Credibility Gap: Toxic Chemicals in Food Packaging

How Green is DuPont's Alternative?

June 9, 2008

Credibility Gap: Toxic Chemicals in Food Packaging: New scientific research shows PFC-related health effects in people

For decades, health surveys of workers at DuPont and 3M fluorochemical plants indicated that exposure to PFCs poses serious health dangers. In 1992, employee surveillance data at the DuPont Washington Works fluorochemical plant revealed a statistically significant excess of cancers of the buccal cavity and pharynx, kidney and other urinary cancers, and leukemia among the workers (DuPont 1992). The next year, a retrospective cohort mortality study was conducted by 3M at the Cottage Grove, Minnesota plant that produced PFOA, reviewing the employee records for the 1947-1983 period. The study found that occupational exposure to PFOA was associated with two-fold higher rate of death from prostate cancer (Gilliland 1993). And in 1998, Cottage Grove workers occupationally exposed to PFOA were found to have abnormal levels of reproductive hormones (higher levels of estradiol in workers with highest PFOA blood levels) (Olsen 1998). In addition DuPont has known since the 1980s that PFOA can cross the placenta and cause developmental abnormalities in children of women exposed to this toxic chemical at work (US EPA 2004; reviewed in EWG 2004). Despite this evidence of health problems in fluorochemical plant workers, industry did not take any precautionary action to protect public health. For decades, data on human health effects of PFOA was suppressed and not submitted to the EPA (US EPA 2004). This is especially egregious considering that animal studies have long linked PFCs with a striking and diverse array of health problems. This incredibly long list includes: a broad range of developmental effects, from smaller birth weight, developmental delays, and organ abnormalities, to stillborn pups and whole litter loss (Andersen 2008; Lau 2007; Lau, Butenhoff 2004); severe liver toxicity (Guruge 2006; Martin 2007; Rosen 2007; Yeung 2007); suppression of the immune system and predisposition to allergies (DeWitt 2008; Fairley 2007; Peden-Adams 2008; Yang 2002; Yang 2000); behavioral changes (Johansson 2008); altered hormonal function, especially thyroid and sex hormones (Lau 2007; Biegel 1995; Bookstaff 1990; Cook 1992; Liu 1996); as well as liver, pancreatic, testicular, and mammary cancers (Sibinski 1987). Everything changed, however, when the studies by both industry and academic researchers revealed that PFOA, PFOS, and other PFCs had become widespread, global contaminants that polluted bodies of humans and wildlife world-wide (Houde 2006; Kannan 2002; Kannan 2004; Prevedouros 2006; Sinclair 2006). Now, not only occupationally exposed workers were at risk from PFCs (Olsen 2004; Joyce 2007), but every American (Calafat, Wong 2007). Especially worrisome, children, the most vulnerable population, appeared to have higher levels of PFCs in their bodies compared to adults (Emmett 2006; Olsen 2004). While the scientific evidence linking PFCs to a wide range of health effects was more than strong enough in 2006 for the EPA to elicit a phaseout of many of these compounds, new studies published in the last two years show even greater cause for concern. Of particular note are three epidemiological studies, all conducted by independent scientists looking at people exposed to PFCs through consumer products and/or through contaminated drinking water, and all showing that exposure to these chemicals may be particularly dangerous for the developing fetus and children. Previous studies from the US, Canada, Germany and Japan have shown that PFCs can cross the placenta and transfer from mother’s body to the fetus (Apelberg, Goldman 2007; Inoue 2004; Midasch 2007; Tittlemier 2004), and are also found in breast milk (Karrman 2007; Kuklenyik 2004; So 2006; Tao 2008; Volkel 2007). Since there have been numerous animal studies demonstrating developmental toxicity of PFCs (Lau 2007; Lau, Butenhoff 2004; Andersen 2008; Fenton 2007) and the links between early developmental problems and health consequences later in life (Lau and Rogers 2004; Needham 2008), the next obvious question was whether any of these same health effects might be seen in human populations. Two sets of researchers set out to answer this question. Unfortunately, both found that the answer was ‘yes.’ The first study, conducted by researchers from Johns Hopkins University and the Centers for Disease control, looked at PFOS and PFOA levels in 293 randomly collected samples of fetal cord blood from babies born in Baltimore, Maryland. The scientists found a statistically significant relationship between the levels of these two compounds and low birth weight and size – even though the blood levels of PFOS and PFOA were within the range found in general population (Apelberg, Witter 2007). These results were featured in the November 2007 issue of the prestigious journal Environmental Health Perspectives (EHP), published by the National Institute of Environmental Health Sciences. The second paper, published in the same issue of EHP, was an even larger study conducted by scientists from University of California, Los Angeles in conjunction with the Institute of Public Health at Aarhus University in Denmark (Fei 2007). The researchers enrolled a randomly selected group of 1,400 Danish women and followed them throughout their pregnancy and birth. They found that levels of PFOA in the mothers’ blood were correlated with their baby having a higher chance of being born with low birth weight. While the term “low birth weight” sounds relatively innocuous, it is a well known harbinger of more serious medical problems. For example, in 2001 a study published in the British Medical Journal (Matte 2001) found that there was a proportional relationship between birth weight and average IQ by age 7, with low birth weight babies scoring lower on IQ tests during development. Another study found a statistically significant link between low birth weight and adverse developmental effects including learning disabilities (Sauve 1998). And a 1999 study found much higher risks of mortality for infants born far underweight (Chye 1999). The most troubling aspect of these two new PFOA studies is that the negative effects of PFOA were seen at levels present in the general population. The participating mothers in these studies were exposed through simple, every-day activities: through contact with PFC-containing products, food, and food packaging that is coated with fluorochemicals. With research conducted by the Centers for Disease Control and chemical manufacturers alike showing that the blood of more than 98% of Americans are contaminated with PFCs (Calafat 2006; Calafat, Kuklenyik 2007; Calafat, Wong 2007; Olsen 2004), these two studies show that there is real reason to be concerned about the ubiquity of PFC exposure. The C8 Health Project: largest PFC study to date The third key epidemiological study would be important if for no other reason because of its sheer size. With 69,000 study subjects, and known as the C8 Health Project, it is by far the largest study ever of PFC health effects in people. But the story of how it came about and its striking findings may make it the most important PFC study to date. In 1981 two of seven children born to PFOA-exposed female workers in DuPont’s Washington Works plant chemical plant in Parkersburg, West Virginia were found to have birth defects involving the eye. But DuPont never told the EPA, even after it learned from 3M that PFOA had been linked to birth defects of the eye in studies of laboratory animals (reviewed in EWG 2004). Three years later, between March and June 1984, DuPont tested for, and found, PFOA in tap water taken from a store in Little Hocking, Ohio, not far from the Washington Works plant. Again, DuPont failed to inform U.S. EPA of the finding. Nor did it inform the Little Hocking Water Association of this finding until seventeen years later in 2001, when it was revealed during the course of establishing a consent order between DuPont and the West Virginia Department of Environmental Protection (reviewed in EWG 2004). These two events were just the beginning of a several decade long cover-up that eventually landed DuPont with a $16.5 million fine, by far the largest in EPA’s history (US EPA 2005). But due to the company’s negligence, unsafe chemical disposal practices, and decades of deception and cover-ups, tens of thousands of people living in Ohio and West Virginia communities near the DuPont plant have been exposed to contaminated drinking water. The C8 Health Project was created to determine what, if any, health effects these community members might be suffering from as the result of this PFC exposure. A number of worrisome trends emerged from the initial analysis of the C8 Health Project data first presented to the public in May of this year (Frisbee 2008; West Virginia University School of Medicine 2008):

  • Children in the study had higher median levels of PFOA in their blood. Among the 69,000 enrolled participants, serum PFOA levels ranged between 0.5 ppb and 22,412 ppb, with a median concentration of 28 ppb (West Virginia University School of Medicine 2008). Among the enrolled children under 10, serum PFOA concentrations ranged between 0.7 ppb and 2,070 ppb, with median concentration of 34 ppb.

  • Higher PFOA concentrations in study children were correlated with higher total cholesterol levels, predisposing these children to future weight problems and accompanying risks of heart disease as well as other illnesses. Similarly, in industry studies elevated cholesterol was one of the hallmark health findings observed in PFC-exposed workers (Olsen 2003; Sakr, Kreckmann 2007; Sakr, Leonard 2007). The problem of obesity in children has now become so severe so as be considered an epidemic. Several prior biomonitoring studies indicated that children tend to have higher serum levels of PFCs compared to adults (Emmett 2006; Olsen, Church 2004). A link between PFCs and elevated lipids thus presents an especial danger to children's health.

  • Higher PFOA levels in study participants were correlated with lower levels of serum immunoglobulin, the key protein that helps the body fight bacteria, viruses, and other pathogenic microorganisms. Similarly, EPA researchers reported PFOA-exposed mice had low immunoglobulin levels (DeWitt 2008). And in animal studies PFC exposure has been linked with death of immune cells and weakening of the body’s ability to protect itself from infection (DeWitt 2008; Peden-Adams 2008; Yang 2002; Yang 2000).

  • Higher PFOA concentration in study participants was also correlated with elevated levels of alanine transaminase (ALT) and aspartate transaminase (AST), two key enzymes used in clinical blood assays to detect liver problems. When the liver is damaged, hepatocytes (liver cells) leak these enzymes into the blood, where higher levels of ALT and AST are then detected. Similar to findings of the C8 project, workers occupationally exposed to PFOA have increased levels of ALT (Olsen and Zobel 2007) and AST (Sakr, Leonard 2007). In addition, serum concentrations of the liver-secreted C-reactive protein, an important element early defense system against infections, decreased with higher PFOA levels in the C8 Health Project cohort.

  • Finally, thyroid function was affected in PFOA-exposed cohort participants. There are two types of thyroid hormones easily measurable in the blood, thyroxine (T4) and triiodothyronine (T3). In the study, researchers analyzed free thyroxine index (FTI), which indicates how much thyroid hormone is free in the blood stream to work on the body. Unlike the T4 alone, FTI is not affected by estrogen levels and can thus be used to assess thyroid function in both genders. FTI is elevated in hyperthyroidism and depressed in hypothyroidism. In the study, FTI response followed an inverted U curve (higher at moderate-high PFOA levels and decreasing again at the highest PFOA levels). This trend is in agreement with worker studies that demonstrate negative association between PFOA serum concentration and free T4 and positive association between PFOA and T3 (Olsen and Zobel 2007).

In summary, C8 Health Project scientists concluded that this pilot analysis of C8 health data points to the association between PFOA and a wide range of adverse health effects including immune function, liver function, cholesterol (especially in children), and thyroid (Frisbee 2008). Results of studies of this size and complexity often take years to make their way into the scientific literature, and the researchers have cautioned that they are preliminary. The trends, however, are so clear and so consistent with previous worker and animal studies that they are deeply worrisome.