Association between fatality rate of COVID-19 and selenium deficiency in China

BMC Infectious Diseases

Table 2 Association between case fatality rate of COVID-19 and the selenium level in China

City selenium level^a	Crude CFR (%) ^b	P value^b	Zero-inflated negative binomial regression^d
City selenium level^a	Crude CFR (%) ^b	P value^b	IRR (95% CIs) ^c	P value
Selenium content in crops (ppm)
> 0.06 (non-selenium-deficient areas)	1.17	0.002	1	–
0.03–0.06 (moderate-selenium-deficient areas)	1.28		1.92 (1.00–3.69)	0.051
< 0.03 (severe-selenium-deficient areas)	3.16		3.88 (1.21–12.52)	0.023
Selenium content in topsoil (mg/kg)
> 0.31 (non-selenium-deficient areas)	0.76	< 0.001	1	–
0.18–0.31 (moderate-selenium-deficient areas)	1.70		2.38 (1.14–4.98)	0.021
< 0.18 (severe-selenium-deficient areas)	1.85		3.06 (1.49–6.27)	0.002

^aWe included only 147 cities from which at least 20 cases were reported
^bThe crude case fatality rates (CFRs) were calculated by dividing the number of deaths by the number of cases within the region. The P values were calculated by chi-square tests
^cIRR (95% CIs): incidence rate ratio with 95% confidence intervals
^dThe Zero-inflated negative binomial regression were used for evaluating the association between the selenium level and case fatality rate at the city level. Only IRR (95%CIs) of count model (negative binomial with “log” link) was shown in the table, and the multivariate analysis had the same results with the univariate analysis due to non-significant effects of demographic, social and medical access related covariates, including population density, GDP per capita, proportion of population over 60 years old, number of hospitals, number of hospital beds, and number of clinical staffs per 1000 people on the case fatality rate for COVID-19

ISSN: 1471-2334