Table 4. Complete list of variables available for multivariable logistic regression analysis.

Variable	Definition	Source
BTBID	Identification number for each premises	Assigned
pp-value sqrtpp Fasciola_pp	Anti-F. hepatica antibody ELISA percentage positivity value Square-root transformation of pp-value Smoothed extracted value of sqrtpp from map	Ref 14 Calculated Calculated
BTBposneg	Whether a farm has tested positive for a bTB test during 2004-07	VETNET, DEFRA
NN_BTB	Log-odds of BTB prevalence of nearest neighbouring farm	Calculated
Total Reactors Total Confirmed	The total number of reactor animals found between 2004-07 The total number of animals confirmed as BTB positive on PME or culture 2004-07	VETNET, DEFRA
SqrtHerd	Herd Size, square-root transformed for this model	VETNET, DEFRA
NUTS2 NUTS1 AHOID PCArea	The NUTS2 classification of the premises The NUTS1 classification of the premises The Animal Health Office classification of the premises The postcode area of the premises	VETNET, DEFRA
NDVI EVI	Normalised Difference Vegetation Index Enhanced Vegetation Index	MODIS
SqrtMOff Sqrt(On+1)	Square-root transformation of cattle movements off the premises Square-root transformation of one plus cattle movements onto the premises	RADAR, DEFRA
MaxT MinT RainyDays RH VP	The maximum temperature recorded on the farm (°C) The minimum temperature recorded on the farm (°C) The mean number of rainy days (rainfall >1mm) per year Mean relative humidity on the farm Mean vapour pressure on the farm	MetOffice
MeanT	The mean temperature on the farm calculated as a mean of maximum and minimum	Calculated
VPD	Vapour Pressure Deficit	Calculated
TVar	Temperature variance, calculated from MODIS fourier curves	Calculated
EngWales	Location of a farm in either England or Wales	Assigned