ARSENIC HEALTH EFFECTS RESEARCH PROGRAM

PROJECTS

Berkeley

University of California

Table S1. Table of Epidemiologic Studies of Ingested Arsenic in Water or Urine and Diabetes Mellitus:

 

 

Study, Location, Design

 

Arsenic exposure levelsa

 

Resultsb

 

Notes

 

 

 

 

 

 

 

HIGH EXPOSURE STUDIES:

 

 

 

 

 

 

Rahman et al., 19981 Bangladesh

 

Unexposed (water)

 

PR=1.0 (ref)

 

Exposed subjects were those with hyperkeratosis due to arsenic.

Cross-sectional

 

0-500 g/L

 

PR=2.6 (1.2-5.7)

 

 

 

 

500-1000 g/L

 

PR=3.9 (1.8-8.2)

 

 

 

 

> 1000 g/L

 

PR=8.8 (2.7-28.4)

 

 

Nabi et al., 20052

 

Mean = 235 g/L (urine)

 

PR=2.8(ng)

 

PR for diabetes mellitus in 115 arsenicosis patients compared to 120 unexposed controls.

Bangladesh

 

 

 

 

 

 

Cross-sectional

 

 

 

 

 

 

Rahman et al., 19993

 

Unexposed (water)

 

PR=1.0 (ref)

 

PRs are for glycosuria in subjects with arsenic-caused skin lesions.

Bangladesh

 

< 500 g/L

 

PR=1.1 (0.5-2.0)

 

 

Cross-sectional

 

500-1000 g/L

 

PR=2.2 (1.3-3.8)

 

 

 

 

> 1000 g/L

 

PR=2.6 (1.5-4.6)

 

 

Tseng et al., 20004

 

< 17,000 g/L-years

 

RR=1.0 (ref)

 

446 initially non-diabetics followed for approximately 3 years. 41 developed diabetes.

Taiwan

 

≥ 17,000 g/L-years

 

RR=2.1 (1.1-4.2)

 

 

Prospective

 

(water)

 

 

 

 

Tsai et al., 19995

 

Median = 780 g/L

 

SMR=1.35 (1.16-1.55) in males and 1.55 (1.39-1.72) in females

 

SMRs for blackfoot disease endemic area. Ecologic study without individual data on exposure and confounding factors.

Taiwan

 

(water)

 

 

 

 

Ecologic

 

 

 

 

 

 

Wang et al., 20036

 

Non-endemic areas

 

3.5% (3.5-3.6%)

 

Prevalence rates of non-insulin dependent diabetes mellitus in arseniasis endemic & non-endemic areas.

Taiwan

 

Endemic areas

 

7.5% (7.4-7.7%)

 

 

Ecologic

 

 

 

 

 

 

Lai et al., 19947

 

0 g/L-years (water)

 

OR=1.0 (ref)

 

ORs for cumulative arsenic exposure.

Taiwan

 

100-15,000 g/L-years

 

OR=6.61 (0.86-51.0)

 

 

Retrospective cohort

 

≥ 15,100 g/L-years

 

OR=10.05 (1.30-77.9)

 

 

MEDIUM EXPOSURE STUDIES:

 

 

 

 

 

 

Coronado-Gonźalez et al., 20078

< 63.5 g/g (urine)

 

OR=1.0 (ref)

 

Dividing arsenic by creatinine could elevate ORs.

 

Mexico

63.5-104 g/g

 

OR=2.16 (1.23-3.79)

 

 

 

Case-control

> 104 g/g

 

OR=2.84 (1.64-4.92)

 

 

 

Lewis et al., 19999

Median = 14-166 g/L (water)

 

Males: 0.79 (0.48-1.22) Females: 1.23 (0.86-1.71)

 

No data on potential confounding variables.

 

Utah

 

 

 

 

 

 

Cohort mortality

 

 

 

 

 

 

Table S1 - Continued

 

 

 

 

 

 

LOW EXPOSURE STUDIES:

 

 

 

 

 

 

Ruiz-Navarro et al., 198810

3.44 2.36 g/L

Diabetics (n=38)

No difference in mean urine arsenic levels

 

 

 

Spain

3.68 2.27 g/L (urine)

Control subjects (n=49)

 

 

 

 

Cross-sectional

 

 

 

 

 

 

Zierold et al., 200411

< 2 g/L (water)

OR=1.0 (ref)

1185 people with 20+ years well water use

 

 

 

Wisconsin

2-10 g/L

OR=1.35 (0.78-2.33)

 

 

 

 

Retrospective cohort

>10 g/L

OR=1.02 (0.49-2.15)

 

 

 

 

Afridi et al., 200812

5.59 0.92 g/L

Diabetics (n=110)

Data given for non-smokers. Similar results seen in smokers. Little difference in mean urine arsenic levels between diabetics and non-diabetics

 

 

 

Pakistan

4.7 0.88 g/L

Non-diabetics (n=115)

 

 

 

 

Cross-sectional

(urine)

 

 

 

 

 

Meliker et al., 200713

Mean = 11 g/L (water)

Males: SMR =1.28 (1.18-1.37)

No individual data on exposure or confounders.

 

 

 

Michigan

 

Females: SMR = 1.27 (1.19-1.35)

 

 

 

 

Ecologic

 

 

 

 

 

 

Abbreviations: ng, not given; OR, odds ratios; PR, prevalence ratio; ref, reference category; RR, rate ratio; SMR, standardized mortality ratios; g/g, microgram of arsenic per gram of creatinine.

aExposure levels in the exposed subjects unless otherwise noted.

bResults are prevalence ratios, rate ratios, SMRs, or odds ratios for diabetes mellitus unless otherwise noted. 95% confidence intervals are in parentheses.

2470 Telegraph Avenue, Suite 301, Berkeley, CA 94704 Tel: (510) 990-8354 Email: asrg@berkeley.edu