Skip to main content
. 2016 Jul 14;144(15):3148–3165. doi: 10.1017/S0950268816001424

Table 4.

Mixed logistic regression of the binomial outcome that a herd test was positive for antibodies against influenza A(H1N1)pdm09 virus infection based on haemagglutination inhibition test

No. of observations (positives) Fixed effects* OR 95% CI P value
Year of test (n = 5)
490 (204) 2010 1
1288 (563) 2011 1·02 0·78–1·33 0·89
1361 (641) 2012 1·07 0·82–1·39 0·62
1401 (587) 2013 0·82 0·64–1·07 0·14
1103 (478) 2014 0·88 0·67–1·15 0·35
Production class (n = 5)
305 (69) Fattening herd 1
362 (123) Nucleus herd 2·78 1·67–4·62 <0·001
627 (297) Multiplier herd 4·72 3·05–7·30 <0·001
4093 (1826) Conventional herd 2·63 1·87–3·70 <0·001
256 (155) Sow pool 24·04 10·98–52·66 <0·001
Herd size in three quantiles (pigs)
1679 (601) Small (<350 pigs) 1
1909 (919) Medium (350–665 pigs) 1·85 1·52–2·26
1851 (881) Large (>665 pigs) 1·63 1·33–1·99
Constant 0·50 0·33–0·75
Random effects
5643 (2473) County (n = 19)
Var(const.) 4·45 2·16–9·15
5643 (2473) County>herd id (n = 1567)
Var(const.) 1·66 1·29–2·15

OR, Odds ratio; CI, confidence interval.

*

Three categorical fixed effects were: (1) year of test, (2) production class, and (3) herd size (based on national registry for subsidy).

County and herd ID were included as random effects to account for non-fixed effects associated with county and the individual herd.