Table 1.

Epidemiological evidence supporting arsenic promotion of diabetes.

Country	Study Population	Age	Adjustments	Duration	As Concentration (In ppb or ppm)	Diabetic Assessment/Methods of Detection	Ref.
Bangladesh	140 diabetic vs. 180 non-diabetic controls recruited with HbAlc level > 7%	≥20 years	Age, sex, family history of diabetes, smoking habit, betel nut chewing, education	2010	69.3–100.9 ppm in drinking water for 9.8–13.6 years	FBG ≥ 200mg/dL	[113]
	115 exposed subjects diagnosed as arsenicosis patients (>50 μg/L As water consumption and skin lesions) and 120 unexposed volunteers	14–85 years	Age, height and body weight	2001–2003	drinking water (0.218 ppm) and spot urine (20.235 ppm)	FBG ≥ 140 mg/dL	[114]
	163 subjects with keratosis exposed to arsenic and 854 unexposed individuals	>30 years	Age, sex and body mass index	NR	0.01–2.1 ppm in drinking water	history of symptoms: previously diagnosed diabetes, glycosuria and blood sugar level after glucose intake (OGTT)	[115]
	1595 subjects depending on drinking water from wells: 1841 drank arsenic-contaminated drinking water but 114 had not	≥30 years	Age, sex and body mass index	NR	well water > 0.05 ppm	Glycosuria	[116]
	40 workers occupationally exposed to arsenic, 26 without any known As exposure and 6 who directly handle As containing products	20–60 years	Sex, occupation, age, smoking habit	NR	22.3–294.5 nmol per mmol of creatinine in urine sample of the exposed group	glycosylated hemoglobin (HbA1c) 5.4% compared to reference group 4.4%	[117]
Chile	population based cancer case-control study of 1301 participants in Northern Chile	≥25 years	Age, sex, race, hypertension, cancer, socioeconomic status, smoking status	2007–2010	>0.8 ppm arsenic water concentration	physician diagnosed diabetes or oral hypoglycemic medication use	[118]
China	2090 women with singleton pregnancy from the Tongji Maternal and Child Health Cohort (TMCHC)	≥25 years	Pregnancy, education, income, ethnicity, fetal sex	2013	0.3 ppb	Urine samples and oral glucose tolerance test, FBG ≥ 92 mg/dL	[119]
	335 gestational diabetes mellitus and 343 controls without GDM based on a prospective cohort established in Beijing, China	<35–≥35 years	Age, ethnicity, education, occupation,	2017–2018	220 ppm	FBG ≥ 5.1 mmol/L, maternal hair samples	[120]
	1527 pregnant women drawn from Mother and Child Microbiome Cohort (MCMC) study	<30–≥30 years	Education, BMI	2017–2018	0.83 ppb	75-g oral glucose tolerance test (OGTT), FBG ≥ 5.1 mmol/L, 1 h postprandial ≥ 10.0 mmol/L, or 2 h postprandial glucose ≥ 8.5 mmol/L	[121]
	3474 women who were part of the Ma’anshan Birth Cohort (MABC) Study conducted from the City of Ma’anshan, Anhui Province of China	≤ 24 years, 25–29 years, ≥ 30 years	Maternal age, BMI, gravidity, parity, income, education	2013–2014	0.0047 ppb	FBG ≥ 5.1 mmol/L; 1 h, ≥10.0 mmol/L; or 2 h, ≥8.5 mmol/L	[122]
Croatia	202 adult urban participants from the city of Osijek in eastern Croatia and city of Zagreb in western Croatia	≥45 years	Age, gender, education, smoking, family history if diabetes, physical activity, dietary consumption, origin of water used for drinking	2018	0.5–361 ppb total urine As	FBG ≥ 3.5 mmol/L, HbA1c ≥ 37 mmol/L, insulin ≥ 15 pmol/L	[123]
India	Natives to Nallampatti, an agricultural village in south India and part of the KMCH-NNCD cross-sectional study	≥20 and ≤85 years	Age, sex, alcohol intake, smoking, tobacco use, BMI, education, occupation, familial diabetic history	2015	4.10–63.30 ppm creatinine units of arsenic	blood investigation included a random glucose, HbA1c, cystatin-c, non-fasting lipid profile, uric acid and hemoglobin	[124]
Italy	3390 art glass workers employed in 17 industrial facilities for at least 1 year	<40, 40–65 and >65 years	Age, sex, history of disease/mortality	1950–1985	3.26 ppb in glassworks (>10 μg/m3 in glassworks)	All causes of death coded according to the 8th revision of the ICD	[125]
	258 subjectswith a minimum of two-year residency in the regions and without occupational exposure to As	≥5 years	Age, sex, source of drinking water	1993–2008	3–215 ppb iAs in drinking water, 2.3–233.7 ng/mL tAs in Urine	FBG ≥ 126 mg/dL, OGTT ≥ 200 mg/dL, HbA1c levels > 7%, self-reported diagnosis, or medication	[103]
	200 diabetic cases and 200 controls	≥30 years	Age, height, weight, body mass index, smoking habit, family history of diabetes, employment, location	1960	intermediate total As concentration in urine (63.5–104 μg/g creatinine)	FBG ≥ 126 mg/100 mL (> or =7.0 mmol/l) or a history of diabetes treated with insulin or oral hypoglycemic agents	[126]
	1160 adults with a minimum 5 year residency in study area	≥18 years	Age, gender, ethnicity, education/occupation, smoking status, alcohol consumption, recent seafood intake, drinking water sources (well, treatment plant or other) and use and medical history	2008–2013	<0.01–419.8 ppb As in drinking water, tAs 0.52–491.5 ppb in urinary As.	FBG ≥ 126 mg/dL, 2HPG ≥ 200 mg/dL, self-reported diagnosis, or medication	[127]
	49 healthy individuals and 77 patients	NR	Age, sex, geographical location history of disease	NR	0.32–9.82 ppb As in diabetic patients, mean As 3.44 ppb	Urine samples of diabetic patients to test As concentration	[128]
	1451 randomly selected participants from Spain (representative sample of a general population)	≥20 years	Age, sex, somking status, education, seafood consumption	2001–2003	3.8 ppb of total plasma As, 106,000 ppb of total urine As, 14,900 ppb μg/g of iAs and 66,500 ppb of Asb in participants with diabetes	FBG ≥ 126 mg/dL and glycosylated hemoglobin (HbA1c) level > 6.5% or physician diagnosis or glucose lowering medication use	[129]
Sweden	43 smelter workers exposed to iAs dust for 13–45 years	44–70 years	age, height, smoking habit, alcohol consumption	1987	1.6–63 ppb As in work-room air at the smelter	self-reported type 2 diabetes	[130]
	12 cases with DM on death certificate and 31 controls employed in a Swedish copper smelter	30–74 years	Age, history of diseas/death	1960–1976	<0.5–>0.5 ppb As	death certificate, medical record	[131]
	5498 art glass workers in southeastern Sweden	≥45 years	Age, occupation (glassworkers vs. glassblowers, other foundry workers and unspecified glass workers)	1950–1982	<1.9 ppb As in Swedish glassworks; <6 μg/m3 As in Swedish glassworks	All causes of death coded according to the 8th revision of the ICD	[132]
Taiwan	891 adults in southern Taiwan village where arseniasis if hyperendemic	≥30 years	Age, sex, body mass index, activity level at work	1960–1970	0.1–15 ppm-year or higher	oral glucose tolerance test (OGTT) or self-reported history of diabetes treated with sulfonylurea or insulin	[104]
	Cancer and noncancer diseases	All age group	Sex, Age	1971–1994	0.25–1.14 ppm As in artesian well water	All causes of death coded according to the 8th or 9th revision of the ICD	[105]
	446 nondiabetic residents in a village in Taiwan	≥30 years	Age, body mass index and cumulative arsenic exposure	1988–1989	median As of artesian well water from 0.7 to 0.93 ppm	FBG ≥ 7.8 mmol/L and/or a 2 h post-load glucose level > or = 11.1 mmol/L.	[133]
	66,667 residents living in endemic areas and 639,667 in nonendemic areas	≥25 years	Age, sex	1999–2000	artesian well water > 0.35 ppm	All causes of death coded according to the 9th revision of the ICD (ICD-9 code 250 and A181)	[134]
	4 townships in southwestern Taiwan where blackfoot disease is endemic	NR	Age, Sex	1971–2000	arsenic concentration of artesian well water ranged from 0.35 to 1.14 ppm with a median of 0.78 ppm	All causes of death coded according to the 8th or 9th revision of the ICD (ICD-9 code 250).	[135]
	1297 subjects from an arsenicosis endemic area in southwestern Taiwan	≥40 years	Age, sex, smoking status, education, exercise, alcohol consumption, betel nut intake	1990, 2002–2003	0.7–0.93 ppm As in well water	FBG, cholesterol, triglycerides, low and high density lipoproteins, urine acid and urine creatinine levels, arsenic methylation patterns and GSTO1 genotypes linked to metabolic syndrome as an early factor for diabetes	[136]
UK	32 insulin treated (ITDM), 55 non-insulin treated (NITDM) diabetic patients and 30 nondiabetic individuals (C-DNM) from Oxford, England	18–78 years	Age, body mass index, glucose, insulin	NR	0.018–0.2 ppm As	Glucose levels and insulin treatment	[137]
USA	4549 American Indian participants	45–75 years	Age, sociodemographic, smoking and alcohol status, height, weight, blood pressure	1989–1991, 1998–1999	5.9–14 ppm iAs 14.3 ppb in Arizona, 11.9 ppb in Dakota, 7 ppb in Oklahoma	FBG ≥ 126 mg = dL, 2HPG ≥ 200 mg = dL, self-reported diagnosis, or medication	[138]
	1393 smelter workers	<20–40+	Age, sex, race, occupation	1946–1977	0.5–5 ppb As of air concentration in the insecticide building	All causes of death coded according to ICD	[139]
	8014 copper smelter workers in Montana	<20–≥30	Sex, Race	<1957,1938–1989	0.29–11.3 ppb of airborne As	All causes of death coded according to the 8th or 9th revision of the ICD (ICD-8 codes 460–519)	[140]
	1827 boys and 1305 girls	2–14 years	Age, sex	1907–1932	140–1600 ppm soil As concentration	All causes of death coded according to death records from the National Death Index, ≥47 and from Washington State (1900–1990), Oregon State (1971–1979), and California State (1960–1990), to locate deaths of cohort members	[141]
	Historical ward membership records of the Church of Jesus Christ of Latter-day Saints (LDS) (also known as the Mormons)	<50–80+	Age, sex	1977	mean As 150 ppb, median As 14 to 166 ppb	Death certificate, mortality from hypertensive heart disease	[142]
	1185 respondents from 19 townships in arsenic contaminated area	≥35 years	Age	1992–1993	2–>10 ppb As, with a median of 2 ppb As	Self reported	[143]
	788 adults aged 20 years or older who participated in the 2003–2004 National Health and Nutrition Examination Survey (NHANES) and had urine arsenic determinations	≥20 years	Age, sex, race, ethnicity; educational, smoking and alcohol consumption status; and dietary recall	2003–2004	7.1 ppb total As, 3 ppb dmAs, 0.9 ppb arsenobetaine	FBG ≥ 126 mg/dL, self-reported physical diagnosis or use of insulin/oral hypoglycemic medication	[144]
	3925 people on tribal tolls in 13 American Indian communities	<55-≥65	Age, sex, education, body mass index, smoking status, alcohol consumption	1989–1991	7.9–24.2 ppb urine As, median urine As 14.1 ppb	Glycated hemoglobin and insulin resistance, fasting glucose level of 126 mg/dL or higher, 2 h glucose levels of 200 mg/dL or higher, hemoglobin A1c (HbA1c) of 6.5% or higher, or diabetes treatment	[145]
	cohort of American Indians in Arizona, Oklahoma, North Dakota and South Dakota	≥30 years	Age, ancestry, family relationships	1998–1999, 2001–2003, 2005–2006, 2014–2015	median exposure of 5.93 ppb	FBG ≥ 126 mg/dL, or use of insulin or oral hypoglycemic medications	[109]
	non-institutionalized civilian resident population from NHANES	≥20 years	Body mass index, age, gender, race/ethnicity, education, income, cigarette use, alcohol intake and physical activity	2011–2014	246–260.6 ng/h	Spot urine samples, FBG ≥ 100 mg/dL or use of medication to treat hyperglycemia	[146]
	4549 members of 13 tribes based in Arizona, Oklahoma, North Dakota and South Dakota	45–75 years	Age, sex, study region, medical history, smoking status	1989–ongoing	10.2–11.2 nmol per mmol of creatinine in urine sample of the exposed group	Urinary arsenic species measured using HPLC to identify differentially methylated position	[110]
	2919 participants recruited by Strong Heart Family Study	≥25 years	Age, sex, education, smoking history, alcohol use, medical history	1998–1999, 2001–2003	median 0.52 ppb	Urine arsenic, FBG ≥ 126 mg/dL, self-reported physician diagnosis or self-reported use of insulin or oral diabetes treatment	[111]
	Pregnant women with and without GDM who received prenatal care at the University of Oklahoma Health Sciences Center (OUHSC) Women’s Clinic and High Risk Pregnancy Clinic	≥18 years	Maternal age, race/ethnicity, education, income, history of GDM diagnosis	2009–2010	1.25 ppb total arsenic	BG ≥ 135 mg/dL	[147]
	688 participants including type 1, type 2 and control participants from SEARCH, a study being conducted in South Carolina, Colorado and Columbia	10–22 years	Age, sex, race, education, height, weight	2003–2006	0.0429–0.0502 ppb iAs	Clinical diabetes assigned by the health provider	[148]
	5114 African-American and white men and women who are part of the CRADIA study living in Birmingham, AL; Chicago, IL; Minneapolis, MN; and Oakland, CA	≥25 years	Age, gender, race, education, smoking status, alcohol consumption, physical activity, BMI, dietary intake	1987–88; 2015–2016	<0.0593–≥0.1692 ppm toenail arsenic level	fasting glucose ≥ 126 mg/dL, non-fasting glucose ≥ 200 mg/dL, 2 h postchallenge glucose ≥ 200 mg/dL, hemoglobin A1c ≥ 6.5%, or use of glucose-lowering medications.	[112]