Evaluation of background exposures of Americans to dioxin-like compounds in the 1990s and the 2000s
Introduction
In April 1991, the United States (US) Environmental Protection Agency (EPA) began a scientific reassessment of the health risks of exposure to 2378-TCDD and dioxin-like compounds (hereafter referred to as the Reassessment). This effort has resulted in a comprehensive multi-volume assessment report (US EPA, 2003a), which was reviewed by the EPA’s Science Advisory Board in 1994 and 2000, and by the National Academies of Science (NAS) beginning in 2004 and concluding in 2006 (NAS, 2006). To date, EPA has not finalized the Reassessment; the NAS review draft (US EPA, 2003a) is the most current draft of the Reassessment. The characterization of background exposures to dioxin-like compounds in the Reassessment included an estimate of a background intake dose, expressed in pg Toxic Equivalent (TEQ)/day or pg TEQ/kg body weight/day, and a body burden, expressed in pg TEQ/g lipid weight (pg/g lwt), or pg TEQ/g whole weight (pg/g wwt), representative of the general population of the US. While information was provided on childhood exposures and exposures to special populations, this update focuses on the background adult exposure estimates provided in the Reassessment.
The NAS Review Draft (US EPA, 2003a) of the Reassessment provided these estimates: the background daily dose was 65.8 pg TEQ/day (42.7 pg of dioxin/furan, or PCDD/PCDF, TEQ and 23.1 pg coplanar PCB TEQ) and the background body burden was provided as 25.4 pg TEQ/g lipid (20.1 pg/g of PCDD/PCDF TEQ and 5.3 pg/g coplanar PCB TEQ). These estimates were developed as arithmetic averages using data generated during the middle 1990s. Congener averages and resulting TEQ concentrations were derived assuming non-detects were equal to one-half the detection limit (ND = ½ DL). These estimates also were derived using the World Health Organization (WHO) 1998 Toxic Equivalency Factor (TEF) scheme (Van den Berg et al., 1998). To avoid confusion, all TEQ values in this study were derived using the WHO-2005 TEFs (Van den Berg et al., 2006). In addition to a rederivation based on a different TEF scheme, these NAS Review Draft estimates were re-derived assuming non-detects are equal to zero (ND = 0).
The purpose of this study is twofold: to provide updated estimates of dose and body burden of dioxin-like compounds, using data collected between 2000 and 2004, and to examine the data more closely to determine whether declines have occurred. This examination highlights the disparity in study design, data sources, and laboratories conducting the trace analyses for the background food samples and blood. These updated dose and body burdens are derived in a manner that is as analogous as possible to the mid-1990s estimates of the Reassessment, so that comparisons of the two sets of estimates can be meaningful. Current estimates will be derived using both substitution methods, ND = ½ DL and ND = 0. The update on the intake dose will use more recent data on PCDD/PCDFs and PCBs in food. The body burden data from the Reassessment included six different regional studies of PCDD/PCDFs and PCBs in background populations, and this will be updated using the more valid national, statistical survey known as the National Health and Nutritional Examination Survey, or NHANES. Statistically based, national averages of congener concentrations in blood were derived from individuals within the 2001/2 NHANES data set, and these averages were derived at ND = 0 and ND = DL/√2.
In addition to these substitutions, a variation is to substitute non-detects with the detection limit (ND = DL), or even more rigorously, to statistically extrapolate below the detection limit based on the distribution of the population above the detection limit. While arguments can be presented for adoption of any of the substitution methods, the Reassessment concluded that the background characterizations were best generated at ND = ½ DL. That is used here for generation of intake doses, although the available analysis of NHANES data used the alternative, ND = DL/√2. If the analytical detection limits are sufficiently low and/or quantifications are possible for a large majority of congener measurements in a population, then average congener concentrations calculated at ND = 0 and any of the other substitution methods will be similar. On the other hand, a meaningful disparity in congener averages (or other statistical measures, for that matter) at ND = 0 versus one of the other methods suggests that the detection limits may be too high for the matrix being studied. For example, if 2378-TCDD (the most toxic congener) is not detected frequently (as is often the case), then the TEQ concentration can be heavily influenced by which substitution method is used for non-detects for 2378-TCDD. A primary focus of this study is to evaluate the impact of the substitution method on the generation of survey average congener and TEQ concentrations, and the subsequent impact to the background characterizations of dioxin dose and body burden.
It is noted that the focus of this evaluation is on TEQ and not on the individual congeners which comprise the TEQ. Such an analysis could mask different trends in the exposure to individual congeners.
Section snippets
Methods
In the Reassessment, dose estimates were provided for inhalation, soil ingestion, soil dermal contact, water ingestion, and for 10 food ingestion categories including beef, pork, poultry, “other meats” (game, lamb, unidentified meat in casseroles, etc.), eggs, milk, dairy, marine fish, freshwater fish, and vegetable oils. Each pathway estimate was derived as a point estimate using an arithmetic average adult per capita contact rate (i.e., inhalation rate, water ingestion rate, food ingestion
Food ingestion rates
The beef/pork/poultry/dairy/milk ingestion rates in the Reassessment were mean adult per capita ingestion rates taken from EPA’s 1997 Exposure Factors Handbook (EFH; US EPA, 1997), and rates in that guidance were developed from the United States Department of Agriculture (USDA) Continuing Survey of Food Intakes among Individuals (CSFII) 1-day 1989–1991 survey. EPA has now analyzed the CSFII 2-day 1994–1996 surveys (US EPA, 2003b) but arrives at essentially the same food consumption rates
Summary and conclusions
This study attempts to provide an update to the background body burden and intake dose estimates described in EPA’s Reassessment (US EPA, 2003a). These estimates were current as of the mid-1990s and the updates are current as of the early 2000s. Besides simply updating the exposure estimates, this study examines the data to ascertain whether a decline in exposures has occurred during this time. Using the same procedures and analogous data when available, the intake dose calculated at 61.0 pg
Disclaimer
The views expressed in this article are those of the authors and do not necessarily reflect the views or policies of the US Environmental Protection Agency, the Centers for Disease Control, and the US Department of Agriculture. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.
Acknowledgments
The authors appreciate the data on NHANES 2001/2002 supplied by Laura Scott (formerly Laura Ferriby of Ferriby et al., 2007) of Chemrisk (10375 Richmond Avenue, Suite 350, Houston, TX 77042). We are also appreciative of the careful reviews provided by EPA reviewers David Cleverly, John Schaum, and Joseph Ferrario.
References (31)
- et al.
Geometric mean estimation from pooled samples
Chemosphere
(2007) - et al.
Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/PCDF) in food samples collected in southern Mississippi, USA
Chemosphere
(1997) - et al.
Concentrations of environmental chemicals associated with neurodevelopmental effects in US population
NeuroToxicology
(2005) - et al.
Total TEQ reference range (PCDDs, PCDFs, cPCBs, mono-PCBs) for the US population 2001–2002
Chemosphere
(2008) - et al.
Levels of dioxins, dibenzofurans, PCB and DDE congeners in pooled food samples collected in 1995 at supermarkets across the United States
Chemosphere
(1997) - et al.
A statistical survey of dioxin-like compounds in United States beef: a progress report
Chemosphere
(1996) On summarizing group exposures in risk assessment: is an arithmetic mean or a geometric mean more appropriate?
Risk Analysis
(2006)- FDA, 2007. Dioxin Analysis Results/Exposure Estimates. Web site posted by Office of Plant and Dairy Foods, Center for...
- et al.
A statistical survey of dioxin-like compounds in the United States poultry fat
Organohalogen Compounds
(1997) - et al.
Evaluation of PCDD/PCDF and dioxin-like PCB serum concentration data from the 2001–2002 national health and nutrition examination survey of the Unites States population
Journal of Exposure Science and Environmental Epidemiology
(2007)
PCDD and PCDF levels in baby food made from chicken produced before and after 1997 in the United States
Organohalogen Compounds
Statistically designed survey of polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans, and co-planar polychlorinated biphenyls in US meat and poultry, 2002–2003: results, trends, and implications
Environmental Science and Technology
Exposure assessment of dioxins/furans consumed in dairy foods and fish
Food Additives and Contaminants
Perspective on serum dioxin levels in the United States: an evaluation of the NHANES data
Journal of Exposure Science and Environmental Epidemiology
Cited by (47)
Polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) in food and human dietary intake: An update of the scientific literature
2021, Food and Chemical ToxicologyDioxin-like compounds in paired maternal serum and breast milk under long sampling intervals
2020, Ecotoxicology and Environmental SafetyCitation Excerpt :Significant declines of PCDD/Fs in serum were observed worldwide during 1985–2008 (Consonni et al., 2012). In recent years, levels of dioxin-like compounds in serum of people from several industrial countries or regions, such as the United States (Lorber et al., 2009), German (Fromme et al., 2016), Japan (Konishi et al., 2009; Kishi et al., 2017), and Taiwan (Hsu et al., 2009), were reported. This information helped us to assess changes in body burden of dioxin-like compounds in general population.
Chemical Contamination of Red Meat
2017, Chemical Contaminants and Residues in Food: Second EditionComparison of industrial emissions and carpet dust concentrations of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans in a multi-center U.S. study
2017, Science of the Total EnvironmentCitation Excerpt :The predominant PCDD/F exposure pathway in the general population is considered dietary ingestion of animal products, which accumulate the lipophilic PCDD/F compounds (Dougherty et al., 2000, U.S. EPA, 1994, USDA, 2015, Charnley and Doull, 2005). Non-dietary exposure pathways, such as non-dietary ingestion, inhalation, and dermal contact of indoor dust, have received relatively little attention, as they are estimated to contribute < 10% of total exposure in the U.S. general population (Lorber et al., 2009). However, exposures from dust could be important contributors to total PCDD/F exposures in certain populations, such as those living in close proximity to PCDD/F-emitting sources, those who do not consume a lot of animal products, those who spend a lot of time at home, or children who spend a large portion of time on the floor and exhibit hand-to-mouth contact (Tue et al., 2013).
Ancestral TCDD exposure promotes epigenetic transgenerational inheritance of imprinted gene Igf2: Methylation status and DNMTs
2015, Toxicology and Applied PharmacologyProposing a scientific confidence framework to help support the application of adverse outcome pathways for regulatory purposes
2015, Regulatory Toxicology and Pharmacology