0.478	0.4198	0.254
Autumn	0.220	0.4218	0.603
Winter	-	-	-
Farm in sampling cluster positive	1.719	0.6085	<0.001
Animal Health Division (AHD)			0.686
Island	−0.907	0.8698	0.298
Highland	0.128	0.6842	0.852
North East	−0.553	0.7340	0.451
Central	0.555	0.6139	0.367
South East	0.827	0.5956	0.166
South West	-	-	-
Farm in sampling cluster positive*AHD			0.004
Farm in sampling cluster positive in Islands	2.215	1.1426	0.053
Farm in sampling cluster positive in Highland	1.801	0.6558	0.006
Farm in sampling cluster positive in North East	2.441	0.7202	0.001
Farm in sampling cluster positive in Central	−0.373	0.6894	0.589
Farm in sampling cluster positive in South East	−0.417	0.5948	0.484
Farm in sampling cluster positive in South West	1.719	0.6085	0.005
Total number of cattle^a	1.401	0.3916	<0.001
No. of movements onto farm in last 8 weeks^a	0.774	0.3658	0.035
Overall OR ^b	2.61 (0.79)

Results of the logistic regression model of risk factors for the presence of E. coli O157 on the 447 Scottish farms that were sampled in both Survey 1 (SEERAD, 1998-2000) and Survey 2 (IPRAVE, 2002-2004). Overall OR gives empirical estimate of odds ratio for the entire model.
^alog₁₀ transformed.
^bMean (SD).

ISSN: 1746-6148