This story is embargoed for republication until July 5, 2021
The U.S. Environmental Protection Agency launched a review into the best-selling Seresto flea and tick collar this spring after media reports about injuries and deaths linked to the product spurred a congressional inquiry, class-action lawsuits and a formal complaint.
The collars, which release small amounts of pesticide onto the fur of cats and dogs for months at a time, were the subject of 75,000 EPA incident reports, including at least 1,698 pet deaths, since their introduction to the market nine years ago.
But any determination by the EPA about the product’s safety is likely to take years – up to a decade or more – based on the agency’s history of foot dragging over complaints about other potentially dangerous pet products.
Since 2006, the EPA has received three public petitions asking it to ban pesticides in pet products over concerns about human health risks. One of them, filed earlier this year, involves Seresto and is ongoing. The other two took an average of 10 years for the EPA to resolve, even as some of the products it ultimately deemed dangerous remained on the shelves.
One of those other cases in particular bears a striking similarity to Seresto.
More than a decade ago, nonprofit Natural Resources Defense Council petitioned the EPA to ban the use of a pesticide called tetrachlorvinphos in pet products like flea and tick collars. The organization cited studies showing that the chemical, a possible carcinogen, had been linked to brain and nerve damage in children.
What followed was 12 years of delayed action, lawsuits, bureaucratic red tape and a new EPA assessment of the pesticide that confirmed some of the nonprofit’s concerns about its use in pet collars. Despite that, tetrachlorvinphos, also called TCVP, is still used in some flea and tick collars today.
The long battle over TCVP shows how hard it can be to remove a chemical from pet products, even when governmental scientists, experts and even federal appeals judges believe it’s harmful to children.
“In sum, the EPA’s years-long delay on this critical matter of public health has been nothing short of egregious,” wrote Ronald M. Gould, a Ninth Circuit Court of Appeals judge in an April 2020 decision mandating a response by the EPA to the Natural Resources Defense Council.
“For more than a decade,” Gould wrote, “the EPA has frustrated NRDC’s ability to seek judicial review by withholding final agency action, all the while endangering the wellbeing of millions of children and ignoring its ‘core mission’ of ‘protecting human health and the environment.’”
In the end, the EPA canceled some uses of TCVP in pet products but allowed others to continue to be sold.
EPA spokesman Ken Labbe said in an email the agency has found pet products containing TCVP, including the popular Hartz UltraGuard collar, to meet registration requirements under federal pesticide laws. In other words, he said, the risk level is acceptable.
Hartz defended the safety of its products. In an emailed statement, the company said its flea and tick collars present no risk to humans or pets when used as directed.
Also defending its products is Elanco, which sells the Seresto flea and tick collars. The nonprofit Center for Biological Diversity petitioned the EPA in April to ban Seresto collars because of the risks associated with its two pesticides – imidacloprid and flumethrin.
Seresto collars were the subject of more than 75,000 reported incidents of pet harm and nearly 1,000 incidents of human harm in the past nine years. Elanco said in previous statements that a vast majority of reported incidents involve minor skin irritations at the site of the collar. The company also said that it investigated each report of pet death and found no link between the collars and the death.
Elanco said its collars are safe and that the overall incident rate is low at 0.3%, meaning one in 300 pets has a reported issue. There is no medical or scientific basis to discontinue their use, the company said.
But the EPA’s pesticide office generally assumes a chemical is safe until it is proven otherwise, rather than taking a precautionary approach, said Dr. Philip Landrigan, a pediatrician, epidemiologist and founding director of the Global Health Initiative at Boston College.
“This is a systemic problem within EPA, and I hope the new administrator confronts it,” Landrigan said. “In my mind as a pediatrician, there's every reason to think tetrachlorvinphos on dog collars is dangerous to children.”
From nerve gas to pet collars
TCVP belongs to a family of chemicals called organophosphates, which were used as nerve agents during World War II. Among the most well-known is sarin gas.
After the war, U.S. companies began marketing the compounds as insecticides. They quickly became the most widely used class of such products available.
TCVP itself was first registered in the United States in 1966 and approved for killing and preventing insects on crops, as well as on livestock and pets in the form of powders, liquids and collars.
Just five years after TCVP was registered for use, Hartz bragged in a television commercial that it had already sold over 20 million of its popular flea and tick collars containing the chemical.
"It would take all of the dogs and cats registered in New York City, Los Angeles, Tampa, Boston, San Diego, Puerto Rico, Chicago and then some to wear all the Hartz 90 Day collars we've sold so far," the ad said. "Over 20 million Hartz 90-day collars. Now, look at your pet, shouldn't he be wearing the collar that conquered the flea?”
It wasn’t until a groundbreaking 1993 National Academy of Sciences study showing how pesticides harmed children that the EPA cracked down on organophosphates.
That study, led by Landrigan, prompted the passage of the Food Quality Protection Act in 1996 and, a few years later, resulted in the ban of nearly all organophosphates from in-home use.
Organophosphate pesticides are still used on crops, however, and they remain one of the leading causes of brain development issues in children in the U.S., according to a peer-reviewed 2020 study, which estimated that between 2001 and 2016 at least 111,000 cases of intellectual disability and 26 million IQ point losses.
The one exception to the EPA’s ban on indoor organophosphates: TCVP.
Years of delays
There are two main ways a pesticide can be banned: The EPA can cancel its registration after determining an unreasonable risk to human health or the environment; or the company that’s registered to use the pesticide can voluntarily relinquish it.
A court also can ban a pesticide, as the Ninth Circuit Court of Appeals did with dicamba in June 2020 after finding the EPA underestimated and ignored the risks it posed.
But that kind of court action is infrequent.
Under federal law, anyone can petition the EPA to ban a pesticide. And the agency must respond.
The nonprofit Natural Resources Defense Council did just that in 2009 when it asked the EPA to cancel the use of TCVP in pet products.
As part of its petition, it cited a 2008 EPA-funded study that found evidence of TCVP in the urine of children who had played with pets wearing the flea and tick collars.
It also cited its own 2009 study, which found that after three days of wearing a collar, at least half of the pets had enough TCVP residue on their fur to cause brain damage to children.
The EPA waited more than five years to respond – and only did so after the nonprofit sued for an answer.
Such delays have been common.
From 2004 to 2015, the EPA received 40 petitions to revoke, cancel or suspend pesticides used for a variety of reasons. Nine of those petitions resulted in unreasonable delay lawsuits because of the agency’s “lack of policies and procedures to manage public pesticide petitions in a transparent and efficient manner,” according to a 2015 Office of Inspector General report.
The Natural Resources Defense Council was behind seven of those lawsuits, including its February 2014 complaint to the D.C. Circuit Court asking it to force a response by the EPA on its petition to ban TCVP.
During that five-year delay, Hartz had continued to sell millions of its flea and tick collars throughout the United States, where the low-cost product dominates the market.
Collars containing other pesticides, like Seresto, sell for as much as $60. Hartz UltraGuard costs less than $5.
Although shampoos, topical spot treatments and oral medications represent more than 90% flea-and-tick prevention products sold, collars are still a big business with nearly $100 million in U.S. sales in 2016 alone, according to the EPA.
TCVP collars represented more than half of that share.
Seven months after getting sued, the EPA responded to the petition: It was denied. The agency said its risk assessment found the levels of TCVP in the products to be of below the threshold of concern.
Like many risk assessments conducted by the EPA, much of the information comes from the same companies that stand to profit from the continued use of the pesticides. Critics have long blasted the agency for this practice, saying it favors profits over safety.
A glaring error in the assessment
The EPA didn’t know it at the time, but its latest risk assessment contained a glaring error.
It had categorized the pesticide in the collars as a “liquid formulation” instead of a “dust/solid formulation,” which is more potent and harmful to human health.
TCVP collars, it turns out, contain both.
Hartz UltraGuard Flea & Tick Collar’s own label states: “As the collar begins to work, a fine white powder will appear on the surface.”
That prompted a new lawsuit in January 2015 arguing that the EPA unlawfully denied the petition based on inaccurate science. The EPA agreed and promised to conduct a new risk assessment.
More than seven years after the original petition was filed, in December 2016, the EPA released that new risk assessment.
It found that children are exposed to TCVP at such high levels in the collars that they are at risk of delayed mental development, lowered IQ scores and increased chances of autism and attention disorders in children.
The agency, which was then under the last days of President Barack Obama’s administration, pledged to take action within 90 days.
Instead, under President Donald Trump’s new administration, the EPA sent a letter to the Natural Resources Defense Council in March 2017 that kicked the can down the road. It said it would make a decision between July and September while also conducting a wide-ranging assessment of TCVP.
September 2017 came and went. So did all of 2018.
In May 2019, a decade after the first petition was submitted, the nonprofit asked the U.S. Ninth Circuit Court of Appeals to force the EPA to respond.
It took nearly a year for the court to act. When it did, a three-judge panel scolded the agency. The April 2020 opinion found that the “years-long delay” was “nothing short of egregious” and “endangering the wellbeing of millions of children.’”
"Repeatedly, the EPA has kicked the can down the road and betrayed its prior assurances of timely action, even as it has acknowledged that the pesticide poses widespread, serious risks to the neurodevelopmental health of children," the opinion stated.
The court ordered the EPA to respond within 90 days.
On the 90th day, in July, the EPA issued a new assessment finding that the risks from most collars to humans were acceptable.
The EPA did reach an agreement with Hartz that one of the seven collar types would be canceled and the rest would undergo minor design and label changes, which describes how to use the collars.
In September, the Natural Resources Defense Council again challenged the approval of TCVP.
In an executive order on his first day in office, President Joe Biden pledged to revamp the EPA to “limit exposure to dangerous chemicals and pesticides.”
But so far, the Biden administration has defended the Trump administration’s actions on TCVP.
“While EPA acknowledges that it took longer than anticipated to respond to NRDC’s cancelation petition, the Agency’s petition denial was informed by more precise data than had ever before been available to evaluate the risks of TCVP pet collar products,” the agency wrote in a recent court filing.
NRDC said this lack of a significant change was disappointing.
“There was an opportunity for the Biden administration to shift course here and put children's health first, and we're not seeing that yet,” said Miriam Rotkin-Ellman, a senior scientist at NRDC.
A cumbersome process
The EPA acknowledges that the process of trying to ban, or cancel, a pesticide is long and cumbersome and that it’s easier to work with the companies that sell it rather than outright cancel its use.
In the document explaining why it denied NRDC’s petition, the agency said any attempt to ban a pesticide triggers an administrative trial before a judge with documents and witnesses. Interested parties, like trade organizations, can seek to intervene alongside the pesticide registrant itself, which is often a multi-billion-dollar company. Together, they can keep the case tied up in court for years while the chemical stays on the market.
The last time the agency began cancellation proceedings was in 2013, but the EPA eventually entered into a voluntary cancellation agreement with product makers, according to the document.
“This process takes much less time and fewer resources,” the agency wrote, therefore the agency uses it much more frequently.
But even voluntary cancellations often take years.
For example, the banning of a pesticide called propoxur that also was used in pet collars took nearly seven years after NRDC filed a petition and three years after EPA scientists first expressed concern.
In March 2014, the agency and the companies using propoxur reached a voluntary agreement to stop selling pet collars containing the chemical because of its link to brain development issues in children and its likelihood to cause cancer.
Even so, the agency allowed the collars to be sold for two years after the cancellation agreement was reached.
As a result of the agency’s unwillingness to take action, the U.S. lags far behind many other countries in banning a wide array of pesticides. A 2019 study found that more than a quarter of all pesticides used in U.S. agriculture are banned in the European Union, and the U.S. also continues to use dozens of pesticides that are banned in China and Brazil.
EPA process ‘not consistent with public health’
When making a determination about whether a pesticide should be used, the EPA must make a cost-benefit analysis, weigh those costs and benefits and make a determination on whether the pesticide should be registered.
The agency is often given significant deference about what goes into that analysis and the final determination on how much weight to give each side, experts said.
For example, one of the TCVP’s closest relatives, chlorpyrifos, has also been the subject of years of court battles and agency delays. It was banned for indoor use in 2001 because of its risks to children but continued to be used as an agricultural pesticide.
The Ninth Circuit Court of Appeals recently ordered the EPA to ban or seriously restrict use of chlorpyrifos.
Landrigan said that chlorpyrifos and TCVP are so closely related that there is little doubt they both harm children. He said the EPA should make the pesticide industry prove that TCVP doesn’t cause harm, rather than make the public prove it does.
But the EPA instead takes a one-by-one approach to pesticides.
“It’s not consistent with public health and protecting America’s children,” Landrigan said.
According to EPA data, at least 163 people and 3,803 incidents have been reported involving TCVP collars, including 388 pet deaths. “However, it is possible that not all human incidents were reported,” Labbe, the EPA spokesman, said.
This is 20 times fewer overall incidents reported on Seresto, and four times fewer pet death incidents.
These incidents are almost certainly an underestimate, said Bettina Francis, an environmental toxicologist and emeritus professor at the University of Illinois, because of how difficult it is to tell when a child suffers from these effects.
Both Landrigan and Francis said they would apply the “precautionary principle” when it comes to pesticide assessments, but TCVP is too dangerous of a chemical to use when there are good alternatives.
“You shouldn't be exposing a child to a neurotoxin,” Francis said. “It’s insane.”
This story is a collaboration between USA TODAY and the Midwest Center for Investigative Reporting. The center is an independent, nonprofit newsroom based in Illinois offering investigative and enterprise coverage of agribusiness, Big Ag and related issues. Gannett funds a fellowship at the center for expanded coverage of agribusiness and its impact on communities.