Table 1.

Summary of studies that examined associations between rice consumption and human exposure to arsenic.

Order	References	Study design	Study population	Rice intake assessment	Rice arsenic level	Water arsenic level	Biomarkers	Results
1	He and Zheng, 2010	Experimental	Female adults (n = 2).	Mass balance of arsenic metabolites. Wheat diet (5 days) vs rice-based diet (418 g of rice/day - 5 days)	101 μg/kg of i-As, 25 μg/kg DMA, and 7.5 μg/kg other arsenic species.	≤1 μg/kg	Mixture of two spot urine samples, the first in the morning and last in the evening.	Urinary t-As concentration doubled with rice diet in the two participants (from the mean value of 8.2 μg to 16.3 μg). DMA dominated urinary arsenic speciation. The mean percentage of rice-diet ingested arsenic excreted in urine was ~63%.
2	Meharg et al., 2014	Experimental	Male adults (n = 9).	Mass balance of arsenic metabolites. 6 participants that followed a rice diet (300 g/day - 5 days) vs 3 participants that did not consume rice (5 days).	99 μg/kg of i-As, 99 μg/kg of DMA, and 3 μg/kg MMA.	Distilled water	Spot urine, and 24-hour urine pass for participants consuming rice.	Urinary t-As was 7.3-fold higher after 4 days consuming rice (from 6.8 μg/L to 49.9 μg/L). Percentage of urinary DMA, and MMA increased 65%, and 6%, respectively after rice consumption. The percentage of urinary i-As decreased ~ 20% after rice consumption. The percentage of rice ingested arsenic excreted in urine was ~ 40%.
3	Lovreglio et al., 2012	Prospective cohort study	Adults Caucasian males resident in a coastal area of southern Italy (n = 12).	Daily food diary starting 3 days before the start of the study and continuing throughout the study period.	-	Refers to regulation (<10 μg/kg)	Morning spot urine samples for 10 consecutive days.	Rice intake was not shown to influence either urinary excretion of i-As and its methylated species. Rice played a minor role in participants’ diet.
4	deCastro et al., 2014	Cross-sectional study	Adults, adolescents, and children from NHANES (n = 8300).	24-Hour dietary recall including rice, rice cakes/crackers, and rice beverage consumption.	-	-	Spot urine	Children consumption of rice and adolescents consumption of rice cakes/crackers was related to an increase in urinary DMA (positive regression slopes of 115 and 872, respectively). An increase in urinary MMA was also associated with an increase of rice beverages in adults, and an increase of rice cakes/crackers in children and adolescents (positive regression slopes of 6.4, 65.5, and 103, respectively).
5	Davis et al., 2012	Cross-sectional study	Children from the NHANES (n = 2323).	48-Hour dietary questionnaire. Rice eater defined as 1/4 cup of cooked rice per day.	-	-	Spot urine	Children “rice eaters” had 1.6-fold higher t-As urinary compared to those non “rice eaters” (from 8.9 μg/L to 5.5 μg/L). An increase of 14.2% of urinary t-As was associated with each 1/4 cup of cooked rice consumption (14.1 g dry weight).
6	Kordas et al., 2016	Cross sectional study	Young children of 5–8 years of age (n = 328) residing in Montevideo, Uruguay.	Two 24-hour dietary recalls including rice consumption. Children defined as consuming <5 g rice/day vs children ≥5 g rice/day	-	0.45 μg/L	Spot urine	Sum of urinary arsenic concentrations (iAs, MMA, and DMA) increased from 11.0 μg/L to 13.5 μg/L with rice consumption ≥5 g rice/day.
7	Davis et al., 2014a	Cross-sectional study	Adults and children from NHANES (n = 20.497).	48-Hour recall dietary questionnaire including rice consumption.	-	-	Spot urine	Association between rice consumption and urinary t-As among both children and adults. An increase of 10 g (dry weight) rice per day was associated with an increase of 9.6% urinary t-As, and an increase of 8.6% in urinary DMA.
8	Wu et al., 2015	Cross-sectional study	U.S. adults (n = 6677) belonging to NHANES.	48-Hour recall dietary questionnaire including white and brown rice consumption.	-	-	Spot urine	Intake of white or brown rice were both associated with ~ 1.6-fold higher urinary sum of arsenic concentrations excluding arsenobetaine
								compared to those who did not consume rice (from 7.9 μg/L to 13.1 μg/L). Among rice eaters, urinary sum of arsenic metabolites excluding arsenobetaine did not differ between participants who primarily consume white rice or brown rice.
9	Gilbert-Diamond et al., 2011	Cross sectional study	Pregnant women (n = 229) residing in New Hampshire.	72-Hour recall dietary questionnaire including rice consumption.	-	Range from ≤0.07 μg/L to100 μg/L. Highly right-skewed.	Spot urine	Urinary t-As was 5.3 μg/L for rice consumers and 3.4 μg/L for non-consumers. Rice consumers had 0.07 μg/L higher i-As, 0.18 μg/L higher MMA, and 1.25 μg/L higher DMA than non-rice consumers. Rice consumption increased urinary loge t-As with a β of 0.009 (95% CI: 0.005, 0.013).
10	Davis et al., 2014b	Cross-sectional study	Mother-infant pairs (n = 170) residing in New Hampshire, and mother-infant pairs (n = 130) belonging to a validation cohort from Rhode Island.	72-Hour recall dietary questionnaire including rice consumption.	-	-	Toenails	Toenail t-As concentration from infants exposed to rice consumption during pregnancy was 0.09 μg/g compared to 0.06 μg/g for those non-exposed. An increase of 1/4 cooked rice cups/day during pregnancy was associated with a 16.9% increase in infants toenail t-As concentration.
11	Cascio et al., 2011	Prospective cohort study	Adults living in United Kingdom (U.K.) (n = 49).	Food frequency questionnaire. Bangladeshi with high consumption of rice (n = 37) vs white Caucasians with lower consumption of rice (n = 12)	-	Refers to regulation (<10 μg/kg)	Spot urine	U.K. Bangladeshi community had 30-fold higher rice consumption than U.K. white Caucasians. Similar urinary t-As concentrations were found for the Bangladeshi group (28.4 μg/L) and the white Caucasians group (20.6 μg/L). Urinary DMA and i-As concentrations for the U.K. Bangladeshi community was 5- and 2.5-fold higher than for the U.K. white Caucasians.
12	Wei et al., 2014	Cross-sectional study	U.S. adults (n = 3027) belonging to the NHANES.	Food frequency questionnaire including rice consumption. Adults classified as low and high rice consumers (<twice a week vs ≥twice a week).	-	-	Spot urine	Participants that consumed rice more than twice a week had 1.1- and 1.2-fold higher urinary t-As (from 2.21 to 2.42 μg/g creatinine) and DMA (from 1.32 to 1.58 μg/g creatinine) compared to those that consume rice less than twice a week, respectively.
13	Melkonian et al., 2013	Prospective cohort study	Adults living in Araihazar, Bangladesh (n = 18.470, with a subset of 4517 with urinary As metabolites).	Food frequency questionnaire including rice consumption.	-	50% participants with access to well water <50 μg As/L.	Spot urine	Rice consumption was positively associated with urinary t-As (Multivariate model β = 0.041) (95% CI: 0.032, 0.051).
14	Park and Lee, 2013	Cross sectional study	Korean adults (n = 3404) residing in South Korea belonging to the Korean NHANES.	Semiquantitative food frequency questionnaire including rice consumption.	-	-	Spot urine	Rice consumption was associated with a higher urinary t-As concentration. The adjusted means urinary t-As for low, moderate, and high rice consumption were 99.3, 105.4, and 120.1 μg/g creatinine.
15	Rivera-Nuñez et al., 2012	Cross-sectional study	Adults (n = 343) residing in Michigan belonging to the case-control study of bladder cancer in Michigan.	Food frequency questionnaire including rice consumption.	-	≤50 μg/L	Spot urine	Rice consumption was positively associated with sum of arsenic concentrations (i-As, MMA, and DMA) with β and R2 of 7.51 and 0.008, respectively.
16	Karagas et al., 2016	Cross sectional study	Infants (n = 759) belonging to the NHBCS.	Questionnaire and diaries including rice and rice-based products.	Ranging from 4.6 to 201 μg i-As/kg.	Mainly <10 μg/kg	Spot urine	Urinary t-As was nearly double (from 4.97 μg/L to 9.53 μg/L) for infants consuming rice-based products compared to those that did not consume rice. Infant rice cereals and rice-based snacks were positively associated with urinary log10 t-As according to the estimate, β = 0.22 (95% CI: 0.11, 0.34) and 0.10 (95% CI: 0.04, 0.15), respectively.
17	Cleland et al., 2009	Descriptive study	Korean women adults (n = 108) living in Washington State.	Food frequency questionnaire including rice consumption rates, which is an important food item in the Korean community.	-	<2 μg/L	Spot urine and hair	Estimated arsenic exposure from rice was a statistically significant predictor of sum of the urinary arsenic species, but only explained a small percentage of the variability.
18	Cottingham et al., 2013	Cross-sectional study	Adults residing in New Hampshire residents (n = 852) belonging to the case-control study of bladder and skin cancer in New Hampshire.	Annual food frequency questionnaire recording average daily consumption of 120 different diet items.	-	<1 μg/L vs ≥ 1 μg/L	Toenail	Toenail t-As increased with household water arsenic concentration. No clear relationship was found between toenail t-As and rice consumption, which may be related to the low rice consumption of the study population (adjusted model β for brown rice = 0.23, p-value = 0.29; β for white rice = 0.065, p-value = 0.67).
19	Slotnick et al., 2007	Cross-sectional study	Adults control participants (n = 440) belonging to the case-control of As in drinking water and bladder cancer in 11 counties o Michigan.	Food frequency questionnaire including rice consumption.	-	Range from <LOD to 99.9 μg/L with a median of 0.38 μg/L.	Toenail	Rice consumption was negatively associated with As toenail concentration (β = −0.137, R2 = 0.017).
20	MacIntosh et al., 1997	Cross sectional study nested within prospective cohort	Females and males adults (n = 969) belonging to the 1984–1986 Nurses’ Health Study and the Health Professionals Follow-Up Study.	Semiquantitative food frequency questionnaire.	-	-	Toenail	Regression coefficient for brown rice consumption per 1 unit increase in daily average servings 0.494 (p-value = 0.0457)