Table 4.
Logistic mixed model results using fast (y=1) or normal (y=0) outbreaks as outcome, with a cut-off <16 or ≥16 days from primary case for categories definition, based on Nearest-neighbor kernel function.
| Outcome = fast outbreak detected by Nearest-neighbor kernel function | |||
|---|---|---|---|
| Variable | ORadj | 95% CI | p |
| Type of farm | |||
| Indoor | Ref. | ||
| Outdoor | 1.36 | 1.12–3.77 | 0.044 |
| Distance < 3.87 km | 8.56 | 4.90–14.98 | <0.0001 |
| Days for outbreak confirmation | 1.03 | 1.01–1.05 | 0.004 |
| N. animals with symptoms | 1.08 | 1.02–1.11 | 0.005 |
| Type of fence | Ref. | ||
| Fenced§ | Ref. | ||
| Not fenced | 2.65 | 1.90–3.69 | <0.0001 |
| N. pigs | |||
| >10 ≤10 |
Ref. 0.53 |
0.31–0.89 |
0.016 |
| Relationship with other farms | |||
| Yes Not |
Ref. 0.49 |
0.25–0.96 |
0.028 |
| Disposable clothing | |||
| Not Yes |
Ref. 0.81 |
0.70–0.94 |
0.005 |
| Farmer as a hunter | |||
| Yes Not |
Ref. 0.45 |
0.25–0.83 |
0.002 |
| Random effect | Est | SE | 95% CI |
| Cluster | 1.29 | 0.28 | 0.83–1.98 |
| LR test vs. logistic regression: | 30.54, p < 0.0000 | ||
| Residual mean (SD) | 3.32 * 10−6 (1.12 * 10−6) | ||
| Spearman's correlation coefficient | 0.850, p < 0.0001 | ||
| Root MSE training dataset | 0.191, p = 0.63 | ||
| Root MSE test dataset | 0.188, p = 0.72 | ||
Data are presented as adjusted Odds Ratio (ORadj), 95% confidence intervals (95% CI) and p-value.
§
Fenced is referred to all types of fences (double, single solid or metal fences).