Table 4 Estimates effects of long-term PM_2.5 exposure in quartile and Temperature on COVID-19 mortality rate using the negative binomial regression model, adjusted for all potential confounders (From 01May to 31 December, 2020)

Annual average Long-term PM_2.5 exposure (μg/m³) in quartile (ref. = Q1 (7.2 μg/m³))

Q2

(9.0 μg/m³)

1.134 (0.758–1.699)

1.144 (0.779–1.680)

1.134 (0.775–1.660)

1.127 (0.771–1.649)

1.126 (0.773–1.640)

1.144 (0.796–1.643)

1.105 (0.807–1.513)

1.142 (0.876–1.490)

1.190 (0.922–1.535)

Q3

(10.7 μg/m³)

2.408^b (1.508–3.845)

2.302^b (1.474–3.597)

2.307^b (1.484–3.586)

2.295^b (1.476–3.568)

2.317^b (1.498–3.582)

2.234^b (1.467–3.404)

1.624^b (1.129–2.336)

1.366^a (1.007–1.853)

1.469^b (1.099–1.963)

Q4

(13.4 μg/m³)

2.797^b (1.439–5.436)

2.678^b (1.414–5.070)

2.697^b (1.433–5.074)

2.682^b (1.424–5.052)

2.680^b (1.433–5.012)

2.596^b (1.416–4.760)

1.712^a (1.013–2.893)

1.141 (0.734–1.776)

1.140 (0.749–1.736)

Annual average Temperature over 12 years

0.919 (0.835–1.012)

0.914 (0.835–1.001)

0.914^a (0.836–1.000)

0.915 (0.837–1.001)

0.918 (0.841–1.003)

0.921 (0.846–1.004)

0.903^b (0.839–0.972)

0.867^b (0.814–0.923)

0.855^b (0.805–0.908)

Number of intensive care and resuscitation beds (per 100,000)

1.015 (0.991–1.039)

1.010 (0.988–1.033)

1.009 (0.987–1.031)

1.008 (0.986–1.031)

1.008 (0.986–1.030)

1.006 (0.985–1.027)

1.006 (0.987–1.024)

1.000 (0.984–1.015)

1.001 (0.986–1.017)

Medical density (per 100,000)

0.998 (0.996–1.001)

0.999 (0.997–1.001)

0.999 (0.997–1.001)

0.999 (0.997–1.002)

0.999 (0.997–1.002)

1.000 (0.998–1.002)

1.000 (0.998–1.002)

1.001 (0.999–1.002)

1.001 (0.999–1.002)

% People aged 60 or more

1.073^a (1.009–1.140)

1.078^a (1.017–1.144)

1.080^b (1.019–1.145)

1.083^b (1.021–1.148)

1.079^a (1.018–1.143)

1.079^b (1.020–1.141)

1.032 (0.986–1.081)

1.017 (0.977–1.059)

1.013 (0.975–1.053)

% Males

1.436 (0.986–2.092)

1.494^a (1.040–2.147)

1.530^a (1.068–2.191)

1.561^a (1.091–2.235)

1.559^a (1.093–2.223)

1.543^a (1.092–2.180)

1.456^a (1.080–1.963)

1.308^a (1.010–1.694)

1.278 (0.997–1.637)

% Unemployment

1.027 (0.889–1.186)

1.014 (0.884–1.163)

1.015 (0.886–1.163)

1.015 (0.886–1.164)

1.020 (0.891–1.167)

0.989 (0.870–1.126)

0.997 (0.891–1.115)

1.008 (0.914–1.111)

1.014 (0.924–1.114)

Rate of Poverty (per cent)

1.000 (0.910–1.100)

1.000 (0.914–1.094)

1.002 (0.917–1.096)

1.002 (0.916–1.095)

1.000 (0.916–1.093)

1.018 (0.935–1.109)

1.017 (0.944–1.095)

1.006 (0.944–1.072)

1.003 (0.944–1.065)

% Urban population (proportion of people living in the great urban areas)

1.025^b (1.009–1.040)

1.024^b (1.009–1.039)

1.024^b (1.010–1.039)

1.025^b (1.010–1.040)

1.024^b (1.010–1.039)

1.025^b (1.011–1.039)

1.014^a (1.003–1.026)

1.009 (0.999–1.019)

1.006 (0.996–1.015)

Population density (inhab/square)

1.000^b (1.000–1.000)

1.000^b (1.000–1.000)

1.000^b (1.000–1.000)

1.000^b (1.000–1.000)

1.000^b (1.000–1.000)

1.000^a (1.000–1.000)

1.000^a (1.000–1.000)

1.000^a (1.000–1.000)

1.000^a (1.000–1.000)

Standardized Prevalence of Diabetes (%)

1.191 (0.831–1.708)

1.235 (0.879–1.735)

1.247 (0.890–1.745)

1.253 (0.895–1.753)

1.250 (0.897–1.741)

1.244 (0.901–1.717)

1.288 (0.973–1.705)

1.353^a (1.066–1.716)

1.352^b (1.077–1.698)

Criteria for assessing goodness of fit:

Deviance

1.1782

1.2091

1.2114

1.2134

1.2086

1.2147

1.1827

1.1982

1.1927

Scaled Deviance

1.1782

1.2091

1.2114

1.2134

1.2086

1.2147

1.1827

1.1982

1.1927

Pearson Chi-Square

1.2419

1.2275

1.2168

1.2148

1.2164

1.1875

1.1898

1.1747

1.1834

Scaled Pearson X2

1.2419

1.2275

1.2168

1.2148

1.2164

1.1875

1.1898

1.1747

1.1834

BIC (smaller is better)

733.6727

769.6559

775.9792

779.4617

781.455

789.3626

813.6141

885.5104

927.3749