Table 2.
The results of the multivariable linear regression
| Covariate | Estimate | Standard error | P |
|---|---|---|---|
| Intercept | 0.090 | 1.9 × 10−4 | <0.001 |
| SE22 (Sydsverige) | Baseline | ||
| SE21 (Småland med öarna) | −0.039 | 2.3 × 10−4 | <0.001 |
| SE23 (Västsverige) | −0.037 | 2.3 × 10−4 | <0.001 |
| SE12 (Östra Mellansverige) | −0.021 | 2.3 × 10−4 | <0.001 |
| SE11 (Stockholm) | −0.054 | 2.3 × 10−4 | <0.001 |
| SE31 (Norra Mellansverige) | −0.053 | 2.3 × 10−4 | <0.001 |
| SE32 (Mellersta Norrland) | −0.064 | 2.3 × 10−4 | <0.001 |
| SE33 (Övre Norrland) | −0.047 | 2.3 × 10−4 | <0.001 |
| Quarter 1 | Baseline | ||
| Quarter 2 | −0.011 | 1.6 × 10−4 | <0.001 |
| Quarter 3 | −0.002 | 1.6 × 10−4 | <0.001 |
| Quarter 4 | 0.011 | 1.6 × 10−4 | <0.001 |
The coefficients to assess the relationship between the proportion infected holdings and the NUTS 2 region and quarter of the year. The multivariable regression is calculated from the outcome in a stochastic simulation to explore the spread of Verotoxigenic Escherichia coli O157:H7 (VTEC O157) in the entire Swedish cattle population based on data reported to the Swedish Board of Agriculture during the period 2005-07- 01 to 2013-12-31.