Table 2

Pooled Estimated Relative Risk and 95% CI—All Studiesa

Exposure AreaStudies IncludedPooled Estimated Relative Risk (95 % CI)
CVDCHDStrokePADb
High Exposure Areas
Overallc[23,24,27,28••,29•,30,3338,40,5357]1.32 (1.05–1.67)1.89 (1.33–2.69)1.08 (0.98–1.19)d2.17 (1.47–3.20)
Internal comparisons[27, 28••,29•,30,33,56]1.46 (1.19–1.80)2.62 (1.09–6.32)1.03 (0.55–1.94)--e
External comparisons[34,37,53]1.01 (0.88–1.17)1.39 (1.07–1.80)1.07 (0.95–1.20)1.69 (0.62–4.63)
Urine arsenic[24,28••,35,40,54]1.55 (1.01–2.37)3.25 (0.97–10.93)1.03 (0.53–2.02)1.86 (1.46–2.37)
Water arsenic[23,25,27,28••,29•,30,33,3638,5557]1.30 (1.04–1.62)2.02 (1.33–3.07)1.16 (0.98–1.37)3.21 (2.27–4.53)
Low to Moderate Exposure Areas
Overall[31•,32•,41,43,44,5962]1.06 (0.99–1.14)1.06 (0.89–1.26)1.07 (0.96–1.20)1.13 (0.77–1.66)

CHD: Coronary heart disease; CI: Confidence interval; CVD: Cardiovascular disease; PAD: Peripheral arterial disease. RR: relative risk.

aExcluding Rapant et al. 2009 [42] and Varsanyi et al. 1991 [49] because they did not report enough data to calculate confidence intervals.
bIncludes peripheral arterial disease and diseases of the arteries, arterioles, and capillaries.
cIf a study included results for multiple media, either urine and water arsenic (Chen et al. 2011 [28••] and Tseng et al. 1996 [25]/Tseng et al. 2005 [24]) or urine and hair arsenic (Afridi et al. 2010 [39]; 2011[40]), only the urine result was used in the overall pooled RR.
dAfter removing an influential study, the estimated pooled relative risk for stroke was 1.12 (95 % CI: 1.04, 1.22).
eNo studies examined for this exposure level, study design, and outcome.
-