Table 3 Summary of the relative contributions (%) of predictor variables for the SFTS data in the boosted regression trees model

From: A National Assessment of the Epidemiology of Severe Fever with Thrombocytopenia Syndrome, China

Variable

Boosted regression trees

Relative contribution (mean)

Relative contribution (sd)

Temperature*

12.44

1.72

Rainfall*

13.82

1.39

Relative humidity*

8.29

1.97

Sunshine hours*

13.36

2.14

Elevation*

19.15

2.31

Percentage coverage of forest*

5.16

1.22

Percentage coverage of shrub

2.57

1.04

Percentage coverage of cropland

3.85

1.18

Cattle density*

6.24

1.65

Goat density

2.43

0.73

Population density

2.52

1.24

Haemaphysalis longicornisa*

9.91

1.52

Rhipicephalus microplus a

0.26

0.20

  1. *Variables whose relative contribution in the BRT models more than 5 were considered to be significantly contributes to the occurrence of human infection with SFTSV.
  2. aHaemaphysalis longicornis and Rhipicephalus microplus are binary variables, whether the tick presence or absence in each county.
Back to article page