Generating social network data using partially described networks: an example informing avian influenza control in the British poultry industry

Table 5 Results from multivariable logistic regression: scenario 1 analyses (n = 348 farms)

Farm-level predictor	Predictor levels^a	OR^b	lower 95% CI^c	upper 95% CI^c	p-value
Intercept^d	-	0.608	0.264	1.401	0.243
House count/Bird count	hbSS^e	1	-	-	-
	hbLS	0.161	0.041	0.636	0.009
	hbSL	0.507	0.207	1.246	0.139
	hbLL	0.317	0.141	0.711	0.005
Indoor/Free-range	ifYN^e	1	-	-	-
	ifNN	1.810	0.275	11.927	0.537
	ifNY	12.185	3.815	38.913	<0.001
	ifYY	3.072	0.725	13.018	0.128
Integration status	Non-integrated^e	1
	Integrated	0.681	0.307	1.512	0.345
Geographical location	Region: North^e	1	-	-	-
	Region: East	1.337	0.567	3.152	0.507
	Region: Scotland	0.324	0.038	2.752	0.302
	Region: Wales	0.126	0.012	1.329	0.085
	Region: West	0.321	0.135	0.761	0.010
Interaction terms	ifNN*Integrated	0.900^f	0.075	10.778	0.934
	*ifNY Integrated**	0.128 ^f	0.028	0.594	0.009
	ifYY* Integrated	0.965^f	0.048	19.526	0.982

^aSee tables 3 and 4 for definitions; ^bOR = odds ratio; ^cCI = confidence interval of the OR; ^daverage odds in the baseline predictor groups; ^ereference-level category; ^fOR for farms in the baseline category of the other interacting variable; results significant at 5% error level are indicated in bold.

ISSN: 1746-6148