Table 4 Estimated regression coefficients, relative infection risks and 95% credible intervals for the effects of each risk factor on COVID-19 infection risk. The relative risks relate to realistic increases in each covariate, which are given in brackets in column 1 of the table.
From: A Bayesian spatio-temporal model of COVID-19 spread in England
Variable | Regression coefficient | Relative risks (95% CI) |
|---|---|---|
Annual household income (\(\pounds\)1000) | 0.0008 | 1.0008 (1.0005, 1.0012) |
Unemployment rate (1%) | 0.0027 | 1.0027 (1.0024, 1.0030) |
Log(population density) | 0.0146 | 1.0146 (1.0129,1.0164) |
Percent of Chinese (1%) | − 0.0160 | 0.9841 (0.9822, 0.9858) |
Percent of Indian (1%) | − 0.0005 | 0.9995 (0.9992, 0.9998) |
Percent of Pakistani (1%) | − 0.0016 | 0.9984 (0.9982, 0.9990) |
Percent of Bangladeshis (1%) | − 0.0014 | 0.9986 (0.9982, 0.9990) |
Percent of African (1%) | − 0.0093 | 0.9907 (0.9901, 0.9913) |
Percent of Caribbean (1%) | 0.0022 | 1.0022 (1.0009, 1.0036) |
Percent of age 18–29 (1%) | − 0.0050 | 0.9950 (0.9947, 0.9954) |
Percent of age 45–64 (1%) | 0.0031 | 1.0031 (1.0024, 1.0039) |
Percent of 65 years old and over (1%) | − 0.0066 | 0.9934 (0.9929, 0.9938) |
Annual mean \(\text {PM}_{2.5}\) (1 µg \(\text {m}^{-3}\)) | 0.0125 | 1.0126 (1.0083, 1.0167) |
Care home beds (0.01) | 1.3295 | 1.0134 (1.0121, 1.0147) |
Emergency facilities (TRUE) | 0.0007 | 1.0007 (0.9958, 1.0057) |
