Table 4.

Summary results from cohort studies reporting on arsenic exposure and the risk of bladder cancer

Study [reference] (Table from original publication)	Study locale	Outcome	ICD1	Arsenic exposure assessment	Exposure [comments]	Outcome measure	Cohort size	Cases	Risk estimate (95% CI)	Covariates assessed
Chen et al. 1988 [70] (Table Six)	4 neighbouring townships from Blackfoot disease (BFD) endemic area, Taiwan	Morality 1968-83	N/A	Group level	Median arsenic content of artesian well and (range): 0.78 ppm (0.35–1.14); in shallow well: 0.04 (0.00-0.30). General population used as reference. 95% CI obtained from IARC 2012 review [23].	SMR	871	15	38.8 (21.7–64.0)
Chiou et al. 1995 [32] (Table Four)	4 neighbouring townships from BFD endemic area, Taiwan	Incidence 1988 (Follow-up period ranged 0.05 to 7.7 years)	N/A	Individual level ‘estimated’	Cumulative arsenic exposure (mg˙•L-1˙•year):	RR	2,556	29		age, sex, cigarette smoking
					0				1.0
					0.1–19.9				1.57 (0.44–5.55)
					> 20				3.58 (1.05–12.19)
					unknown				1.25 (0.38–4.12)
					[Median arsenic content of artesian well and (range): 0.78 ppm (0.35–1.14); in shallow well: 0.04 (0.00-0.30). Histories of residential address and duration of drinking well water used to derive cumulative exposure.]
*Tsuda et al.2 1995 [34] (Table Three)	Niigata, Japan	Mortality 1959-92 (Recruitment in 1959, followed until 1992)	Transitional cell carcinoma	Individual level ‘measured’	Arsenic water concentration (μg˙•L-1):	SMR	443			age, smoking habits
					< 50			254	0.00 (0–12.50)
					50 – 990			76	0.00 (0–47.05)
			ICD9 188, 189 ICDO histology N/A		≥ 1,000			113	31.18 (8.62–91.75)
					[Arsenic-polluted area. Exposure to be between 1955-59. All 34 wells in the area were sampled and arsenic concentration ranged from non detectable to 3,000 μg˙•L-1).]
Lewis et al. 19993[40] (Table Four)	Millard County in Utah, USA	Mortality (Recruitment 1900–1945)	N/A	Group level	Cumulative arsenic exposure derived from: low exposure (< 1000 ppb-year); medium (1,000-4,999 ppb-year); high (≥ 5,000 ppb-year):		4,058			Individual data on cofactors not available. However, the cohort was assembled from historical membership records of the Church of Jesus Christ of Latter-day Saints (Mormons) which prohibits tobacco use and the consumption of alcohol and caffeine.
						SMRmale		–	0.42 (0.08–1.22)
					< 1,000 ppb•year	SMRfemale		–	0.81 (0.10–2.93)
					≥ 5,000 ppb•year	SMRmale		–	0.4
					[Residential history combined with local water records used to assess exposure. High variability in exposure estimates in each community with median arsenic concentrations ranging from 14 to 166 ppb. Records of arsenic measurements dating back to 1964.]	SMRfemale		–	1.18
						SMRmale		–	0.95
						SMRfemale		–	1.10
*Chiou et al. 20013[33] (Table Five)	18 villages in four townships in Lanyang Basin, North-eastern Taiwan	Incidence 1991-1994 (Follow-up period from time of enrollment to Dec.1996)	Urinary organs	Individual level ‘estimated’	Arsenic water concentration (μg˙•L-1):	RR	8,102			age, sex, cigarette smoking, duration of well water drinking
					0–10.0	Urinary organs		3	1.0
			ICD9 188, 189
					10.1–50.0			3	1.5 (0.3–8.0)
					50.1–100.0			2	2.2 (0.4–13.7)
			Transitional cell carcinoma		> 100.0			7	4.8 (1.2–19.4)
					Arsenic water concentration (μg˙•L-1);	RR Transitional cell carcinoma
					0–10.0			1	1.0
			ICDO1 8120.2, 8120.3, 8130.3		10.1–50.0			1	1.9 (0.1–32.5)
					50.1–100.0			2	8.2 (0.7–99.1)
					> 100.0			6	15.3 (1.7–139.9)
					[Arsenic levels in shallow well ranging from < 0.15 to 3,590 μg˙•L-1 and collected from 3,901 well water samples between 1991–94.]
† Baastrup et al. 2008 [96] (Table Three)	23 municipalities in Copenhagen & Asrhus areas, Dannemark	Incidence 1993-1997 (Follow-up from enrollment until date of first cancer diagnosis, emigration, death, or Aug. 2003)	N/A	Individual level ‘estimated’	Cumulated arsenic exposure (5 mg˙):	IRR	56,378	214	1.0 (0.98–1.04)	smoking status, smoking duration, smoking intensity, education, occupation
					Time-weighted average exposure (μg˙•L-1):	IRR		214	1.01 (0.93–1.11)
					[Average arsenic exposure from 0.05 to 25.3 μg˙•L-1, with mean of 1.2 μg˙•L-1. Average arsenic concentrations obtained from 4,954 samples from 2,487 water utilities collected, 1987–2004, with most samples dating 2002–04. Residential history 1970–2003.]
*†Huang et al. 2008 [53] (Table Two)	3 villages in Putai Township, in BFD endemic area of southern Taiwan	Incidence 1989 (Average follow-up period of 12 years)	Urothelial carcinoma	Individual level ‘estimated’	Arsenic water concentration (μg˙•L-1):	RR	1,078			age, sex, cigarette smoking, education
					0–400			1	1.0
			ICDO3 M-codes 8120/3, 8230/3		401–700			14	5.2 (0.7–39.8)
					710–900			9	6.7 (0.8–53.4)
					≥ 900			7	6.5 (0.8–53.1)
					Cumulative arsenic exposure (mg˙•L-1•year):	RR
					0			0	–
					0.1–11.9			2	1.0
					12.0–19.9			9	4.6 (1.0–21.8)
					≥ 20.0			20	7.9 (1.7–37.9)
					[Period of arsenic water samples collection not reported. Participants used artesian well water more > 30 years when recruited. Information from interview included history of well-water consumption, residential history, lifestyle factors].
*†Chen et al. 20105[60] (Tables One, Two)	Taiwan	Incidence 1991-1994 (Average follow-up period of 11.6 years)	Urothelial carcinoma	Individual level ‘measured’	Arsenic water concentration (μg˙•L-1):	RR	8,086			age, sex, cigarette smoking status, education, alcohol consumption at enrolment, and whether subject started drinking well water from birth
			ICDO histology		< 10	Urothelial carcinoma		3	1.0
			N/A		10–49.9			6	1.85 (0.45–7.61)
			Urinary organs		50–99.9			3	2.19 (0.43–11.1)
			ICD9 188, 189, 189.1-189.9		100–299.9			7	5.50 (1.39–21.8)
					≥ 300			10	10.8 (2.90–40.3)
					unknown			7	4.34 (1.06–17.7)
					Cumulative arsenic exposure (μg˙•L-1•year):
					< 400	RR		6	1.0
					400– < 1,000	Urinary organs		3	1.16 (0.29–4.64)
					1,000– < 5,000			12	2.44 (0.91–6.50)
					5,000– < 10,000			5	3.88 (1.18–12.7)
					≥ 10,000			11	7.55 (2.79–20.4)
					Unknown			8	2.90 (1.01–8.37)
					[Arsenic concentration ranged < 0.15 to > 3,000 μg˙•L-1 and was estimated using 3,901 water samples from residence of participants at time of interview. Other measures of arsenic exposure included, duration of exposure, age starting/ending drinking well water, and cumulative exposure.]
*†Chung et al. 20136[65] (Table One)	3 villages in Putai Township, in BFD endemic area of southern Taiwan	Mortality 1996-2010 (Average follow-up period of 17.8 years)	ICD9 188	SMR based analyses:	Median arsenic content of artesian well (range: 700–930 μg˙•L-1) measured in the early 1960s.	SMRmale	1,563	24	2.9 (27.5–63.8)	SMR adjusted for age
						SMRfemale		19	59.4 (35.7–92.7)
				Group level
					[Used age-adjusted mortality rate in Taiwan as standard rates.]
				HR based analyses: Individual level ‘estimated’	Average arsenic concentration in artesian well (μg˙•L-1):	HR				HR adjusted for age, gender, education, smoking habits
					< 50			1	1.0
					50–710			15	4.35 (0.56–33.52)
					> 710			22	7.22 (0.95–55.04)
					[Duration of drinking artesian well water and history of residential address obtained from questionnaires. Authors found a significant association with duration of well water drinking.]

*Study included in meta-analyses.

†Recent study not included in the International Agency for Research on Cancer 2012 review (Monograph 100C [23]).

¹ICD = International Classification of Disease. ICD for cancer site abstracted which included bladder and urothelial/transitional cell carcinoma of the bladder or kidney. Transitional cell carcinoma of the renal pelvis often share the same etiology as bladder cancer, and as such, have been treated as bladder within the meta-analyses as recommended by IARC [23]. N/A = Not available.

²Cases = number of persons exposed between 1955-1959.

³95% Confidence intervals not available for data at low and high exposure.

⁴Results for transitional cell carcinoma were included in the meta-analysis.

⁵Results for urothelial carcinoma were included in the meta-analysis.

⁶Results from SMR were included in the meta-analyses.