Arsenic in Soil: A Critical and Scoping Review of Exposure Pathways and Health Impacts

Abstract

Arsenic (As) in soil, such as mining waste, is a concern for communities with legacy contamination. While the chronic health effects of As exposure through drinking water are well documented, the association between As in soil and population-wide health impacts is complex, involving factors like soil accessibility, soil properties, and exposure modes. This review summarizes evidence of associations between As in soil and human health, as well as biomarker and bioaccessibility evidence of exposure pathways. Fourteen studies were included in the final analysis. Reviewed studies reported associations between As in soil and birth outcomes, neurological effects, DNA damage, and cancer. Some of these health outcomes are not known to be linked to As in drinking water and were reported over a range of soil concentrations, indicating inconsistencies. Higher soil As concentrations are associated with higher As in human biospecimens, suggesting direct and indirect soil ingestion as primary exposure pathways. The subpopulations more likely to be exposed include younger children and those involved in soil-based activities. Future research should focus on standardized epidemiological studies, longitudinal studies, soil exposure and mitigating factors, combined exposure biomarker studies, the behavior of the different As species, soil dose related to bioavailability/bioaccessibility, and effects with other elements.

1. Introduction

Arsenic (As) is a naturally occurring element found in the Earth’s crust, and it is commonly present in air, water, and soil. Average concentrations in the Earth’s crust range from 1.5 to 5.7 mg kg⁻¹ [1,2,3,4,5,6]. In soils and sediments, As levels vary worldwide, with average concentrations ranging from 0.1 to 40 mg kg⁻¹ [7,8,9].

Human activities have significantly altered many landscapes, contributing to increased background levels of As in soil. This has occurred through the use of products such as copper–chrome–arsenate-treated timber and garden pesticides and production processes such as farming, railways, smelting, gold mining, and others [10,11,12,13,14,15]. Many urban communities now live in these altered environments, where legacy contamination poses a concern. Although the health impacts of arsenic in drinking water are well established, the effects of arsenic in soil are less clear. They depend on factors such as how easily people come into contact with the soil, the soil’s physical and chemical properties, and the ways people might be exposed.

To manage these risks, regulatory authorities around the world have established guidelines to assess As contamination in soil. Jennings identified 387 soil values globally, with a median value of 20 mg kg⁻¹ [16]. They found non-random clusters of As guideline values at 100, 50, 30, 20, and 0.39 mg kg⁻¹ across different countries and regions. Most of these guidelines are set above background concentrations of As in soil [16].

Arsenic exists in both inorganic and organic forms. Inorganic As (iAs) is commonly found in minerals, rocks, and mine tailings, commonly as arsenate (AsV) and arsenite (AsIII). The predominant organic As species, monomethylarsonic acid (MMA) and dimethylarsinic acid (DMA), are less toxic than iAs and are often found in fish and shellfish. The order of toxicity of As species is AsIII > AsV > MMA > DMA. However, all As compounds have been classified as human carcinogens by the International Agency for Research on Carcinogens (IARC) [17]. As is known to cause skin, bladder, and lung cancers, and possibly liver cancer [18,19]. Arsenic’s ability to damage DNA (genotoxicity) has been demonstrated in both laboratory and animal studies, and several mechanisms have been proposed [18]. However, regulatory approaches differ. Some jurisdictions treat As as a “threshold” chemical (with a safe exposure level), while others classify it as “non-threshold” (assuming any exposure poses a risk) [20,21]. The IARC based its classification on a wide range of evidence, including human epidemiological studies, animal research, and lab-based studies of chemical characteristics, metabolism, and modes of carcinogenicity [17].

Research shows that As, particularly As(III), exhibits toxicity due to its high binding affinity for sulfhydryl groups (also known as thiols) in key enzymes. This binding can interfere with enzyme function by displacing critical metal co-factors, such as selenium, or reducing the reactivity of groups. As a result, As disrupts important biological processes, including DNA repair and mitochondrial respiration [22]. As(V) is thought to impact DNA synthesis [23]. The genotoxic mechanisms of iAs are still being actively studied, with various properties and biomolecular interactions now characterized [24].

Humans can be exposed to As through many pathways. These include drinking contaminated water, eating contaminated food, using contaminated water in food preparation, occupational exposure, smoking tobacco, ingesting or inhaling contaminated soil and dust, and skin contact with contaminated surfaces [20,22]. Globally, As is the most common contaminant in drinking water [22]. Some foods, such as rice and other grains, salt, and seaweed, naturally contain higher levels of iAs and are regulated in Australia and New Zealand due to known risks [25,26].

Exposure to As in soil, including from mining waste, occurs through several pathways: direct ingestion (e.g., incidental swallowing and soil pica), skin contact, and indirectly through eating food grown in contaminated soil or food with soil particles attached. Exposure can also occur through products from animals that feed on contaminated soil and vegetation. As-contaminated soil can be transported indoors on shoes, clothing, pets, or through air-conditioning systems, contributing to indoor exposure, especially through everyday hand-to-mouth behaviors [27,28]. Citizen science projects have further demonstrated links between outdoor contamination and indoor exposure pathways [29].

According to the Australian National Environment Protection (Assessment of Site Contamination) Measure, the ingestion of soil and dust, particularly from commercial, residential, and recreational environments, accounts for 72% to 90% of human exposure to As from soil [21]. Homegrown produce contributes 9%, inhalation less than 1%, and dermal absorption between 5% and 28% [21]. Practically, this means that in areas with high soil As concentrations, the ingestion of soil and dust is the primary source of exposure.

Soil is a more complex medium than water, and the bioavailability of As can vary considerably depending on the soil type and the specific forms of arsenic present [30]. While the inherent toxicity of As itself does not change much between soil and water, exposure pathways, frequency, dose, and potential health effects may differ when exposure comes from the soil. This is largely because environmental factors and physicochemical characteristics of soil influence how much arsenic is available for absorption [31]. Factors like soil pH, redox, organic matter, clay content, and the presence of carbonates influence the availability and distribution of As in soil [32,33].

Adding to the complexity, human-made soils, such as from mine tailings, can significantly increase exposure concentrations. These processed materials, including gray sands and calcined sands, often contain much higher concentrations of As than natural soils. Gray sands can have As levels as high as 3160 mg kg⁻¹, while calcined sands can reach up to 47,100 mg kg⁻¹ [33]. Despite these high levels, there is a lack of understanding of the interaction between soil characteristics, how people are exposed to arsenic, and how these affect health.

Blood, hair, nails, and urine are commonly used biological samples to assess As exposure. Blood and urine are useful for detecting recent exposure since As and its metabolites decline relatively rapidly after exposure ceases, typically over a period of days. However, blood concentrations are usually much lower than in other biological samples, making blood tests more suitable for identifying high-level or continuous exposures. Urine, on the other hand, is more effective for assessing short-term and ongoing exposures, including metabolites. Urine, on the other hand, is more effective for evaluating short-term and ongoing exposure, and is considered the most reliable biomarker for recent As exposure [23]. A 24 h urine sample is the gold standard, as As clears from the blood within hours, making blood a poor long-term indicator [23].

Hair and nail samples are often used to assess long-term exposure. As they grow, As is deposited, creating a timeline of exposure. Hair, due to its faster growth rate, can provide a more detailed exposure history than nails. However, both hair and nails can be affected by external contamination, though nails are less prone to this [34]. The accurate assessment of arsenic exposure, particularly from contaminated soil, depends on selecting the most practical and reliable methods for each situation.

Despite regulatory thresholds for water, and a growing understanding of arsenic’s toxicological effects, knowledge gaps remain in relation to arsenic in soil—particularly regarding health outcomes, exposure levels, and how soil characteristics influence bioavailability. There is a need for a clearer understanding of soil-specific exposure pathways and the associated health impacts.

This scoping review assesses the current evidence related to soil–As exposure, which remains an under-researched pathway compared to water and food. It seeks to address the following key questions:

• What health impacts have been observed in relation to As in soil?
• What concentrations of As in soil are associated with adverse health effects?
• What other soil or exposure parameters are important to understanding As’s impact on human health?
• How robust are the existing studies on the health impacts of As in soil?
• What methods are used to assess As exposure, and how reliable are they?

This review summarizes, synthesizes, and evaluates the following: (1) epidemiological evidence of associations between As in soil and human health and (2) biomarker and bioaccessibility evidence of human exposure pathways. Bioaccessibility, in this context, refers to the fraction of compounds that become accessible for absorption [35]. By focusing on soil-specific pathways, this review advances the understanding of how arsenic in soil contributes to population-level exposure and identifies critical areas for future research.

2. Method—Literature Search and Selection Strategy

The protocol for this scoping review was drafted following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols for Scoping Reviews) and registered retrospectively with the Open Science Framework (https://osf.io/dsn5z).

Databases PUBMED, SCOPUS, Web of Science, and Embase were searched using a combination of keywords, drafted by reviewer SG (see Supplementary Materials). The search aimed to identify studies that assessed markers of exposure and adverse health effects due to chronic exposure to As from soil. The search terms were refined and peer reviewed by reviewers CI, MD, MM, PN, SM, and TC. Search limits were set to include all peer-reviewed English language journal articles reporting data from human participants and published prior to 11 November 2022. References were uploaded into the bibliographical management software EndNote^® v20, and duplicates were deleted. References were then exported into Rayyan^®, a web and mobile app for organizing and managing reviews.

Publications were screened in three phases. In the first phase, 2863 titles were identified, and the abstracts were screened by a single reviewer (SG). Studies that clearly did not reflect the review questions were rejected. In the second phase, two reviewers (SG and SDM) independently reviewed the remaining 156 titles and abstracts to identify studies that met the following criteria: Studies were included if they demonstrated human exposure pathways to As-containing soil and/or associated health impacts. Studies were excluded if they lacked original research (i.e., reviews, opinions, book chapters, letters, conference abstracts), were laboratory-based or animal studies (i.e., in vitro or in vivo studies), were about acute poisoning from As ingestion, only involved probability modeling of adverse health effects, reported As intake from secondary exposure pathways related to food ingestion, reported As exposures from smoking or drinking water, reported occupational As exposure, or were case studies/case reports involving fewer than 10 people. Reviewers CI and SM were consulted in the final screening of the studies.

In the third phase, 14 full texts were retrieved from the total of 2863 publications screened (Figure 1). These 14 studies were included in the final analysis.

We synthesized results using a narrative review framework, which included developing a preliminary synthesis of findings, such as environmental As levels, As exposure, and potential adverse health effects of As exposure, and exploring the relationships within and between studies (i.e., patterns and trends) where possible.

For the 9 studies that provided information on As exposure in soil, we extracted data on locations, study type, population number, environmental As concentrations, biological As exposure concentrations, adverse health effects, and adjustment factors. For biological As in hair, nails, and urine, we extracted data on the location, sample type, sample size, As concentration in the environment, and As concentration in the biological samples.

The robustness of the studies was assessed qualitatively by examining the consistency of health outcomes and methodological limitations.

3. Results

Figure 1 presents the screening process and results of the literature search. We identified nine studies that investigated associations between As in soil and/or As in blood, hair, nails, or urine and effects on health

Table 1. Observed health effects from exposure to As in soil (n = 9 studies).

Health Effects	Reference	Location	Study Type	Sample Size/Population	Environmental As Conc a	Biological As Exposure Conc b	Adverse Health Effect	Adjustment Factors
Birth Outcomes	[37]	South Carolina, USA	Retrospective cohort study. General metals in soil.	n = 9920 mother-child pairs c	Soil at maternal address: 2.56 mg kg−1 (IQR); soil at LBW mother’s address: 3.47 mg kg−1; soil at normal birth weight mothers address: 3.25 d	Not measured	The As variable was nonlinear in relation to LBW, and the association between higher concentrations of As with LBW was strong (p = 0.002)	Characteristics: Maternal: age, race, ethnicity, parity, tobacco use, and alcohol use. Child: sex, birthweight, and weeks of gestation. Neighborhood: density per square mile and median age of residential building.
Neurological Effects	[36]	Georgia, USA	Retrospective cohort study. Windblown dust and impacted soil from pesticide plant.	n = 85 adults exposed	Soil: range: 2–1845 mg kg−1; mean: 46.6 mg kg−1; standard deviation: 164.1 mg kg−1	Not measured	Strong association between As-containing dust exposure and peripheral neuropathy in adults (OR 5.1; p = 0.004)	None mentioned
DNA Damage	[38]	Villa de la Paz, Mexico	Cohort study. Gold and other metal/mineral mining sites.	Exposure Villa de la Paz: n = 20 children e	Exposure Villa de la Paz: Soil: range: 141–11,930 mg kg−1; mean: 2462 mg kg−1; n = 26 dust: range: 352–9950 mg kg−1; mean: 2231 mg kg−1; n = 18	Exposure Villa de la Paz: FVU: range: 87–323 µg g−1 geometric mean: 136; creatinine; n = 20	100% of the children had AsU > health guidelines f Comet study: tail length and tail moment in cells were significantly higher than the cells from the control site in Matehuala (p < 0.05) g	None mentioned
				Control Matehuala: n = 35 children e	Control Matehuala: Soil: range: 51–6866 mg kg−1; mean: 1019 mg kg−1; n = 23	Control Matehuala: FVU: geometric mean: 34; range: 8–60 µg g−1 creatinine; n = 35
Cancer	[43]	Former lumber treatment facility; Kentucky, USA	Cross-sectional study.	n = 84 adults and children	Soil: range: 10.3–100.6. median: 64.8 mg kg−1 h	Toenail: median: 0.43 mg kg−1 (IQR: 0.12–0.9) i	Cancer incidence: no difference found for lung and bladder cancer j	Gender, age, race/ethnicity, occupational history, tobacco use, soil and gardening exposures, and well water.
	[39]	Victoria, Australia	Ecological study.	People living in high soil As areas and cancer data from Victorian Cancer Registry data as per ICD-9 codes	Soil > 100 mg kg−1 median soil range: 1–16,800 mg kg−1	Not measured	Cancer relative risk rate (95%CI): All cancers 1.06 (1.03–1.09) Prostate cancer 1.14 (1.05–1.23) Kidney cancer 1.16 (0.98–1.37) Melanoma 1.36 (1.24–1.48) Chronic myeloid leukemia 1.54 (1.13–2.10) Breast cancer in females 1.10 (1.03–1.18)	Gender, geographical location and As soil and water concentrations.
	[41]	Gold mining region; Victoria, Australia	Ecological study.	People living in the goldfields region and peripheral regions (for comparison). Victoria Cancer Registry data as per ICD-10 code	Soil range: 1.4–1857 mg kg−1	Not measured	Cancer relative risk rate (95%CI): Males: all cancers 1.21 (1.15–1.27) Males: melanoma 1.52 (1.25–1.85) Females: all cancers 1.08 (1.03–1.14) Females: melanoma: 1.29 (1.08–1.55)	Gender and socioeconomic disadvantage.
	[42]	USA	Ecological study.	Individual patient cases from the Surveillance Epidemiology and End-Results (SEER) database	Soil and sediment range: <1.476 mg kg−1 to >14,525 mg kg−1	Not measured	Cancer incidence rate: 1 mg kg−1 increase in soil As conc was associated with 1.04% increase in lung cancer rate (p < 0.0001)	Age, income, tobacco use, As exposure and residential and geographical location.
	[44]	Suzhou, China	Ecological study.	All resident death records from 2005 to 2010. Mortality classified using ICD-10; expressed as per 100,000 people/year	Soil range: 6.14–16.52 mg kg−1; Mean: 9.08 mg kg−1;	Not measured	Age-adjusted mortality rates: Colon cancer 1.083 (1.027, 1.142) Gastric cancer 1.111 (1.061, 1.165) Kidney, cancer 1.129 (1.039, 1.228) Lung cancer 1.050 (1.001, 1.102) Nasopharyngeal cancer 1.086 (1.028, 1.148)	Age, “heavy metals”, tobacco use, and education.
	[40]	Spain (including the Canary and Balearic Islands)	Ecological study.	Mortality data from National Statistics Institute for 27 tumors as per ICD-9 and ICD-10 codes	Soil range: 0.10 to 2510 mg kg−1	Not measured	Mortality: Cancers of the stomach, pancreas, lung, and brain, and non-Hodgkin lymphomas statistically significant (RR > 1) in males and females	Geographical location, sociodemographic indicators such as urban/rural zoning, occupation, number of people in household, literacy levels, unemployment, and income.

Notes: IQR = interquartile range. Reported parameters and number of significant figures may differ as we present data as reported in each study. ^a environmental concentrations presented in the paper. All As concentration values adjusted to mg kg⁻¹. Similar statistical values not provided in all papers. ^b FVU concentration values are in ug/g creatinine. Toenail concentrations reported in mg kg⁻¹. ^c analysis excluded multiple births and missing data. Pregnant women insured by South Carolina Medicaid between 1996 to 2002. ^d all values estimated. Soil sampled across 1.0–3.0 km grids, and As concentrations estimated for maternal addresses using Bayesian kriging. ^e Villa de la Paz: mean age, 4.9 years, who resided in the area/house for at least 2 years. Control: mean age, 4.4 years, who resided in the area/house for at least 2 years. ^f significantly higher when compared to control site at Matehuala (p < 0.05). ^g tail length higher than 6.0–60% in exposure group vs. 14% in control group. ^h total sample number not specified. A total of 4–24 samples per residence. ⁱ 63.7% exhibited conc > 0.2 µg g⁻¹ and 19.7% had conc > 1 µg g⁻¹; 10.7% of data not analyzed. ^j the ZIP code-level SIRs for bladder and lung cancer were both numerically elevated (SIR < 1.2), but the analysis did not have sufficient power to detect such small differences in rates.

(observed health effects from exposure to As in soil) and

Table 2. Summary of biological arsenic determinations in hair, nails, and urine (without assessments of adverse health effects).

Reference	Location	Study Type	Sample Size	Environmental As Conc a	Biological As Exposure Conc b
[45]	Gold mining region Victoria, Australia	Cross-sectional study	n = 153 adults and children	Soil: range: 9.1–9900 mg kg−1; geometric mean: 123.1 mg kg−1; median: 92.0 mg kg−1. Control soil: range: 1.7–80 mg kg−1; geometric mean: 4.3 mg kg−1; median: 3.3 mg kg−1. Dust: range: 12–1300 mg kg−1; geometric mean: 60.8 mg kg−1; median: 53.0 mg kg−1	Toenail: range: 3.20–477 mg kg−1; geometric mean: 32.1 mg kg−1 (no median provided). Control (low personal): range: 1.30–7.70 mg kg−1; geometric mean: 3.35 mg kg−1 (no median provided). Hair: range: 0.4–27.3 mg kg−1; geometric mean: 3.31 mg kg−1 (no median provided). Control (low personal) hair: 0.20–4.80 mg kg−1; geometric mean: 1.27 mg kg−1 (no median provided).
[46]	Gold mining region Victoria, Australia	Cross-sectional study	Exposure: n = 55 adults Control: n = 52 adults	Soil: range: 9.1–9900 mg kg−1; geometric mean: 123.1 mg kg−1; median: 92.0 mg kg−1. Control soil: range: 1.7–80 mg kg−1 Dust: range: 12–1300 mg kg−1; Geometric mean: 60.8 mg kg−1; median: 53.0 mg kg−1. Control dust: range: 2.2–21 mg kg−1; geometric mean: 3.9 mg kg−1; median: 3.9 mg kg−1.	FVU: exposure: <DL-28.4 ug/L; geometric mean: 1.64 μg/L; (no median provided). Control: <DL-2.81 μg/L; geometric mean: 1.18 μg/L; (no median provided).
[47]	Gold mining region Victoria, Australia	Cross-sectional study	n = 29 children 5–13 yrs	Soil: range: 3.3–130 mg kg−1; geometric mean: 11.5 mg kg−1	Toenail: range: 0.15–2.1 µg g−1; geometric mean: 0.49 mg kg−1.
[48]	Gold mining region Victoria, Australia	Cross-sectional study	n = 24 children 5–17 years	Soil: range: 3.4–97.1 mg kg−1; geometric mean: 9.53 mg kg−1; median: 5.93	Toenail: range: 0.03–0.712 a mg kg−1; geometric mean, 0.173 a mg kg−1; median: 0.166.
[49]	Hubei Province, China	Cross-sectional study	n = 120 children 3–17 years	Soil: range: 3.1–17.2 mg kg−1; Geometric mean: 8–10.7 mg kg−1 b	Ave first morning urine: 15.5 µg/L; mean 24 h urine: 11.4 µg/L (R = 0.143, p = 0.166); Creatinine-adjusted urine: FVU: mean: 22 µg/L; 24 h: mean: 18.1 µg/L (R = 0.450, p < 0.01).
[43]	Lumber treatment facility c	Cross-sectional study	n = 84 adults and children	Soil: range: 10.3–100.6. mg kg−1 median: 64.8 mg kg−1 d	Toenail: median: 0.43 mg kg−1 (IQR: 0.12–0.9) e.

Notes: DL = Detection limit. Reported parameters and number of significant figures may differ as we present data as reported in each study. ^a values were inconsistently provided in the journal article. These values are taken from “Table 1 Summary statistics for arsenic concentration in residential soil and toenail clippings recorded in 2011”. ^b geometric mean for age ranges for three age groups broken down by sex, representing six ranges. ^c Yaffee et al., 2019 included a cancer assessment [43]. ^d total sample number not specified. A total of 4–24 samples per residence. ^e 63.7% had conc > 0.2 µg g⁻¹ and 19.7% had conc > 1 µg g⁻¹; 10.7% of data not analyzed. An increase of 1 mg kg⁻¹ of As conc was associated with a 0.003 µg g⁻¹ (95% CI: 0.002–0.004, p < 0.001) increase in toenail concentration. Those that engaged in digging were 0.68 µg g⁻¹ (95% CI: 0.06–1.3, p = 0.03) higher compared to those that did not.

(summary of biological arsenic determinations in hair, nails, and urine) present the details of these studies, their observed health effects, and adjustment factors. Three cohort studies investigated associations between As from soil and corresponding non-cancer health effects [36,37,38], and five ecological studies and one cross-sectional study investigated the association between increased As concentration in soil and cancer incidence rates [39,40,41,42,43,44]. Six studies investigated exposure to soil and the intake of As via outdoor activities [43,45,46,47,48,49]. Summaries of the two types of studies’ health effects from As in soil, and biomarker and bioaccessibility from exposure to As in soil, and their limitations are set out in Section 3.1 and Section 3.2 below.

3.1. Observed Health Effects from Association with As in Soil

3.1.1. Birth Outcomes in Areas with <4 mg kg⁻¹ As in Soil

A large cohort study conducted in South Carolina, USA, by McDermott et al., investigated the association between residential soil As concentrations and adverse pregnancy and birth outcomes [37]. This study included 9920 mother–infant pairs from a socioeconomically disadvantaged area, focusing on the risk of low birth weight, defined as 2500 g or less. The researchers used Bayesian kriging to estimate soil As (interquartile range of 2.46 mg kg⁻¹) and other metal concentrations at the maternal address during pregnancy. Kriging is a geostatistical interpolation method to estimate values at unobserved locations. These estimates were then linked to birth weight data from medical birth records.

The study found a statistically significant association between higher soil As and lower birth weight (p = 0.002). The estimated soil As concentration at the addresses of mothers of low birth weight infants was 3.47 mg kg⁻¹, compared to 3.25 mg kg⁻¹ (crude odds ratio with only one predictor in the model: 1.04 (low: 1.02, normal: 1.07)) for mothers of non-low birth weight infants [37].

Despite these findings, the study had several limitations. Soil As concentrations were not directly measured but estimated statistically, and no biomonitoring data were collected. The study population was limited to low-income families, and maternal address data were from the sixth month of pregnancy, potentially missing other exposures. Additionally, although some environmental factors were considered, drinking water sources were not included. The authors noted that the biological mechanisms by which metals might affect birth weights remain unclear [37].

3.1.2. Neurological Effects Reported in People Living near a Pesticide Plant

Gerr et al. conducted a cross-sectional study in Georgia, USA, to estimate historical concentrations of As exposure from a former pesticide plant [36]. As the area had been remediated in 1995, biological markers of exposure were not available, so the researchers used historical measures to estimate past As concentrations. These included house dust, windowsill dust, attic dust, and soil. Neurological assessments were conducted on 85 exposed and 118 non-exposed (control) participants, evaluating both peripheral and central nervous system function.

The study found a strong association between elevated As concentrations, consistent with very contaminated environments (defined as soil As > 100 mg kg⁻¹) and peripheral neuropathy (OR 5.15; p = 0.004). However, several limitations were noted, including the absence of data on other potential exposure sources (e.g., drinking water and home-grown vegetables), potential testing bias (though considered low), and selective survival bias, as those most severely affected may have been less likely to participate.

While the findings support an association between chronic exposure to soil As and neurodevelopmental outcomes, the study’s limitations restrict the strength of the conclusions [36].

3.1.3. DNA Damage in Children Exposed to Soil As from Mine Tailings

A population health study by Yáñez et al. investigated the association between soil arsenic (As) exposure and DNA damage in children living in Villa de la Paz, Mexico [38]. This prospective cohort study included healthy children aged 3–6 years from a high-exposure site (n = 20) and a control site in Matehuala (n = 35). Soil samples were collected from areas frequently used by children, such as schools and homes. The average soil As concentration at the exposure site was 2462 mg/kg, compared to 1019 mg kg⁻¹ at the control site. Household dust samples from randomly selected homes in Villa de la Paz, where mine tailings may have been used in construction, had a mean As concentration of 2231 mg/kg. Dust was not analyzed in homes at the control site.

First Void Urine (FVU) samples were collected from all children, and DNA damage was assessed using comet assays [38]. Children from Villa de la Paz had significantly higher urinary As concentrations (range: 87–323 µg/g of creatinine; geometric mean: 136 µg/g, p < 0.05) and greater DNA damage (p < 0.05) compared to children from the control area [38]. Despite the elevated As and lead (Pb) levels in both sites, the differences in DNA damage were statistically significant. The study reported soil As concentrations ranging from 141 to 11,930 mg kg⁻¹ and dust concentrations between 352 and 9920 mg kg⁻¹ at the high-exposure site [38].

A limitation of this study was that both high exposure and control sites had very high concentrations of As (mean soil As exposure site, 2462 mg kg⁻¹ vs. control, 1019 mg kg⁻¹) and Pb (mean soil Pb exposure site, 748 mg kg⁻¹ vs. control, 410 mg kg⁻¹), which reduces the contrast between groups and limits the generalizability to lower-exposure populations. Additionally, comet assays cannot distinguish between DNA damage caused by Pb vs. As. To address this, the authors referenced an in vitro study showing that As induces apoptosis at concentrations two orders of magnitude lower than Pb in human peripheral blood mononuclear cells [44]. There were higher levels of apoptotic activity in Villa de la Paz children, observed through comet assay findings [38]. Finally, this study did not account for potential confounding factors such as exposure to As from drinking water and social–cultural influences, which may also contribute to the observed health outcomes [38].

3.1.4. Cancer Outcomes from Six Ecological Studies with Different As Impact Scenarios

We identified six ecological studies that investigated associations between soil As and cancer. The first, by Yaffee et al., performed a spatio-specific (single-affected street) cross-sectional study of exposures to a single industrial source of soil As [43]. The remaining five studies reported population-wide environmental soil As concentrations and cancer incidence/cancer mortality rates [39,40,41,42,44]. These five studies examined populations living in areas with elevated As but did not include measurements of biological As exposure concentrations. Most studies adjusted for socioeconomic and demographic factors, and all but one study reported cancer relative risk or incidence rates as cancer outcomes. Instead, Nunez et al. estimated differences in cancer mortality rates [40].

Yaffee et al. investigated the association of cancer incidence with exposure to residential soil As from a former lumber treatment facility [43]. Soil samples were randomly collected from 19 affected households in a Kentucky, USA, neighborhood, and toenail clippings were collected from the 84 people living in those households. The households were not exposed to well water. The median household soil As concentration was 64.8 mg kg⁻¹, and the median toenail As concentration was 0.43 mg kg⁻¹ (IQR: 0.12–0.9). After correcting for socioeconomic factors, tobacco use, and well water, elevated toenail As concentrations were significantly related to digging and weeding in soil, pets that split time between indoors and outdoors, and the consumption of homegrown fruits and vegetables. The authors conducted an ecological assessment of lung or bladder cancer incidence using Kentucky Cancer Registry data for 2000–2013 but revealed no significant association with soil As, despite three (4%) of the residents reporting skin, lung, and unspecified cancer diagnoses. The authors reported elevated standardized incidence rates (SIRs) for bladder and lung cancer at the ZIP code level, but their study lacked the power to detect differences in rates. The authors also noted that toenail clippings are not a standardized biomarker for health outcomes, and the potential contamination of toenail surfaces could have influenced the results [43].

Two Australian studies reported SIRs for areas with surface soil iAs concentrations of >100 mg kg⁻¹ (up to 16,800 mg kg⁻¹) and/or drinking water As concentrations > 0.01 mg L⁻¹ using Victorian Cancer Registry data [39,41]. Analyses of data from 1982 to 1991 showed a 1.06% increase for all cancers, and when stratified into exposure categories, SIRs of prostate cancer and chronic myeloid leukemia were significant, at 1.20 (1.06 ± 1.36) and 1.54 (1.13–2.10) for the high soil plus high water category [39]. No statistically significant increases in cancer incidence were observed in the high soil As category, although the SIR in this category was raised for melanoma and chronic myeloid leukemia. The generalized use of soil data instead of data from individual residential addresses may have led to exposure misclassification in the high soil As category. No exposure–response relationships were observed in any other category. The authors also noted limitations, including low statistical power and confounding from factors such as cigarette smoking, other contaminants in drinking water, pesticide use, “heavy metals”, socioeconomic status, and racial mix [39].

Pearce et al. built on the above work of Hinwood et al. by analyzing cancer registry data from 1984 to 2003 using a geographical correlation approach (kriging) alongside soil data from sites with historical mining activity or environmental concern [41]. Socioeconomic disadvantage was used as a proxy for smoking prevalence. To address potential exposure misclassification, the study compared actual, predicted, and back-transformed soil As values. Among males, increased risks were observed for colon and prostate cancers, leukemia, and melanoma, while females showed increased risk for melanoma and colon cancers. The authors suggested that the association with melanoma may indicate that As acts as a co-carcinogen with ultraviolet (UV) radiation, potentially by inhibiting DNA repair mechanisms. No significant association was found with lung cancer.

The study also identified increasing cancer risk across quintiles of soil As concentration, with stronger associations in socioeconomically disadvantaged areas [41]. However, the authors acknowledged potential anomalies and biases related to small sample sizes in certain statistical areas, limitations of the interpolation method, and residential confounding by socioeconomic status. Key limitations included exposure misclassification, a lack of individual-level confounder adjustment, and the inability to account for population mobility. The authors recommended that a case–control study of exposure and cancer risk would be the best way to confirm these findings [41].

Putila and Guo obtained As soil and stream sediment data (soil and sediment range: <1.476 mg kg⁻¹ to >14,525 mg kg⁻¹) from the US Geological Survey and health and population data from national databases, individual state health departments, and the US Census [42]. Soil As concentrations were calculated by taking the average of all individual point measurements within each county area. They used a Poisson regression model to determine the relationship between soil As concentrations and lung cancer incidence in the USA population by county, after controlling for age, geographical location, smoking, and income. They reported a positive association between As and lung cancer after adjusting for smoking and income, with a 0.4% increase in lung cancer rates (OR: 1.004, 95% CI 1.004–1.004) per 1 mg kg⁻¹ increase in soil As. A regression analysis of smokers with increasing soil As concentrations showed excess lung cancer rates beyond those related to individual factors. Individually, As was found to contribute up to 5297 cases of lung cancer per year when accounting for the various exposure levels. The authors note, however, that some of the counties with the highest associations were sparsely populated and not well served by health facilities. In this study, soil As concentrations were used as a proxy for As contamination of groundwater, which was the assumed source of drinking water in areas of elevated soil and sediment As concentrations. But this assumption was not tested, and no assessment of aquifers and parameters related to groundwater use was reported. These data corroborate the relationship between soil As and cancer; however, the depth of soil sampling was not provided, precluding certainty of the human exposure pathway.

Chen et al. associated “heavy metals” and As (mean: 9.08 mg kg⁻¹; range: 6.14–16.52 mg kg⁻¹) in soil with cancer mortality rates during the years 2005–2010 in Suzhou, China. Cancer associations were tested with soil metal concentrations estimated by averaging and kriging interpolation [44]. Both averaging and kriging exposure estimates were associated with cancer mortality rates (per 100,000 people), and after adjusting for age, “heavy metals”, smoking prevalence, and education, As concentrations in topsoil were significantly (using a 95% confidence interval) associated with deaths from colon (RR = 1.083), gastric (RR = 1.111), kidney (RR = 1.129), lung (RR = 1.050), and nasopharyngeal (RR = 1.086) cancers. Their focus on cancer mortality may have excluded prostate and bladder cancers, because these cancers cause fewer deaths than other cancers. The limitations of the study were noted as a misclassification of exposure and the inability to address exposure duration and adjust for individual-level confounding factors. The authors noted that individual-level studies with exposure measurements and confounder information would confirm their findings. The contamination status of drinking water was not mentioned.

Nunez et al. conducted an ecological cancer mortality study on data from Spanish mainland towns over the years 1999–2008 [40]. The study assessed the health effects of low-dose (mean: 15.16; range: 0.1–2510.0 mg kg⁻¹ As) chronic exposures to metals, according to distributions of metals as defined by geochemical soil compositions from the Geochemical Atlas of Spain and estimated in towns by kriging, at depths of between 0 and 20 cm. After adjusting for sociodemographic variables and proximity to industrial emissions, they demonstrated associations between As soil levels and mortality due to cancers of the stomach, pancreas, lung, brain, and non-Hodgkin lymphomas in men and women. Prostate cancer was also associated with As in soil. Noted limitations included exposure misclassification due to the absence of actual exposure data and the lack of data on other confounding variables such as smoking, although efforts were made to control for these factors.

Four of the six ecological studies reviewed suggest an association between the increased rates of cancer incidence in areas with increasing As concentrations in soil. However, the types of cancers reported do not align across the studies and there were significant limitations, such as exposure misclassification and the inability to assess exposure (duration and mode) and adjust for various confounding factors.

The IARC classifies As and iAs compounds as carcinogenic to humans [18]. There is growing interest in the association between environmental exposure via soil and cancer incidence rates, with several studies reporting significant increases in cancer incidence in areas with elevated As. Socioeconomic disadvantage in the study areas was associated with increased cancer risk where elevated soil As concentrations were present, suggesting the importance of considering relevant socioeconomic factors. However, associations between soil As concentrations and cancer outcomes suffer from various confounders and may not be relevant to individual-level considerations [35]. Future mechanistic studies are needed to help validate the findings reviewed here.

3.2. Exposure to As in Soil–Human Biomarker and Bioaccessibility Studies

Of the six studies that investigated exposure to soil and the intake of As via outdoor activities, four reported biomarker analyses of hair, nails, and urine in people exposed to As from gold tailings in Australia (Table 2) but did not include adverse health effects [45,46,47,48]. The fifth study related biomarker analysis of As in toenails with cancer incidence and is discussed above in the cancer section [43]. In the sixth study, low As exposures from soil (3.1–17.2 mg kg⁻¹ As) and other sources were assessed in children, and the relationship between 24 h urine samples and first-morning spot urine or FVU biomarkers was analyzed [49]. They reported a significant correlation between As concentrations in soil and those in biological samples such as toenails.

Hinwood et al. conducted two cross-sectional studies and found that people living in gold mining areas with elevated As soil concentrations (median: 92.0 mg kg⁻¹) and indoor dust (median 53.0 mg kg⁻¹) had significantly more iAs in their toenails, hair, and FVU than people living in areas without historic gold mining and lower soil iAs concentrations (median, soil: 3.3 mg kg⁻¹; dust: 3.9 mg kg⁻¹) [45,46]. In statistical analyses stratified by soil iAs concentrations above and below 100 mg kg⁻¹, age, gender, season, hours of contact with soil, and excavation activities, the main predictor of iAs FVU concentrations was soil As concentrations above 100 mg kg⁻¹. Seasonal factors, in contrast, were not significantly associated with elevated iAs FVU concentrations. Moreover, iAs concentrations were not increased in FVU samples from participants who provided information that they use pesticides, consume home-grown produce (washed or unwashed), or may be exposed to As occupationally [38]. The authors noted a potential bias from non-compliance with the sampling schedule and recall bias for survey questions on lifestyle factors. Their analysis, nonetheless, showed an exposure–response pattern indicating that iAs concentrations in hair (0.4–27.3 mg kg⁻¹), toenails (3.2–477 mg kg⁻¹), and FVU (<DL–28.4 ug/L) increased with increasing soil As concentrations. In the Hinwood et al. study (n = 153), children (n = 5) had higher iAs concentrations in hair and toenails than adults (n = 22) living in the same areas, suggesting that behavior influences exposure [45]. Although sample contamination precluded a correlation of hair and toenail iAs with As absorption in the children of the 2003 study, the results did link hair and toenail concentrations with total exposures to iAs (surficial and absorbed). The authors concluded that in the absence of definitive exposure and health effects studies, efforts to limit exposures to the soil would be prudent [45]. They also recommend future assessments of chronic As exposures and health impacts.

A cross-sectional study by Pearce et al. examined iAs in soil, and toenail clippings of n = 29 children aged 5–13 years studying at two primary schools, in a historical gold mining area in rural Victoria, Australia [47]. Residential soil iAs ranged from 3.3 to 130 mg kg⁻¹ (geometric mean, 11.5 mg kg⁻¹; n = 22), and children’s toenail iAs concentrations ranged from 0.15 to 2.1 mg kg⁻¹ (geometric mean, 0.49 mg kg⁻¹; n = 29). The study included an assessment of iAs speciation, using X-ray absorption spectroscopy, X-ray absorption near-edge structure, and Secondary Ion Mass Spectrometry to differentiate between As species, including AsIII and AsV. They also assessed the potential direct diffusion of As from mine waste into toenails and measured residual surface contamination. The assessment of iAs species in nails found that iAs were irregularly deposited, and episodic intake during nail growth coincided with the warm season, likely reflecting children’s outdoor play activities. The direct diffusion of iAs into toenails could not be ruled out as an alternative intake pathway to surface contamination. The authors concluded that there was a positive correlation between As concentrations in residential soil and in toenail clippings of resident children.

Martin et al. collected toenail clippings from children (n = 24) of 5 to 17 years at 14 addresses in rural Victoria (Australia) [48]. Residential soil As levels were 3.4–97.1 mg kg⁻¹ (geometric mean, 9.5 mg kg⁻¹; n = 14) and were found in toenail clippings at 0.03–0.71 mg kg⁻¹ (geometric mean, 0.17 mg kg⁻¹; n = 24). Longitudinal analysis revealed that children who spent 1–5 h outdoors per day had significantly higher toenail As than children who spent less than 1 h outdoors per day and demonstrated an inverse correlation with age. Given that the children in this study were all aged >5 years and had limited hand-to-mouth behaviors, these observations were likely related to the proximity of breathing zones to the ground and larger breathing volumes relative to size in younger children [48]. Gender, the frequency of seafood and home-grown vegetable consumption, domestic tobacco smoke exposure, and age were not significant determinants of toenail As levels. Interestingly, a comparison with their study’s 2006 results showed an overall decrease in iAs intake, potentially reflecting increased community awareness since the media coverage of the 2006 study [47,48]. In addition, a significant negative relationship was identified between toenail As concentrations and soil pH, suggesting increased bioavailability in acidic soils. The small sample sizes of the four As exposure studies identified in this review likely undermined many of the associations tested. But together, these studies suggest that high soil iAs concentrations are associated with significant human exposures in the Australian setting [45,46,47,48].

Wang et al. assessed As exposure to young children (n = 120) in central China [49]. No significant impact of soil exposures was demonstrated, and instead, dietary intake was the highest contributor to As in 24 h urine samples and FVU. After adjusting for creatinine, the two measures were significantly correlated, warranting the use of FVU as a more convenient method for measuring As concentrations in urine.

4. Discussion

This review identified and evaluated 14 studies exploring potential adverse health effects from chronic exposure to soil As, as well as biomarker studies related to soil exposure. The small number of studies and publication dates highlight the lack of ongoing research in this area.

Soil As concentrations and exposure

Cohort and cross-sectional studies identified relationships between adverse health effects and soil As concentrations. Effects ranged from adverse birth outcomes, DNA damage in children, neuropathy in adults, and various cancers, with As soil concentrations ranging widely from less than 5 mg kg⁻¹ to 11,930 mg kg⁻¹ [36,37,38,43]. Ecological studies suggest that soil As, even at low levels (<30 mg kg⁻¹), is associated with cancer incidence and cancer mortality rates [40,42,44]. McDermott et al., Putila and Guo, Chen et al., and Nunez et al. assessed administrative data sets from large geographical areas using epidemiological approaches to assess population health at large scales [37,40,42,44]. Linking soil concentrations of any contaminant, including As, across large geographical areas to health outcomes presents significant challenges from an exposure perspective. This is due to the heterogeneous nature of soil and the variability in exposure pathways, which may not remain constant. Variable exposure pathways, combined with the short residence time of As in the body, can complicate the detection of health effects in large-scale ecological studies. To better understand the risk of adverse health outcomes associated with soil As, it is crucial to consider factors such as the soil’s properties (e.g., pH, redox, organic matter, soil type), the As species present, co-contaminants, and the concentrations to which individuals are exposed (bioavailability and bioaccessibility), along with the various exposure pathways.

Cancer types

The types of cancers reported in the ecological studies were inconsistent, raising concerns about the limitations of the study methods. Typically, when a chemical exposure is linked to cancer, there is consistency in the types of cancer outcomes observed. However, the studies reviewed here did not show such consistency, suggesting a need for further investigation to determine if the health outcomes are truly related to soil As concentrations. To clarify whether these specific cancer types are associated with As in soil, more rigorous study designs are necessary. The clearer identification of cancer outcomes is important for understanding this pathway and warrants further research. Due to the shortcomings of exposure assessments in these studies, the causes and validity of the associations remain uncertain. Additionally, Putila and Guo noted that counties with the strongest health–As associations were underserved by healthcare facilities [42]. Although such findings may be predictable, they provide an impetus for policymakers to intervene.

Methods

The studies on health effects from As in soil have several methodological limitations, including the use of statistical methods to estimate soil As concentrations, the lack of biomonitoring data, poor control groups, and limited data that may lead to exposure misclassification. Additionally, confounding factors could undermine the evidence in some of these studies. For most health outcomes, there were few, if any, studies identified in this review.

Ideally, we would have conducted meta-analyses to synthesize the evidence across different studies. However, we were unable to do this because we were unable to meet the meta-analysis criteria of two or more studies with comparable exposures and outcomes [50]. A further limitation is that the epidemiological studies identified as either cross-sectional or ecological and did not assess long-term As exposure. These existing studies are therefore limited in their ability to provide evidence of causation.

Cohort studies are generally considered methodologically stronger than case–control and cross-sectional studies to assess exposures and outcomes, and this review included three of these types of studies [36,37,38]. Although some of the studies herein are consistent with other well-accepted evidence of the health effects of chronic As exposure from drinking water studies, including neurological and birth outcomes, and cancers (including lung, skin, and kidney) [17], others indicated health outcomes (including prostate and gastric cancer, and chronic myeloid leukemia) that were not consistent, pointing to methodological limitations. These methodological limitations included the use of statistical methods to estimate As concentrations, difficulties in assessing the duration of exposure, and the inability to exclude some confounding factors. Standardization of such study methods, including computational methods, adjustments, and assumptions, to enable meta-analysis, could improve the strength of such findings.

Modes of exposure and uptake

This review has identified at-risk subpopulations in areas where soil As concentrations are elevated and the main exposure modes for these populations. Critically, higher exposure groups engage in soil-based activities (e.g., digging, gardening, and playing [43]) and are younger children who spend more time playing outdoors [48]. Given the play and mouthing habits of younger children, the doses of As relative to body weight are often higher than those of adults [43]. Breathing zone proximity to the soil for children and the different metabolic processes of children also likely influence As uptake [48,51].

Although relevant modes of uptake were proposed, their pathways were not directly assessed, and contributions to total uptake were not defined. Evidence shows that increasing soil As concentrations resulted in higher As in people’s blood, FVU, hair, and toenail clippings showed an exposure–response pattern with biomarkers and increasing soil As concentrations [38,39,43,45,46,47,48]. In addition, Hinwood et al. noted that iAs concentrations were not significantly influenced by a variety of factors, including the use of pesticides and the consumption of home-grown produce [46]. These observations support the UK, Australia, and Canada guideline values, which consider direct soil ingestion as the primary exposure pathway [20,52,53].

Biomarkers

This review found that FVU samples are a reliable method for assessing recent As intake, presenting an opportunity for future studies of As intake from soil. Although previously, 24 h urine analysis has been considered the best indicator of recent exposure [23], it is more onerous to collect than FVU, which is a demonstrated proxy for As exposure. As deposition in hair and nails also produces useful biomarkers of chronic As exposure, although very low levels of As are found in these biosamples, and they do not indicate how recently exposure might have occurred [23]. In addition, there are questions regarding the usefulness of hair and nail samples due to the difficulty in separating exogenous and endogenous exposures. Notwithstanding, because soil exposures are likely to be intermittent due to variable exposure (e.g., seasonal variability in outdoor activities), a combination of recent and chronic exposure methods is more likely to elucidate potential actual exposures. Using a dual sampling approach in conjunction with biokinetic models may improve the extrapolations of actual exposures.

Arsenic speciation was determined in one biomarker study because the As species taken up into the body from soil may be an important aspect of understanding health outcomes in specific environments [41].

Many of the biomarker studies had limited sample sizes for specimens or specific age groups, resulting in limited statistical relevance and indicating the need for further studies with larger sample sizes [38,41,45,46,48].

Separating dust and soil exposure

Studies reporting elevated As in dust inside households and the related health effects have elucidated the importance of the wind-blown and household dust uptake pathway [36,38]. But accounting for individual exposures to dust and soil remains difficult, as articulated previously [46]. Overall, the findings of the reviewed exposure assessment studies indicate that both residential soil and household dust are sources of uptake. But mechanisms of exposure are difficult to elucidate due to the absence of implementable methods to differentiate the two sources, hence the reliance on surveys and uptake assays.

Chemistry

Most studies use total As concentrations, and none distinguish between As species and health effects. Hence, future population and cohort studies could differentiate the effects of the As species, soil–As complexes, and soil pH [40] on exposure and subsequent human health risk. The bioavailability and bioaccessibility of As from soil also requires further study, as the transformation of As species in the gastrointestinal tract governs the dose of As, particularly in children [48]. Only one study investigated differing exposure contributions of As intake pathways (inhalation, ingestion, and drinking water) and combined them to understand the true risk of exposure, albeit in an area with low natural soil As concentrations of <17.2 mg kg⁻¹ [49]. Similar studies in areas with elevated soil As concentrations would better test the cumulative uptake risks from multiple sources. In addition to As, several other soil pollutants may influence As absorption into the human body [37,44]. Hence, it is also important to include a broader pollutant range to understand the exposure risk for humans. Mine tailings are known to contain other contaminants, such as Pb and mercury (Hg). The effects of As as a potential co-carcinogen have been suggested, and there may be synergistic effects with other contaminants not elucidated in these studies [38].

Built environment

Research focusing on exposure in regions with elevated levels of As can provide valuable additional lines of evidence regarding health risks. Building on previous studies to understand the relationship between exposure and lifestyle/property factors that potentially influence exposure like groundcover (or how much bare soil there is on a property), the frequency of soil-based activities, wearing shoes indoors, pets, children (who may also track soil into the house), and the use of air conditioners (as an example) can offer insights into effective mitigation strategies [29]. This knowledge is instrumental in developing guidance for communities residing in affected areas, aiding them in adopting appropriate measures to safeguard their health.

Climate

One area of potential interest that was not identified in this review is the climatic effects of As in soil and the impact on human health. People living in temperate, dry, and windy areas may have different risk profiles than those living in colder and wetter areas.

Summary of evidence and research gaps

The primary gaps in ecological, cohort, cross-sectional, and biomarker studies on As exposure and health impacts are multifaceted. The ecological studies reviewed had methodological limitations, confounding factors, and a lack of consistency that limited the ability to compare studies and to assess exposure, making it difficult to establish causality. The cohort and cross-sectional studies, while methodologically stronger, were limited by small sample sizes, potential confounding factors, and challenges in assessing long-term exposure, the total uptake relative to various modes and durations of exposure, and mostly lacked biomonitoring data. The biomarker studies were hindered by small sample sizes and difficulties in separating exogenous and endogenous exposures.

5. Conclusions

This review advances the understanding of As exposure and its potential health impacts linked to soil exposure by synthesizing the limited existing research on this topic. The literature highlights associations between As soil concentrations and various health outcomes, including birth outcomes, neurological effects, DNA damage, and cancer. However, clear links to causality were limited, particularly in studies with low concentrations of As in soil. Epidemiological studies are an initial step in identifying potential health impacts, but it is essential that these studies follow consistent methodologies to ensure that the individual findings are comparable and contribute to the overall body of evidence. To strengthen the evidence, more robust and comprehensive exposure assessments are needed to evaluate the pathways of exposure, dosage, and associated health effects. Future research should focus on standardized epidemiological studies to facilitate meta-analysis, long-term studies, exposure-mitigating factors such as groundcover and weather, combined chronic and recent exposure biomarker studies, and the behavior of the different As species and soil–As complexes. Additionally, the further exploration of soil redox, pH, and soil dose related to bioavailability and bioaccessibility, and the cumulative/synergistic effects of As in combination with other elements is essential.