Skip to main content
Figure 3 | International Journal of Health Geographics

Figure 3

From: Proof of concept of a method that assesses the spread of microbial infections with spatially explicit and non-spatially explicit data

Figure 3

Non-spatial and composite (non-spatial and spatial) microbial diffusion. The number of EP-specific interfarm isolates correlated significantly between repeated and single testing of the same animals (r = .87, P < 0.01, as indicated in Table 4, columns B and D, a). Four indices that estimate bacterial diffusion are described: the (single-cow testing based) number of isolates/EP (as indicated in Table 4, column E, b), the distance assumed each EP disseminated over time (EP distance, as indicated in Table 2, column C, c); the EP diffusion velocity (EPspeed, as indicated in Table 2, column E, d); and the product of (fully adjusted) interfarm isolates/EP and EP speed (the fully adjusted geo-temporal or EPGeotemp index, as indicated in Table 4, column N, e). Diagonal lines indicate the expected distribution under the hypothesis of linearity. The null hypothesis of lack of normality was rejected (Ryan-Joiner [RJ] test > 0.05). Significant correlations were observed (i) between the isolates/EP index (adjusted for single-cow testing, number of farms, and time) and EPdist (r = .59, P < 0.02, as indicated in Table 4, columns J and L, f), (ii) between EPdist and EPspeed (r = .72, P < 0.01, as indicated in Table 2, columns C and E, g), and (iii) between both versions of the EPGeotemp index (non-repeated vs. repeated testing of the same cow, r = .98, P = 0.01, as indicated in Table 4, columns M and N, h). Numbers in italics identify EPs above the 75th percentile or displaying the highest values.

Back to article page