The case of the wildly varying degrees of toxicity in wildlife

A real doctor treats more than one species.

At least that's how one of veterinary medicine's favorite sayings goes—and there's a lot of truth to it, given the way veterinarians' versatility provides them advantages in many fields, including environmental science, wildlife health, and ecological toxicology. Few toxicologists other than veterinarians are taught about infectious diseases, body systems pathology, and clinical pathology in a differential diagnosis context.

Because the multiple permutations that occur in the real world cannot all be replicated efficiently in the laboratory, that comprehensive approach can be a real benefit. For example, the concentration at which chemicals are toxic not only to mammals, birds, fish, reptiles, and amphibians but also to invertebrates and native microbes is an important consideration in evaluating risks from the release of those chemicals into streams and other areas of nature. 

Dr. Anne Fairbrother, a veterinarian who worked for the Environmental Protection Agency for more than a decade, has used this knowledge in a large collaborative effort to develop ecological soil screening levels, known as Eco-SSLs. These provided a source of toxicity reference values, or TRVs, for various species and chemicals that were hoped would improve consistency and streamline the screening process for risk assessments at potentially contaminated sites. But that was nearly two decades ago, and no progress has been made toward developing more predictive and precise soil cleanup values for the amount of soil that actually needs to be cleaned up, resulting in highly variable results among ecological risk assessments conducted throughout the United States.

"Using the Eco-SSLs to drive cleanup often results in cleaning up places that don't need to be cleaned up," said Dr. Fairbrother, who has more than 30 years of experience in ecotoxicology, wildlife toxicology, assessment of contaminated sites, and regulatory science.

"Thinking about plants and animals, cleaning up (contaminated sites) is hugely disruptive. If you say, ‘Let's remove the soil or take out contaminated plants,' you change the habitat. Habitat modification is a major stressor for wildlife. That's how animals end up on the endangered species list," she said.

Pharmacology 101

Dr. Fairbrother has always been interested in wildlife. She was one of the first students to focus on nondomestic animals at the University of California-Davis School of Veterinary Medicine, where she also helped out in a virology research laboratory.

After receiving her DVM degree in 1980, she pursued her master's at the University of Wisconsin-Madison, where she studied under Dr. Thomas Yuill, now professor emeritus of pathobiological science, forest and wildlife ecology. For five years, they measured how stress factors interact with and have the potential to create problems for animal populations—such as how chemicals can cause immune system problems and increase disease susceptibility.

Dr. Fairbrother took a job soon afterward at the EPA's Western Ecology Division Laboratory in Corvallis, Oregon. She led the Ecotoxicology Branch for eight years and researched ways to measure biopesticides' effects on wildlife and developed methods for assessing immunosuppression in birds exposed to environmental chemicals. During that time, she and other researchers also started to develop approaches to determine ecological risks at Superfund sites. The Superfund—a federal program designed to fund the cleanup of sites contaminated with hazardous substances and pollutants—was established as part of the Comprehensive Environmental Response, Compensation, and Liability Act of 1980.

At Superfund and other potentially contaminated sites, risk assessors conduct an ecological risk assessment to determine which chemicals and how much of them are in the environment, the potential exposure of humans and animals to those chemicals, and whether the site is contaminated enough to spend money to clean it up.

When it comes to wildlife, risk assessors must determine what species could be impacted by chemicals at a site. Once that is known, a risk assessor measures what is in the environment to see if the animals have, in fact, been exposed, then compares the results with the amount of each chemical that is toxic in those species. The risk assessor must use a predictive TRV, combined with exposure assumptions such as an organism's body weight and food ingestion rate.

"Where we run into a problem is that we have lots of different animals, from shrews and mice to coyote and deer," Dr. Fairbrother said. "They have differences in physiology, they absorb different amounts, and their bodies have different ways of responding to chemicals—the way (a chemical) moves through the body, the degrading enzymes each animal has, and how they excrete it."

Yet, risk assessors commonly take a TRV for one species and assume other animals have the same degree of sensitivity, meaning they are overpredicting risk for some animals and underpredicting for others.

"At the very least, they have to acknowledge high levels of uncertainty. At best, they should try to find ways to appropriately extrapolate across species," Dr. Fairbrother said.

Taking notes

When Dr. Fairbrother returned to the EPA in 2002 as assistant director for the ecology laboratory in Corvallis, she wanted to do something about this issue.

So, she helped lead the EPA Office of Emergency and Remedial Response in a collaboration among federal, state, consulting, industry, and academic participants. The result was the derivation process for ecological soil screening levels. The idea was to have standard operating procedures that provide a transparent and repeatable method on a national scale.

"So if you say, ‘I'm sure this chemical'—let's say copper or PCBs (polychlorinated biphenyls)—‘is not going to cause problems,' you want to be really sure about that," Dr. Fairbrother said. "It could be if you had an amount above a certain level, it might cause minor problems or be a gray area, or if you had an amount at a level where it's very high, that for sure will cause problems. When screening, you want it to be below that gray area. We were able, through the Eco-SSLs, to standardize that and move people through the process cheaper and faster."

The EPA bases its TRVs on survival, growth, and reproduction endpoints; behaviors; biomarkers; and up or down regulation of genes.

TRVs are specific to each species that has been tested. The 2011 paper "Assessment of risks to ground-feeding songbirds from lead in the Coeur d'Alene Basin, Idaho, USA" in the journal Integrated Environmental Assessment and Management provides an example of a TRV for songbirds at a lead mining site in Utah.

Dr. Fairbrother said she's proud of what the working group accomplished, developing Eco-SSLs for 17 inorganic and four organic contaminants that are frequently found in soil at Superfund sites. At the same time, she is "terribly frustrated" at the lack of progress toward developing species-appropriate TRVs and soil cleanup values.

In 2007, she left the EPA, moved to Seattle, and rejoined the private sector. Dr. Fairbrother continues to perform risk assessments as a consultant. She and David Mayfield wrote the article "Efforts to standardize wildlife toxicity values remain unrealized," published in IEAM, describing the challenges that ecological risk assessors face when trying to employ toxicity values for wildlife.

They found less than half of risk assessments used TRVs developed through the Eco-SSLs.

"We are still doing it the same way we did it 20 years ago, even though there is so much more knowledge in the world to be brought to bear to do it accurately," Dr. Fairbrother said. "What the EPA is doing across different parts of the country, it is not using the same numbers for different sites. Each site has a different approach to cleanup values for wildlife (toxicity reference values)."

One reason for the variation is that some risk assessors use one TRV for the screening stage and another TRV for more detailed stages of the risk assessment process. Another reason is that the people who oversee the cleanup are generally not wildlife biologists or ecologists. "They are almost always engineers, and their primary focus at contaminated sites is human health. Maybe in the last year of the cleanup process they remember to do the ecological and wildlife part, and it just ends up being a paper exercise," Dr. Fairbrother said.

But it doesn't have to be this way, as a better model exists, she said. When looking at human health risks, assessors can refer to the EPA's Integrated Risk Information System. For each chemical, EPA staff does a thorough literature review of studies on each chemical. After that is compiled, a board of experts reviews it and comes to a consensus on the value for that chemical.

"The fact that we don't do a wildlife IRIS across the country wastes lots of money" and potentially endangers animals in those areas through unnecessary habitat destruction, Dr. Fairbrother said.