Skip to main content
. 2021 Mar 24;8:661063. doi: 10.3389/fvets.2021.661063

Table 1.

Results of general additive modeling (GAM) explaining the number of outbreaks of infectious diseases, zoonotic or vector-borne, per country and per year, with forest cover or oil palm areas, human population size and the matrix of longitude/latitude of country centroids as independent variables, using a negative binomial link and controlling for spatial and temporal effects, with approximate significance of smooth terms (edf, estimated residual degrees of freedom; df, degree of freedom) (see Methods for the models, data from GIDEON, FAOSTAT and World Bank) (see Figure 3).

Response variable Explanatory variables edf Chi2 (df) P-value Deviance (in%), R2 adjusted, n
1. Outbreaks of zoonotic diseases (Forest cover, year) 19.7 106 (29) <0.0001
(Population size, year) 19.9 1,250 (28) <0.0001
(Longitude, latitude) 24.1 645 (29) <0.0001 53.4, 0.65, 5311
2 Outbreaks of vector-borne diseases (Forest cover, year) 13.4 52.5 (29) <0.0001
(population size, year) 17.0 619.8 (28) <0.0001
(Longitude, latitude) 21.2 348.1 (29) <0.0001 40.5, 0.49, 4784
3. Outbreaks of zoonotic diseases (Oil palm area, year) 0.001 0.0 (29) 0.54
(Population size, year) 15.1 488 (28) <0.0001
(Longitude, latitude) 11.7 161 (29) <0.0001 55.1, 0.65, 1126
4. Outbreaks of vector-borne diseases (Oil palm area, year) 1.74 6.79 (29) 0.0029
(Population size, year) 11.05 127.34 (28) <0.0001
(Longitude, latitude) 11.25 81.74 (29) <0.0001 49.3; 0.54, 1072