Table 1.

Association between the risk of upper tract urothelial carcinoma and consumption of aristolochic acid-containing herbal medicines in arseniasis-endemic and non-endemic areas

		Controlsa		Cases		P-value	Crude OR	95%CI		P-value	Adjusted ORb	95%CI		P-value
		n	%	n	%
Whole population	Subject number	11684		2921
	AA consumption (mg)					<0.0001
	0	6744	57.72	1447	49.54		1.00				1.00
	1–100	4241	36.30	1154	39.51		1.28	1.18	1.40	<0.0001	1.26	1.15	1.38	<.0001
	101–200	377	3.23	167	5.72		2.10	1.73	2.54	<0.0001	2.05	1.69	2.49	<.0001
	> 200	322	2.76	153	5.24		2.24	1.83	2.74	<0.0001	2.21	1.80	2.70	<.0001
							P-value for trend < 0.0001				P-value for trend < 0.0001
Arsenic grade 1–3 areas	Subject number	2252		563
	AA consumption (mg)					0.0002
	0	1353	60.08	282	50.09		1.00				1.00
	1–100	723	32.10	220	39.08		1.49	1.22	1.83	0.0001	1.48	1.20	1.81	0.0002
	101–200	81	3.60	26	4.62		1.61	1.01	2.57	0.0458	1.57	0.98	2.51	0.0587
	> 200	95	4.22	35	6.22		1.82	1.20	2.75	0.0046	1.80	1.19	2.72	0.0056
							P-value for trend < 0.0001				P-value for trend < 0.0001
Arsenic grade 0 areas	Subject number	9432		2358
	AA consumption (mg)					<0.0001
	0	5391	57.16	1165	49.41		1.00				1.00
	1–100	3518	37.30	934	39.61		1.24	1.12	1.37	<.0001	1.22	1.10	1.34	0.0001
	101–200	296	3.14	141	5.98		2.22	1.79	2.74	<.0001	2.17	1.76	2.69	<.0001
	> 200	227	2.41	118	5.00		2.41	1.91	3.04	<.0001	2.36	1.87	2.98	<.0001
							P-value for trend < 0.0001				P-value for trend < 0.0001

AA, aristolochic acid; AA consumption is defined as the cumulative dose of AA from 1997 to 2003 or to one year prior to the diagnosis of upper urinary tract urothelial carcinoma.

^aFactors for matching controls: sex, age, economic status, renal insufficiency, hepatitis, chronic obstructive pulmonary disease, diabetes mellitus, hyperlipidemia

^badjusted for use of aspirin, NSAIDs, and acetaminophen

The relative excess risk due to interaction (RERI) = 0.12, 95% CI. −0.13 to 0.39, p = 0.339