The Chemical Body Burden of Environmental Justice Leaders
Pollution in 5 Extraordinary Women
The Chemical Body Burden of Environmental Justice Leaders
An unprecedented two-year study commissioned by the Environmental Working Group and conducted by four independent research laboratories in the United States, Canada and the Netherlands has documented up to 481 toxic chemicals in the blood of five minority women leaders in the environmental justice movement.
The women leaders, from New Orleans, Green Bay, Corpus Christi and Oakland, have spent years deeply engaged in battles to rid their communities of air and water pollution from local manufacturing plants, hazardous waste dumps, oil refineries and conventional agriculture.
Who's in the Study — 5 Environmental Justice Leaders in 4 States
New Orleans, LA
Works to protect citizens from pollution in "Cancer Alley"
Green Bay, WI
Protects natural resources for Oneida Nation
Corpus Christi, TX
Defends her community from oil refinery pollution
Corpus Christi, TX
Helped expose local polluters violating the Clean Air Act
Leads environmental justice efforts for Asian Americans
75 chemicals tested
The study, sponsored by EWG in conjunction with Rachel's Network, a nationwide organization of women environmental leaders, tested the five women last year for 75 chemical contaminants.
Testing was targeted toward compounds that are heavily used in everyday consumer products but that have escaped effective regulation under the antiquated Toxic Substances Control Act (TSCA). The results underscore the widespread and systemic failure of current law to protect the public from chemicals, many of which persist in the environment for decades or far longer, that are associated in animal studies with cancer, reproductive problems and behavioral effects.
All of the women were contaminated with flame retardants, Teflon chemicals, synthetic fragrances, the plastics ingredient bisphenol A and the rocket fuel component perchlorate.
Though they live thousands of miles apart, come from distinctive cultural traditions and confront different environmental hazards outside their homes, the women's differences are only skin deep.
Their body burdens of environmental pollutants, a mix of industrial chemicals, synthetic cosmetics ingredients and chemicals used to treat consumer products, are strikingly similar - and roughly equivalent to the body burdens of other Americans surveyed by governmental and independent research organizations.
- Tested positive for 35 to 60 percent of the 75 chemicals on the search list.
- Had a high body burden of at least one controversial chemical whose lack of regulation and widespread presence in American life is fueling debate over reform of the nation's toxic chemical policies.
The laboratory analyses, which offer a snapshot of the toxic body burdens of women on the front lines of the environmental health and environmental justice movements, set the stage for larger, population-scale research projects that could determine how exposure to chemicals in water, food and consumer products may vary across minority populations; what other industrial compounds may also be present in Americans' bodies; and any health risks those pollutants may pose, alone or in combination.
Even now, the findings could well mark a turning point in the increasingly heated debate unfolding in Washington over the reform of the nation's 33-year-old chemical safety law that, in a recent shift, even the chemical and plastics industries concede must be modernized.
As researchers have mapped more and more of the "human toxome" over the past decade, steadily expanding the list of pollutants found in people, the techniques of human biomonitoring have been embraced by scientists, public health experts and key Congressional and legislative leaders as the logical foundation for revolutionary changes in the way government scientists test and regulate industrial chemicals.
For instance, the principal chemical reform proposal on Capitol Hill would prioritize safety assessments of tens of thousands of hazardous substances currently in commerce by focusing first on those that show up in people. Production or uses of toxic chemicals found in umbilical cord blood would be phased out unless rigorous, expedited testing showed them to be safe for developing infants.
Environmental Justice Leaders
A sociology professor and New Orleans native, joined the environmental justice movement after a visit to "Cancer Alley," as many residents call the Lower Mississippi River Industrial Corridor, an 85-mile stretch of oil refineries and petrochemical plants between New Orleans and Baton Rouge.
In her forthcoming book, Race, Place, and Environmental Justice After Hurricane Katrina, with Robert D. Bullard, director of the Environmental Justice Resource Center at Clark Atlanta University and author of the ground-breaking 2000 book Dumping in Dixie, Dr. Wright recalls that whites had fled the area, but poor African Americans remained behind, living in the shadow of the petrochemical plants. She blames "a pattern of discrimination and exclusion based on a culture of segregation and racism that allowed these polluting facilities and local government to respond to the needs of white citizens while ignoring the needs of black citizens."
To redress the balance, Wright helped found New Orleans-based Deep South Center for Environmental Justice (DSCEJ) in 1992.
After Hurricane Katrina devastated her own neighborhood, Dr. Wright wrote In the Wake of the Storm, published in 2006. Her book, Toxic Wastes and Race at Twenty was published in 2007.
- Tested positive for 36 to 391 of 75 chemicals.
- Perfluorchemicals (PFCs) - 87th percentile. Higher than all but 13 percent of Americans tested2 for perfluorochemicals (PFCS), used in non-stick coatings such as Teflon, water and stain-resistant textiles coatings such as Goretex and Stainmaster and grease-resistant food packaging.
- Mercury - 87th percentile. Higher than all but 13 percent of Americans tested3 for mercury, a potent neurotoxin especially dangerous to the developing fetus and infants.
- Also found - Bisphenol A (BPA), a plastics chemical and synthetic estrogen found to disrupt the endocrine system; perchlorate, a rocket fuel ingredient and common tap water and food contaminant; lead, a neurotoxic heavy metal found in older homes and tap water; polycyclic musks, synthetic fragrances associated with hormone disruption in animal studies; polybrominated diphenyl ethers (PBDEs), flame retardants found in foam furniture, computers, and televisions.
Policy analyst at the 65,000-acre Oneida Nation near Green Bay, Wisconsin,and a senior fellow with the non-profit Environmental Leadership Program, signs her emails wahnislateni ne yohantsya^teni -- every day is earth day.
She has spent 12 years overseeing cleanup of the reservation's fishing streams and ground waters and working to restore the tribe's traditional fishery. That mission says Hill-Kelley, is"near to my heart" because back in 1822, the Oneida people moved to the area around sparkling Duck Creek near Lake Michigan's Green Bay.
Today, Duck Creek, which bisects the reservation, is polluted with polychlorinated biphenyls (PCBs), industrial chemicals that local paper mills dumped by the ton into nearby Fox River and the bay itself. In 1977, the U.S. government banned the manufacture of PCBs, by that time classified as known human carcinogens, but the waters in and around Green Bay remain contaminated by PCBs and other toxins such as mercury, dioxins and the pesticide DDT.
"The fishing tradition is integral to our culture," says Hill-Kelley, "We want to be able to fish again on the reservation."
Some tribe members defy the no-fishing warnings, but Hill-Kelley, a former environmental laboratory technician, says she"realized very early that fish were a source of chemical contamination for myself and my daughter. I love to eat fish but limit myself and my family to occasional meals of fresh wild Alaskan salmon (store bought, of course) or Lake Michigan salmon. I don't eat tuna at all."
While the PCB cleanup grinds along, Hill-Kelley works with state and federal authorities and local farmers to reduce agricultural runoff and control sedimentation from urbanization.
Oneida's work is paying off. Today, Oneida Nation's streams are clearer, wetlands have been restored, and birds, trout and other wildlife are returning.
- Tested positive for 37 to 401 of 75 chemicals chemicals.
- Bisphenol A (BPA) - 81st percentile. Higher than all but 19 percent of Americans tested4.
- Mercury - 68th percentile. Higher than all but 32 percent of Americans tested3.
- Polycyclic musks - 91st percentile. Higher than all but 9 percent of Americans tested5.
- Also found - Perchlorate, lead, perfluorochemicals (PFCS), polybrominated diphenyl ethers (PBDEs).
Returned home to Corpus Christi to spend time with her sister, Diana Bazan, as she was dying of cancer at the age of 42.
At Diane's funeral, Ms. Canales was struck by the number of family friends who confided that they, or family members, were waging their own battles with cancer. She began to wonder if her sister's cancer, her other two sisters' hysterectomies and other diseases in the community might have something to do with the old dump in the predominately Hispanic neighborhood where she had grown up. Ms. Canales' health was fine - but was that because, as a Navy wife, she had been away from the neighborhood for two decades?
Canales and other family members founded Citizens for Environmental Justice and began investigating the dump and, eventually, all of Corpus Christi. Because of its key role in the energy industry - it is home to six oil refineries, several chemical plants and a network of related oil and gas businesses -- Corpus holds coveted slot on Forbes Magazine's list of 20 best U.S. cities for jobs [http://www.dallasnews.com/sharedcontent/dws/bus/stories/DN-topcities_21eco.ART.State.Edition1.4a8e8ca.html].
But Canales has become convinced that this prosperity has come at a human price.
"We found lots of clusters of cancer around the city," she says. Her organization instigated a study that found the city's birth defect rate to be 84 percent higher than in other parts of Texas.
Two of the Corpus Christi children born with birth defects were Canales' own grandsons.
Canales and her organization have engaged in a long-running battle to prevent the Citgo Petroleum Corp. from expanding its refining operations in Corpus Christi. So far, they have held the giant oil company at bay. The organization is currently in a legal battle with Flint Hills refinery over an air permit that would increase emissions into the surrounding community.
Ms. Canales is the recipient of the Congressional Hispanic Caucus Institute Award for Outstanding Achievements in Environmental Justice.
- Tested positive for 26 to 291 of 75 chemicals.
- Bisphenol A - 85th percentile. Higher than all but 15 percent of Americans tested.4
- Polycyclic musks - 86th percentile. Higher than all but 14 percent of people tested5.
- Also found - Perchlorate, lead, perfluorochemicals (PFCS), polybrominated diphenyl ethers (PBDEs), mercury.
Moved into Corpus Christi's Hillcrest neighborhood in 1962. It was a comfortable, largely African American working class community with spacious houses and big yards. She hardly noticed the Citgo oil refinery across the chain link fence from Hillcrest.
Until 2001, when Salone says she was awakened by a strong chemical stench. When the smell didn't subside, she joined Citizens for Environmental Justice. Salone chaired a biomonitoring study, with the Texas A&M School of Rural Public Health, that found that Hillcrest residents had elevated blood and urine concentrations of benzene, a chemical associated with oil drilling and refining and listed by the U.S. government as a known human carcinogen. Of special concern, the study found, were high levels of volatile organic compounds, chemicals generated by the petrochemical industry, in the urine of neighborhood children.
Salone wondered if her own bout with breast cancer might have been triggered by refinery emissions. Given the current state of medical science, there was no certain answer.
Even so, Salone was determined to take action.
In 2006, Citgo was indicted in federal court for criminal violations of the federal Clean Air Act. Salone became a key witness for the U.S. Justice Department.
In July 2007, the company was found guilty of two counts of operating open equipment without emissions controls required by law, the first time a U.S. oil refiner had been convicted in a federal criminal trial for Clean Air Act violations. (In 2001, Koch Petroleum Group, operator of another refinery near the Hillcrest neighborhood, had pleaded guilty to covering up environmental violations by disconnecting a device that controlled benzene emissions. It was ordered to pay $20 million in fines and support for environmental projects.
Salone was disappointed that Citgo was acquitted on two other counts of emitting benzene.
"They should have been convicted on all the charges," says Salone. "But it was great to let the refineries know that they weren't above the law."
- Tested positive for for 40 to 451 of 75 chemicals chemicals.
- Bisphenol A - 77th percentile. Higher than all but 23 percent of Americans tested4.
- Perfluorochemicals (PFCs) - 80th percentile. Higher than all but 20 percent of Americans tested2.
- Polybrominated diphenyl ethers (PBDEs) - 89th percentile. Higher than all but 11 percent of Americans tested2.
- Also found - Perchlorate, lead, mercury, polycyclic musks.
Chang has spent a dozen years as a community organizer specializing in environmental justice issues, until recently as executive director of the Asian Pacific Environmental Network (APEN), based in Oakland, California.
Her work, she says, is her way of honoring her grandmother, a onetime garment worker in Taiwan who immigrated to California in 1965. "I know what it took for my grandmother and parents to put food on the table," she says. Chang's parents earned professional degrees. Chang earned a Masters degree in urban planning from the University of California-Los Angeles
Like other immigrants starting at the bottom of the economic ladder, Chang says Asians struggle with low-paying jobs, dilapidated housing in polluted neighborhoods - and more. "Distinct to Asians," she says, "is their isolation and invisibility to the regulatory agencies. Many lack the ability to pick up the phone to report something because nobody speaks their language. They may not know their rights."
Her first major success: organizing the large Laotian community in Richmond, California, to confront environmental problems caused by the local Chevron oil refinery.
"We're one of the richest nations in the world," says Chang. "There's no reason why people can't have decent housing that's not contaminated with lead paint or built on top of a Superfund site, no reason why people can't have clean drinking water."
- Tested positive for for 40 to 451 of 75 chemicals chemicals.
- Mercury - 91st percentile. Higher than all but 9 percent of Americans tested3.
- Polycyclic musks - 84th percentile. Higher than all but 16 percent of people tested5.
- Polybrominated diphenyl ethers (PBDEs) - 86th percentile. Higher than all but 14 percent of Americans tested2.
- Also found - Perchlorate, bisphenol A (BPA), lead, perfluorochemicals (PFCS).
1 Numbers are expressed as a range because several PBDEs are tested in groups of 2,3, or 4 chemicals; a positive result may mean one, some, or all of the chemicals are present. The minimum and maximum number of possible positive results are provided in the range.
2 Perfluorchemicals (PFCs) results are compared to a national biomonitoring survey in which U.S. Centers for Disease Control tested 3,959 Americans.
3 Mercury results are compared to a national biomonitoring survey in which U.S. Centers for Disease Control tested 8,373 Americans.
4 Bisphenol A results are compared to a national biomonitoring survey in which U.S. Centers for Disease Control tested 2,612 Americans.
5 Polycyclic musk results are compared with an EWG biomonitoring survey in which 42 Americans were tested. CDC has not tested Americans for synthetic musks
Sample acquisition: Environmental Working Group (EWG) scientists obtained Institutional Review Board (IRB) approval for study design and protocols from Chesapeake Research Review, Inc (CRRI) in September, 2007. Each participant submitted a signed Informed Consent document to EWG verifying their understanding of the scope and purpose of the study, and what was required of participants.
Local phlebotomy laboratories collected whole blood and serum samples from each of the 5 participants on a single day in July or August 2008, depending on the participant. Each participant independently collected a first-morning urine void on the day of their blood draw. The samples were frozen upon collection at -20 degrees Celsius and shipped from the phlebotomy labs where they to AXYS Analytical Services (Sydney, British Columbia). All samples were stored at AXYS at 4 degrees Celsius.
Sample preparation: After all samples were received at AXYS, portions of each sample were sent to secondary laboratories for analysis (musks, perchlorate, bisphenol A, and metals). Samples were stored at -20 degrees Celsius prior to extraction and analysis.
Analysis of PBDEs: Samples were analyzed in one batch alongside QC samples including three procedural blanks and a spiked matrix sample. Analysis was performed by AXYS Method MLA-033 Analytical Method for the Determination of Polybrominated Diphenylethers by High Resolution GC/MS. Samples were spiked with the isotopically labeled surrogate standards prior to liquid-liquid extraction with ethanol, saturated ammonium sulphate and hexane mixture. Extracts were cleaned up using AXYS method MLA-033 for Polybrominated diphenylether analysis using an automated fluid management system (FMS) with silica, acid/base layered silica and alumina columns. In addition, samples were given a manual alumina finishing column. Gravimetric lipid analysis was conducted using an aliquot of the extract prior to clean up and analysis and surrogated were corrected accordingly.
The final extracts were spiked with isotopically labeled recovery (internal) standards prior to instrumental analysis. Analysis was performed on a high-resolution mass spectrometer (HRMS) coupled to a high-resolution gas chromatograph (HRGC) equipped with a DB-5HT chromatography column (30 m, 0.25 mm i.d., 0.10 um film thickness). The method was carried out in accordance with the protocols described in EPA method 1614 with some additional internal AXYS guidelines applied.
Calculations for PBDEs: Target concentrations for each analysis were determined by isotope dilution or internal standard quantification procedures using Micromass OPUSQUAN software. Sample specific detection limits (DL’s) were determined from the analysis data by converting three times the height of the average noise signal to a response, using the area/height ratio of the labeled standard, and then to a concentration following the same procedures used to convert target peak responses to concentrations. If the OPUSquan or MassLynx software selected an unrepresentative area for the detection limit calculation, the data interpretation chemist or the QA chemist made corrections.
Analysis of Perfluorinated Chemicals (PFC) and Tetrabromo Bisphenol-A (TBBPA)- AXYS method MLA-049/042 and AXYS Method 4226: Blood samples were spiked with isotopically labeled PFC and TBBPA surrogate, extracted in acetonitrile, cleaned up on SPE cartridges, split into two portions (1) PFC and (2) for TBBP-A and submitted for separate instrumental analysis runs. Samples were analyzed by liquid chromatography/mass spectrometry (LC-MS/MS). Analysis of sample extracts for perfluorinated organics was performed on a high performance liquid chromatograph column (Agilent Zorbax XDB Reverse phase C18, 7.5cm, 2.1mm i.d., 3.5 um particle size (or equivalent) coupled with a triple quadrupole mass spectrometer, running MassLynx v.4.0 software. Final sample concentrations were determined by isotope dilution/internal standard quantification against matrix calibration standards carried through the analysis procedure alongside the samples.
Calculations for PFCs and TBBPA: Target concentrations for each analysis were determined by isotope dilution or internal standard quantification procedures using Micromass MassLynx software. Sample specific detection limits (DL’s) were determined from the analysis data by converting three times the height of the average noise signal to an area using the area/height ratio of the labeled standard, and then to a concentration following the same procedures used to convert target peak responses to concentrations. If the MassLynx software selected an unrepresentative area for the detection limit calculation, the data interpretation chemist or the QA chemist made corrections. Reporting limits were equal to the greater of the lowest calibration standard concentration equivalent or the sample specific detection limit (SDL).
Analysis of BPA- AXYS method MLA-059: Urine samples were spiked with isotopically labeled Bisphenol A and 4-methylumbelliferyl glucuronide solution and B-glucuronidase enzyme (for deconjugation of glucuronidated forms of the target analytes) and extracted and cleaned up using solid phase extraction (SPE). The final extracts were reconstituted with methanol and spiked with isotopically labeled recovery (internal) standards prior to instrumental analysis. Analysis of extracts was performed by LC-MS/MS on a high performance liquid chromatograph (HPLC) coupled with a triple quadrupole mass spectrometer in the Multiple Reaction Monitoring (MRM) mode using a Waters Xterra C18MS, 10.0 cm, 2.1mmi.d., 3.5 um particle with C18 opti-guard column.
Calculations for BPA: Target concentrations for each analysis were determined by isotope dilution or internal standard quantification procedures using Micromass MassLynx software. Sample specific detection limits (DL’s) were determined from the analysis data by converting three times the height of the average noise signal to an area using the area/height ratio of the labeled standard, and then to a concentration following the same procedures used to convert target peak responses to concentrations. If the MassLynx software selected an unrepresentative area for the detection limit calculation, the data interpretation chemist or the QA chemist made corrections. Reporting limits were equal to the greater of the lowest calibration standard concentration equivalent or the sample specific detection limit (SDL).
Analysis of musks: A serum sample is weight into a clean glass 60 ml vial. Methanol, 0.1 M HCI and a set of internal standards (one or more for each group of chemicals) is added to the sample. The sample is extracted three times with a hexane-diethyl ether mixture and centrifuged after each extraction to separate the organic phase. The combined extracts are washed with a 1% KCI-solution and dried with anhydrous sodium sulphate. The serum extract is concentrated to a small volume and purified using a florisil clean-up procedure. The purified extracts are concentrated to a small volume and an injection standard is added. The final extracts are ana lysed with gas chromatography coupled with mass spectrometry (GC/MS) in the selected ion monitoring mode (SIM). The identification of analytes is based on correct retention times and/or qualifier ion ratios, compared to an external standard. The quantification was based on an external standard analysed together with the samples. The recovery of added internal standards (musk xylene-d 15 and tonalide-d3) were used to determine the performance ofthe analysis, but not to correct the results of the target compounds. The results are expressed in ng/g matrix. The matrix is serum and blood.
Analysis of lead: Whole blood samples were diluted 50x with a 1% HNO3. Digests are analyzed using Inductively Coupled Plasma - Mass Spectrometry (ICP-MS) for the analysis of Lead (Pb). Results were blank corrected as per Brooks Rand SOPs for EPA 1638 Modified method.
Analysis of total mercury: All samples were prepared and analyzed in accordance with the Appendix to EPA Method 1631E. Blood samples were first digested with nitric acid/sulfuric acid (HNO3/H2SO4) and further oxidized with bromine monochloride (BrCl). All samples were analyzed with stannous chloride (SnCl2) reduction, gold amalgamation and cold vapor atomic fluorescence spectroscopy (CVAFS) using a BRL Model III CVAFS Mercury Analyzer. Summarized sample results were blank corrected as described in EPA Method 1631 E.
Analysis of methyl mercury: Blood samples were prepared by potassium hydroxide/methanol (KOH/MeOH) digestion followed by distillation. All samples were analyzed by aqueous phase ethylation, Tenax trap collection, gas chromatography separation, isothermal decomposition, and cold vapor atomic fluorescence spectrometry (CVAFS). The samples were analyzed by a modification of EPA Draft Method 1630, as detailed in the BRL SOP BR-0011. All results were blank corrected as described in the method.
Analysis of perchlorate: Urine samples were spiked to a final concentration of 5ppb with an isotopically labeled perchlorate internal standard 5mL of sample was added to 0.5g of Dowex HCR-W2 cation exchange resin that was pre-washed with methanol and water and vortexed for 60s. 0.5 mL of a 10% ammonium hydroxide solution was added to the resulting liquid and the sample was passed through 3cm x 1.5cm diameter column of basic activated alumina. The resulting liquid was passed through a 0.45μm syringe filter and placed into an autosampler vial for analysis. Samples were analyzed using IC-MS/MS using Dionex AS-16 (2mm x 250mm) column with AG-16 guard column. A Quantum Discovery Max ESI-MS with HESI probe was used in the MS/MS mode for quantitation.
QA/QC: All organic analyses were conducted in accordance with AXYS’ accredited QA/QC program including regular analysis of QC samples and participation in international inter-laboratory comparison programs. Each analysis batch included a procedural blank to demonstrate cleanliness and a spiked laboratory control sample to monitor precision and recovery. The sample results were reviewed and evaluated in relation to the QA/QC samples worked up at the same time. The sample surrogate standard recoveries and detection limits, procedural blank data and the laboratory control sample data were evaluated against method criteria to ensure acceptable data quality. The laboratory flagged some values for not meeting certain analytical criteria. These related to ion abundance ratios and the method calibration limit. We used these values, but note the data quality flags in the data section of our Human Toxome website.
Four background samples were analyzed for each of the contaminants studied. Background contamination was detected only in the PBDE family.
The following criteria were applied to account for background contamination:
- Analyses reported as detections in this study exceed the 95% confidence interval and the maximum value for the 4 background samples.
- Where applicable, reported concentrations are adjusted for background concentration. Reported concentrations are taken as the difference between the sample concentration reported by the laboratory and the highest reported concentration in laboratory blank samples.
The number of chemicals detected is reported as a range due to some co-eluting chemicals in the PBDE families. The minimum value in the range considers each co-eluting group of chemicals to represent a single chemical, while the maximum value in the range represents the case where both co-eluting chemicals are present.
Antignac JP, Cariou R, Zalko D, Berrebi A, Cravedi JP, Maume D, et al. 2009. Exposure assessment of French women and their newborn to brominated flame retardants: Determination of tri- to deca- polybromodiphenylethers (PBDE) in maternal adipose tissue, serum, breast milk and cord serum. Environ Pollut 157: 164-73.
CDC. 2009. National Report on Human Exposure to Environmental Chemicals. Available online at: http://www.cdc.gov/exposurereport/
de Boer J, Cofino WP. 2002. First world-wide interlaboratory study on polybrominated diphenylethers (PBDEs). Chemosphere 46(5): 625-33.
EWG. 2005.Body Burden: The Pollution in Newborns. Washington DC: Environmental Working Group. Available: https://www.ewg.org/reports/bodyburden2/
EWG. 2008. Fire Retardants in Toddlers and Their Mothers. Washington DC: Environmental Working Group. Available: https://www.ewg.org/reports/pbdesintoddlers
EWG. 2009 Human Toxome Project. Available online at: https://www.ewg.org/sites/humantoxome/
Fischer D, Hooper K, Athanasiadou M, Athanassiadis I, Bergman A. 2006. Children show highest levels of polybrominated diphenyl ethers in a California family of four: a case study. Environ Health Perspect 114(10): 1581-4.
Herbstman JB, Sjodin A, Apelberg BJ, Witter FR, Patterson DG, Halden RU, et al. 2007. Determinants of prenatal exposure to polychlorinated biphenyls (PCBs) and polybrominated diphenyl ethers (PBDEs) in an urban population. Environ Health Perspect 115(12): 1794-800.
Sjödin A, Wong L-Y, Jones RS, Park A, Zhang Y, Hodge C, et al. 2008. Serum Concentrations of Polybrominated Diphenyl Ethers (PBDEs) and Polybrominated Biphenyl (PBB) in the United States Population: 2003–2004. Environ Sci Technol 42(4): 1377-84.