Evaluating the effectiveness of fish consumption advisories: modeling prenatal, postnatal, and childhood exposures to persistent organic pollutants

Considerations and challenges in support of science and communication of fish consumption advisories for per- and polyfluoroalkyl substances

Federal, state, tribal, or local entities in the United States issue fish consumption advisories (FCAs) as guidance for safer consumption of locally caught fish containing contaminants. Fish consumption advisories have been developed for commonly detected compounds such as mercury and polychlorinated biphenyls. The existing national guidance does not specifically address the unique challenges associated with bioaccumulation and consumption risk related to per- and polyfluoroalkyl substances (PFAS). As a result, several states have derived their own PFAS-related consumption guidelines, many of which focus on one frequently detected PFAS, known as perfluorooctane sulfonic acid (PFOS). However, there can be significant variation between tissue concentrations or trigger concentrations (TCs) of PFOS that support the individual state-issued FCAs. This variation in TCs can create challenges for risk assessors and risk communicators in their efforts to protect public health. The objective of this article is to review existing challenges, knowledge gaps, and needs related to issuing PFAS-related FCAs and to provide key considerations for the development of protective fish consumption guidance. The current state of the science and variability in FCA derivation, considerations for sampling and analytical methodologies, risk management, risk communication, and policy challenges are discussed. How to best address PFAS mixtures in the development of FCAs, in risk assessment, and establishment of effect thresholds remains a major challenge, as well as a source of uncertainty and scrutiny. This includes developments better elucidating toxicity factors, exposures to PFAS mixtures, community fish consumption behaviors, and evolving technology and analytical instrumentation, methods, and the associated detection limits. Given the evolving science and public interests informing PFAS-related FCAs, continued review and revision of FCA approaches and best practices are vital. Nonetheless, consistent, widely applicable, PFAS-specific approaches informing methods, critical concentration thresholds, and priority compounds may assist practitioners in PFAS-related FCA development and possibly reduce variability between states and jurisdictions. Integr Environ Assess Manag 2024;00:1-20. © 2024 SETAC.

Assessing awareness and compliance with fish consumption advisories on the upper Hudson River: Implications for risk management of the Hudson River Superfund site

Polychlorinated biphenyls (PCBs) are a common environmental contaminant. The NYS Department of Health (DOH) issues fish consumption advisories to limit consumption of PCB-contaminated fish. Fish consumption advisories are utilized as institutional controls within the Hudson River Superfund site to limit exposure to PCBs. There is a "Do Not Eat" advisory for all species caught in the upper Hudson River, from Glens Falls, NY to Troy, NY. The section of the river below Bakers Falls also has a catch-and-release regulation issued by the NYS Department of Environmental Conservation. There is limited research on the effectiveness of these advisories in preventing consumption of contaminated fish in the context of Superfund site risk management. We surveyed individuals actively fishing on the upper Hudson River in the area with a "Do Not Eat" advisory, specifically between Hudson Falls and the Federal Dam in Troy, NY. The goal of the survey was to assess knowledge of the consumption guidelines, and if the guidelines are effectively preventing exposure to PCBs. A subset of individuals continue to consume fish caught from the upper Hudson River Superfund site. Awareness of advisories was inversely related to fish consumption from the Superfund site. Age, race, and possession of a fishing license were associated with overall awareness of fish consumption guidelines; age and possession of a license were associated with awareness of the "Do Not Eat" advisory. While institutional controls appear to have a beneficial impact, there is incomplete awareness and compliance with advisories and regulations aimed at preventing exposure to PCBs from fish consumption. Risk assessment and management strategies for contaminated fisheries should consider imperfect adherence to fish consumption guidelines.

An introduction to the sources, fate, occurrence and effects of endocrine disrupting chemicals released into the environment

Many classes of compounds are known or suspected to disrupt the endocrine system of vertebrate and invertebrate organisms. This review of the sources and fate of selected endocrine disrupting chemicals (EDCs) in the environment includes classes of compounds that are "legacy" contaminants, as well as contaminants of emerging concern. EDCs included for discussion are organochlorine compounds, halogenated aromatic hydrocarbons, brominated flame retardants, per- and polyfluoroalkyl substances, alkylphenols, phthalates, bisphenol A and analogues, pharmaceuticals, drugs of abuse and steroid hormones, personal care products, and organotins. An exhaustive survey of the fate of these contaminants in all environmental media (e.g., air, water, soil, biota, foods and beverages) is beyond the scope of this review, so the priority is to highlight the fate of EDCs in environmental media for which there is a clear link between exposure and endocrine effects in humans or in biota from other taxa. Where appropriate, linkages are also made between the fate of EDCs and regulatory limits such as environmental quality guidelines for water and sediments and total daily intake values for humans.

Methods for the analysis of endocrine disrupting chemicals in selected environmental matrixes

Endocrine disrupting chemicals (EDCs) are heterogenous in structure, chemical and physical properties, and their capacity to partition into various environmental matrixes. In many cases, these chemicals can disrupt the endocrine systems of vertebrate and invertebrate organisms when present at very low concentrations. Therefore, sensitive and varied analytical methods are required to detect these compounds in the environment. This review summarizes the analytical methods and instruments that are most used to monitor for EDCs in selected environmental matrixes. Only those matrixes for which there is a clear link between exposures and endocrine effects are included in this review. Also discussed are emerging methods for sample preparation and advanced analytical instruments that provide greater selectivity and sensitivity.

Risk tradeoffs associated with traditional food advisories for Labrador Inuit

The traditional Inuit diet includes wild birds, fish and marine mammals, which can contain high concentrations of the neurotoxicant methylmercury (MeHg). Hydroelectric development may increase MeHg concentrations in traditional foods. Consumption advisories are often used to mitigate such risks and can result in reduced intake of traditional foods. Data from a dietary survey, MeHg exposure assessment and risk analysis for individuals in three Inuit communities in Labrador, Canada (n = 1145) in 2014 indicate reducing traditional food intake is likely to exacerbate deficiencies in n-3 polyunsaturated fatty acids and vitamins B12 and B2. Traditional foods accounted for < 5% of per-capita calories but up to 70% of nutrients consumed. Although consumption advisories could lower neurodevelopmental risks associated with an increase in MeHg exposure (90th-percentile ∆IQ = - 0.12 vs. - 0.34), they may lead to greater risks of cardiovascular mortality (90th-percentile increase: + 58% to + 116% vs. + 25%) and cancer mortality (90th-percentile increase + 2% to + 4% vs. no increase). Conversely, greater consumption of locally caught salmon mostly unaffected by hydroelectric flooding would lower all these risks (90th-percentile ∆IQ = + 0.4; cardiovascular risk: - 45%; cancer risk: - 1.4%). We thus conclude that continued consumption of traditional foods is essential for Inuit health in these communities.

Differential contribution of animal and vegetable food items on persistent organic pollutant serum concentrations in Spanish adults. Data from BIOAMBIENT.ES project

Diet is considered the main source of Persistent Organic Pollutant (POP) exposure in the general population, although there are still several gaps of knowledge regarding the differential contribution of main food groups. The aim of this study was to identify dietary patterns that contribute to human exposure to organochlorine pesticides (OCPs), polychlorinated biphenyls (PCBs), and Perfluoroalkyl Substances (PFASs). Study population (n=1880, 18-65years old) was recruited during 2009-2010 in all the main geographical areas of Spain. For this work, exposure was estimated by chemical analyses of serum levels of 6 PCBs (n=1880), 13 OCPs (n=934), and 6 (PFASs) in a subsample of 755 (n=755). Dietary habits and covariates were gathered via self-administered questionnaires. Data analyses were performed by means of multivariable linear regression and weighted quantile sum regression. Both the consumption of animal-based and plant-based food were positively associated with the individual concentrations of p,p´-DDE, hexachlorobenzene, and PCB-congeners -138, -153, and-180. The contribution of animal-based products was 2.1-4.0× stronger except in p,p´-DDE, to which both patterns had similar contributions. In PFASs only animal food was positively associated with the exposure levels. The main animal-based contributors to PCB exposure were fish (49-64%) and eggs (19-36%), while OCP concentrations were mainly influenced by dairy products (32-48%) and fish (47-48%). PFOA and PFHxS were mainly explained by cold-meat (34-37%), fish (25-26%), and eggs (19-21%), while PFOS and PFDA were primarily influenced by fish consumption (44-77%). In the case of plant-based items, fruits (25-82%) and vegetables (18-63%) accounted for the majority of the variability of PCB and OCP concentrations. Our results highlight the relevance of dietary POP exposure as well as the need for the consideration of nutritional interventions in public health programs aiming to reduce POP exposure in the general population.

Reproductive factors, lifestyle and dietary habits among pregnant women in Greenland: The ACCEPT sub-study 2013-2015

BACKGROUND

During past decades the formerly active lifestyle in Greenland has become sedentary, and the intake of traditional food has gradually been replaced with imported food. These lifestyle and dietary habits may affect pregnant women.

AIM

To describe age and regional differences in reproductive factors, lifestyle and diet among Greenlandic pregnant women in their first trimester.

METHODS

A cross-sectional study during 2013-2015 including 373 pregnant women was conducted in five Greenlandic regions (West, Disko Bay, South, North and East). Interview-based questionnaires on reproductive factors, lifestyle and dietary habits were compared in relation to two age groups (median age ≤28 years and >28 years).

RESULTS

In total, 72.4% were Inuit, 46.6% had BMI >25.0 kg/m², 29.0% were smoking during pregnancy and 54.6% had used hashish. BMI, educational level, personal income, previous pregnancies and planned breastfeeding period were significantly higher in the age group >28 years of age compared to the age group ≤28 years of age. In region Disko Bay, 90.9% were Inuit, in region South more had a university degree (37.9%) and region East had the highest number of previous pregnancies, the highest number of smokers during pregnancy and the most frequent intake of sauce with hot meals and fast-food.

CONCLUSIONS

Overall a high BMI and a high smoking frequency were found. Age differences were found for BMI and planned breastfeeding period, while regional differences were found for smoking and intake of sauce with hot meals and fast-food. Future recommendations aimed at pregnant women in Greenland should focus on these health issues.

High selenium exposure lowers the odds ratios for hypertension, stroke, and myocardial infarction associated with mercury exposure among Inuit in Canada

BACKGROUND

Selenium (Se) has been reported to protect against the neurotoxicity of mercury (Hg). However, the effect of Se against Hg on cardiovascular diseases remains unclear. Inuit living in the Arctic have high exposure to both Se and Hg through their marine mammal and fish rich traditional diet.

OBJECTIVE

To characterize the co-exposure of Hg and Se among Inuit in Canada and to assess the associations between Hg, Se and cardiovascular health outcomes, including stroke, hypertension, and myocardial infarction (MI).

METHODS

Data was collected from the International Polar Year Inuit Health Survey (IHS) conducted in 2007 and 2008. Blood Se and Hg were measured, and self-report cardiovascular health outcomes were collected through a questionnaire interview from 2169 adults aged 18 and above.

RESULTS

The mean age was 42.4years, and 38.7% of the participants were male. The geometric means (GM) of blood Se and total Hg were 319.5μg/L and 7.0μg/L, respectively. The crude prevalence of heart attack, stroke and hypertension were 3.55%, 2.36%, and 24.47% respectively. Participants were categorized into 4 exposure groups according to blood Hg (high: ≥7.8μg/L; low: <7.8μg/L), and Se (high: ≥280μg/L; low: <280μg/L). The odds ratio (OR) of cardiovascular outcomes were estimated using general linearized models. Results showed the low Se and high Hg group had a higher prevalence of cardiovascular disease (OR=1.76 for hypertension, 1.57 for stroke, and 1.26 for MI. However, the prevalence was decreased in both the high Se and low Hg group (OR=0.57 for hypertension, 0.44 for stroke, and 0.27 for MI) and the high Se and high Hg group (OR=1.14 for hypertension, 0.31 for stroke, and 0.80 for MI).

CONCLUSIONS

The high Se and low Hg group had the lowest prevalence of cardiovascular outcomes, except for stroke. These results provide evidence that Se may exhibit a protective effect against Hg on cardiovascular disease.

New approaches in human health risk assessment

Studies on the precise impact of environmental pollutants on human health are difficult to undertake and interpret, because many genetic and environmental factors influence health at the same time and to varying degrees. Our chapter in the AMAP report was based on new approaches to describe risks and future needs. In this paper, we will introduce the issues associated with risk assessment of single chemicals, and present suggestions for future studies as well as a summary of lessons learned during the health-related parts of the European Union-funded FP7 project ArcRisk (Arctic Health Risks: Impacts on health in the Arctic and Europe owing to climate-induced changes in contaminant cycling, 2009–2014; www.arcrisk.eu).

An update on risk communication in the Arctic

Background

Arctic residents can be exposed to a wide range of contaminants through consumption of traditional (country) foods (i.e. food from wild animals and plants that are hunted, caught or collected locally in the Arctic). Yet these foods provide excellent nutrition, promote social cohesion, meet some spiritual needs for connectedness to the land and water, reinforce cultural ties, are economically important and promote overall good health for many. The risk and benefit balance associated with the consumption of traditional Arctic foods is complicated to communicate and has been referred to as the “Arctic Dilemma”. This article gives an update on health risk communication in the Arctic region. It briefly summarizes some research on risk communication methodologies as well as approaches to an evaluation of the outcomes of risk communication initiatives. It provides information on specific initiatives in several Arctic countries, and particularly those that were directed at Indigenous populations. This article also summarizes some international versus local risk communication activities and the complexity of developing and delivering messages designed for different audiences. Finally, the potential application of social media for risk communication and a summary of “best practices” based on published literature and a survey of Inuit in a few Arctic countries are described.

Conclusion

Several of the risk communication initiatives portrayed in this article indicate that there is only limited awareness of the outcome of risk communication messages. In some cases, risk communication efforts appear to have been successful, at least when effectiveness is measured in an indirect way, for example, by lower contaminant levels. However, due to missing effectiveness evaluation studies, uncertainty remains as to whether a specific risk communication method was successful and could be clearly linked to behavioural changes that resulted in decreased contaminant exposure.

Deterministic modeling of the exposure of individual participants in the National Health and Nutrition Examination Survey (NHANES) to polychlorinated biphenyls

A population's exposure to persistent organic pollutants, e.g., polychlorinated biphenyls (PCBs), is typically assessed through national biomonitoring programs, such as the United States National Health and Nutrition Examination Survey (NHANES). To complement statistical methods, we use a deterministic modeling approach to establish mechanistic links between human contaminant concentrations and factors (e.g. age, diet, lipid mass) deemed responsible for the often considerable variability in these concentrations. Lifetime exposures to four PCB congeners in 6128 participants from NHANES 1999-2004 are simulated using the ACC-Human model supplied with individualized input parameters obtained from NHANES questionnaires (e.g., birth year, sex, body mass index, dietary composition, reproductive behavior). Modeled and measured geometric mean PCB-153 concentrations in NHANES participants of 13.3 and 22.0 ng g^-1 lipid, respectively, agree remarkably well, although lower model-measurement agreement for air, water, and food suggests that this is partially due to fortuitous error cancellation. The model also reproduces trends in the measured data with key factors such as age, parity and sex. On an individual level, 62% of all modeled concentrations are within a factor of three of their corresponding measured values (Spearman r_s = 0.44). However, the model attributes more of the inter-individual variability to differences in dietary lipid intake than is indicated by the measured data. While the model succeeds in predicting levels and trends on the population level, the accuracy of individual-specific predictions would need to be improved for refined exposure characterization in epidemiological studies.

Balancing the benefits and costs of traditional food substitution by indigenous Arctic women of childbearing age: Impacts on persistent organic pollutant, mercury, and nutrient intakes

For indigenous Arctic Canadians, traditional food consumption represents a key source of nutrients and environmental contaminants. Particularly, ingestion of marine mammal blubber and meat may lead to persistent organic pollutant levels and mercury intakes that exceed regulatory thresholds for sensitive populations. We investigated whether temporary adjustments to the consumption of traditional food derived from marine mammals appreciably impacted contaminant exposure and nutrient intakes among indigenous women of childbearing age. Such adjustments can be motivated by the desire to lower contaminant exposure or to increase nutrition, or by the diminishing availability of other traditional food sources. We combined the contaminant fate and transport model GloboPOP with the food chain bioaccumulation model ACC-Human Arctic to simulate polychlorinated biphenyl exposures in female 2007-08 Inuit Health Survey participants. We also calculated daily mercury and nutrient intake rates. Our results suggest that a temporary decrease in marine mammal consumption is largely ineffective at reducing exposure to polychlorinated biphenyls, because of their long elimination half-lives. In contrast, substitution of marine mammals was highly efficient at reducing mercury intake, but also appreciably lowered intakes of iron, manganese, selenium, and ω-3 polyunsaturated fatty acids. The impact of increasing intake of traditional food derived from marine mammals during childbearing age greatly depended on baseline consumption rates; replacement is ill-advised for those who already consume a lot of traditional food due to greater polychlorinated biphenyl and mercury exposures, while replacement was potentially beneficial for those with very limited marine mammal consumption due to increased nutrient intakes. Our calculations primarily suggest that considering baseline traditional food intake rates is critical to devising reproductive dietary adjustment strategies that maximize nutrient intake while minimizing environmental contaminant exposure.

Mechanistic polychlorinated biphenyl exposure modeling of mothers in the Canadian Arctic: the challenge of reliably establishing dietary composition

BACKGROUND

Traditional food (TF) consumption represents the main route of persistent organic pollutant (POP) exposure for indigenous Arctic Canadians. Ongoing dietary transitions away from TFs and toward imported foods (IFs) may contribute to decreasing POP exposures observed in these groups.

METHODS

To explore this issue, we combined the global fate and transport model GloboPOP and the human food chain bioaccumulation model ACC-Human Arctic to simulate polychlorinated biphenyl (PCB) exposure in two indigenous Arctic Canadian communities from the Inuvik region, Northwest Territories and Baffin region, Nunavut. Using dietary survey information from initial (1996-98) and follow-up (2005-07) biomonitoring campaigns in Inuvik and Baffin, we simulated PCB exposures (PCB-118, -138, -153, and -180) for each individual study participant and also whole study populations.

RESULTS

TF intake rates, particularly of marine mammals (MMs), were the most important predictors of modeled PCB exposure, while TF consumption did not associate consistently with measured PCB exposures. Further, reported mean TF intake increased from baseline to follow-up in both Inuvik (from 8 to 183gd(-1)) and Baffin (from 60 to 134gd(-1)), opposing both the expected dietary transition direction and the observed decrease in human PCB exposures in these communities (ΣPCB Inuvik: from 43 to 29ngglipid(-1), ΣPCB Baffin: from 213 to 82ngglipid(-1)). However dietary questionnaire data are frequently subject to numerous biases (e.g., recall, recency, confirmation), and thus casts doubt on the usefulness of these data.

CONCLUSIONS

Ultimately, our model's capability to reproduce historic PCB exposure data in these two groups was highly sensitive to TF intake, further underscoring the importance of accurate TF consumption reporting, and clarification of the role of dietary transitions in future POP biomonitoring of indigenous Arctic populations.

Estimating Time-Varying PCB Exposures Using Person-Specific Predictions to Supplement Measured Values: A Comparison of Observed and Predicted Values in Two Cohorts of Norwegian Women

BACKGROUND

Studies on the health effects of polychlorinated biphenyls (PCBs) call for an understanding of past and present human exposure. Time-resolved mechanistic models may supplement information on concentrations in individuals obtained from measurements and/or statistical approaches if they can be shown to reproduce empirical data.

OBJECTIVES

Here, we evaluated the capability of one such mechanistic model to reproduce measured PCB concentrations in individual Norwegian women. We also assessed individual life-course concentrations.

METHODS

Concentrations of four PCB congeners in pregnant (n = 310, sampled in 2007-2009) and postmenopausal (n = 244, 2005) women were compared with person-specific predictions obtained using CoZMoMAN, an emission-based environmental fate and human food-chain bioaccumulation model. Person-specific predictions were also made using statistical regression models including dietary and lifestyle variables and concentrations.

RESULTS

CoZMoMAN accurately reproduced medians and ranges of measured concentrations in the two study groups. Furthermore, rank correlations between measurements and predictions from both CoZMoMAN and regression analyses were strong (Spearman's r > 0.67). Precision in quartile assignments from predictions was strong overall as evaluated by weighted Cohen's kappa (> 0.6). Simulations indicated large inter-individual differences in concentrations experienced in the past.

CONCLUSIONS

The mechanistic model reproduced all measurements of PCB concentrations within a factor of 10, and subject ranking and quartile assignments were overall largely consistent, although they were weak within each study group. Contamination histories for individuals predicted by CoZMoMAN revealed variation between study subjects, particularly in the timing of peak concentrations. Mechanistic models can provide individual PCB exposure metrics that could serve as valuable supplements to measurements.

Serum concentrations of PCBs and OCPs among prepubertal Korean children

Although children are more vulnerable than adults to environmental hazards, due to their physiological and behavioral characteristics, few studies have assessed their internal exposure to persistent organic pollutants (POPs). The objective of this study was to determine the serum levels of certain polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs) among pre-pubertal children in Korea and examine their variation with age and gender. We analyzed a total of 51 POPs in 214 children (age range, 7-9 years), using data from the Ewha Birth and Growth Cohort Study. To examine change over time, POP concentrations were measured at both 7 and 9 years of age in a sub-cohort of 26 children. The median and interquartile range (IQR, 25th to 75th percentiles) data for the sum of PCB congeners was 26.44 ng/g lipid (18.27-37.63 ng/g lipid). The median (IQR) concentration for the sum of OCPs was 74.82 ng/g lipid (51.11-112.45 ng/g lipid). Our results showed significant increases in serum concentrations of total PCBs with age (p = 0.0001), no gender dependence of total PCBs (p = 0.38) and total OCPs (p = 0.12), and strong correlations between individual POPs. Childhood exposure to POPs was relatively low compared to other studies carried out in different countries. However, even though the use of POPs has been banned since 2001, our results indicated that children in the general population are still exposed to various POPs. Because long-term, low-level exposure to POPs in the general population remains a concern for human health, future research should examine major routes of exposure and identification of risk factors.

Assessing determinants of maternal blood concentrations for persistent organic pollutants and metals in the eastern and western Canadian Arctic

Aboriginal peoples in the Canadian Arctic are exposed to persistent organic pollutants (POPs) and metals mainly through their consumption of a traditional diet of wildlife items. Recent studies indicate that many human chemical levels have decreased in the north, likely due to a combination of reduced global chemical emissions, dietary shifts, and risk mitigation efforts by local health authorities. Body burdens for chemicals in mothers can be further offset by breastfeeding, parity, and other maternal characteristics. We have assessed the impact of several dietary and maternal covariates following a decade of awareness of the contaminant issue in northern Canada, by performing multiple stepwise linear regression analyses from blood concentrations and demographic variables for 176 mothers recruited from Nunavut and the Northwest Territories during the period 2005-2007. A significant aboriginal group effect was observed for the modeled chemicals, except for lead and cadmium, after adjusting for covariates. Further, blood concentrations for POPs and metals were significantly associated with at least one covariate of older age, fewer months spent breastfeeding, more frequent eating of traditional foods, or smoking during pregnancy. Cadmium had the highest explained variance (72.5%) from just two significant covariates (current smoking status and parity). Although Inuit participants from the Northwest Territories consumed more traditional foods in general, Inuit participants from coastal communities in Nunavut continued to demonstrate higher adjusted blood concentrations for POPs and metals examined here. While this is due in part to a higher prevalence of marine mammals in the eastern Arctic diet, it is possible that other aboriginal group effects unrelated to diet may also contribute to elevated chemical body burdens in Canadian Arctic populations.