Table 1.

Poisson regression model between the air pollutant and hemorrhagic fever with renal syndrome

Time lag	Univariate		Multivariate*
	RR (95% CI)	P	RR (95% CI)	P
No time lag	0.998 (0.995–1.000)	.025	1.013 (1.008–1.017)	< .001
1–month lag	0.972 (0.970–0.975)	< .001	1.001 (0.997–1.004)	.785
2–month lag	0.938 (0.935–0.940)	< .001	0.991 (0.987–0.995)	< .001
3–month lag	0.933 (0.931–0.936)	< .001	0.983 (0.979–0.987)	< .001
4–month lag	0.964 (0.961–0.966)	< .001	0.992 (0.988–0.996)	< .001
5–month lag	0.997 (0.995–1.000)	.022	0.991 (0.988–0.995)	< .001
6–month lag	1.024 (1.022–1.026)	< .001	1.005 (1.002–1.008)	.001
7–month lag	1.036 (1.034–1.037)	< .001	1.006 (1.004–1.009)	< .001
8–month lag	1.031 (1.030–1.033)	< .001	1.002 (1.000–1.005)	.036
9–month lag	1.016 (1.014–1.018)	< .001	0.999 (1.012–1.031)	.360
10–month lag	1.005 (1.002–1.007)	< .001	0.993 (0.989–0.997)	.001
11–month lag	1.000 (0.998–1.002)	.827	0.997 (0.993–1.002)	.208
12–month lag	0.990 (0.988–0.992)	< .001	0.999 (0.995–1.003)	.572

CI confidence interval, RR relative risk.

Dependent variable is the occurrence of hemorrhagic fever with renal syndrome.

^*Adjusted for seasonality and climate variables.