If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Changes For Research At EPA

New science chief faces budget challenges and pushes for greater analysis of uncertainty

by Cheryl Hogue
May 15, 2006 | A version of this story appeared in Volume 84, Issue 20

Credit: Photo by Peter Cutts
Credit: Photo by Peter Cutts

The Environmental Protection Agency's nearly $600 million-per-year research program is small compared with the efforts at other federal science offices. Its work, however, can have big impacts on public health and the economy. Work in the agency's Office of Research & Development (ORD) ranges widely, from computational toxicology and homeland security investigations to research on ecosystems and what is likely the world's most widely used database on risks from chemicals. The program, however, faces a dwindling budget while it tries to meet greater demands for information and to engender greater sophistication in its science.

Nevertheless, George M. Gray, who took over the helm of ORD six months ago, nearly bubbles over with enthusiasm about his job.

He tells C&EN he is struck by the passion that EPA researchers bring to their jobs every day, "working hard and really committed to the agency's mission." Gray, who came from Harvard University, says, "It's different from an academic institution where this day-to-day sense of mission is a little more oriented toward your lab group. But here, it's an entire agency that feels like it has a common mission, a common goal that everybody believes in."

Gray has been both EPA assistant administrator for ORD and the science adviser to all of the agency's programs since November 2005. His predecessor at ORD, J. Paul Gilman, left the government for the Oak Ridge Center for Advanced Studies in November 2004, after serving in the Administration of President George W. Bush for two-and-a-half years.

Gray, who turns 43 this month, holds a doctorate in toxicology from the University of Rochester. For the 16 years before coming to EPA, Gray worked at the Harvard School of Public Health, serving his last two years as the executive director of the Harvard Center for Risk Analysis and the two years previous as its acting director. His research there examined the scientific basis for human risk assessment and its application to policy. Gray focused on trade-offs in risk managment.

Some environmental and public interest groups are skeptical about Gray because of his association with the Harvard Center. They criticize the center because, although it carries an Ivy League imprimatur, it gets large chunks of funding from industry. And business, says Linda Greer, director of the health program at the Natural Resources Defense Council, tends to support research that backs its own agenda.

"We have had a dim view of the work done at the Harvard Center for many years," Greer says. And after meeting with Gray in April, "I did feel his agenda was more aligned with the interests of private industry than the interests of the public," she says.

Robert Shull, director of regulatory policy at OMB Watch, says that through its support of the Harvard Center, industry has created "the academic and technical apparatus to attack the regulatory system"—and not just the rules that the federal government issues. Shull points to Harvard Center founding director John D. Graham, who served as President George W. Bush's regulatory gatekeeper at the Office of Management & Budget (OMB) from 2001 until earlier this year (C&EN, Aug. 19, 2002, page 28). Many environmental and progressive groups were critical of Graham, saying he made it more difficult for the government to issue environmental, health, and safety regulations. Gray, Shull says, is "an acolyte of Graham."

Credit: Photos by Peter Cutts
Credit: Photos by Peter Cutts

In contrast, Carol Henry, American Chemistry Council vice president for industry performance programs, says Gray comes to EPA "with a very good reputation" forged at the Harvard Center. Henry says Gray focuses firmly on science without getting lost in ancillary issues. This approach, she says, "will strengthen the foundation for decision-making" at EPA.

As head of ORD, Gray must contend with ever-tightening federal resources. President Bush's budget plan for fiscal 2007 would cut the office's budget by 6%, to $557 million. This reduction has raised concerns in Congress and among the agency's outside science advisers, among others.

Rep. Vernon J. Ehlers (R-Mich.), chairman of the House Science Subcommittee on Environment, Technology & Standards, is concerned not only with the proposed cuts for 2007 but also that ORD's budget has dropped 14% since 2004.

"This trend, together with the rapid growth in spending on homeland security research, which alone accounts for almost 12% of the [EPA] science budget, seems to be making it harder for ORD to continue producing the valuable scientific knowledge that the agency depends on for its regulatory actions," Ehlers said at a March 16 congressional hearing on the office's budget.

M. Granger Morgan, who chairs EPA's Science Advisory Board, is particularly worried about the shrinking funds for basic research at the agency. "Environmental issues are complex and often subtle. If EPA does not continue to invest in a significant amount of basic environmental science, we will likely find ourselves making costly regulatory mistakes in the future," Morgan, who heads Carnegie Mellon University's department of engineering and public policy, said at the March hearing.

Gray tells C&EN that he deeply appreciates the sentiment behind these comments. "That people respect and value the research that ORD does is fantastic. That kind of good will only comes from years and years of good works."

But Gray dismisses the concerns about funding. "It's very clear that we're able to do what we need to do. We have the resources we need to support the agency with science and technology information," he tells C&EN.

"When it comes down to setting the specific priorities, difficult decisions have to be made. We make them in a very thoughtful and careful way, making sure that we can—in fact, do—deliver the results the agency needs," Gray says.

"Priorities change over time. Sometimes that means there are things that we've been doing for a long time that may be mature, that have served their purposes, and we have to make difficult decisions," he says. "These are decisions that are carefully thought out and are part of our strategic thinking about making sure that ORD can provide that scientific and technical support to the agency.

"Efficiency is going to be the name of the game," he adds. "We're making great progress there."

Gray is looking ahead to new initiatives for EPA research. They include improving the agency's database on risks from chemicals, research on making municipal water and sewage infrastructure more sustainable, and nanotechnology.

One of his major goals is changing how EPA scientists present their estimations of health or environmental risk to policymakers. Gray wants them to provide formal, detailed explanations of uncertainty surrounding the risk numbers they provide, rather than offering a single number that is their best estimate.

Unlike in the early days of EPA, when the agency worked to stanch the flow of raw sewage into waterways, Gray says, "this agency, this country, and the world are facing a different kind of environmental question. We're facing more complicated questions, questions where sophisticated scientific techniques, analytic techniques, are necessary to help us."

He adds, "Weighing and balancing risks makes characterization of uncertainty important."

For instance, EPA now is being asked to examine the relative risk of different forms of energy, Gray says. "No form of energy is perfect," he explains, citing concerns ranging from "what wind power does to bats to the implications of biofuels, nuclear, and traditional fossil fuels."

Noting ORD's homeland security research on decontamination following a terrorist attack, Gray provides another example of balancing risks and the need for analyzing uncertainty.

"How clean do you have to get a hospital in order to have it ready for people who have been hurt, who need medical care? Where there are trade-offs, you don't want to put people in danger from some kind of contamination, but there's a real danger to health from not having this facility open," Gray says. "That's where quantitatively characterizing uncertainty, I think, helps."

Formally quantifying uncertainty will also help direct EPA research, he says.

"To really understand uncertainty and to quantify it the way that I'm advocating, you have to take the science apart into the pieces and try to understand how they fit together," Gray says. "One of the things that quantifying uncertainty lets you do is understand which of those pieces contribute most to your uncertainty. That helps you focus your research to understand where there are places where you can reduce that scientific uncertainty."

Sometimes, reducing uncertainty may be extremely difficult, especially when scientists are unable to run an experiment to generate more information, Gray acknowledges. "In other cases, it may be a matter of gathering some data, doing a line of research that can help us." And analytical tools can help determine which data are the most valuable to gather, he says.

Gray is focusing a great deal of attention on uncertainties presented in the Integrated Risk Information System (IRIS), an Internet database that provides EPA's expert judgments on how much exposure to a particular chemical is safe (C&EN, Sept. 25, 2000, page 34). EPA, other federal agencies, state and tribal environmental departments, and even regulators in foreign countries rely on IRIS, which was created in 1986 and has 544 entries. Industry, notably through ACC, has long supported efforts to keep information in the database up-to-date about potential health effects from exposure to chemicals.

"This is a database that other people really respect and want to use," Gray says. "It has become the worldwide standard." He adds, "Fully a third of the hits to our IRIS website are from outside this country."

The carefully considered scientific information in IRIS "is used in lots of different ways, from standard-setting to hazardous waste cleanups to homeland security sorts of decisions" as well as in academic research, he says. Gray adds that he drew on the database when he was in academe.

IRIS evolved as a tool for setting environmental and health standards, he says, and is oriented toward policymakers. Gray says he wants to make IRIS information functional for a variety of uses. "If we do a good job characterizing uncertainty quantitatively, we can support those other uses as well," he argues.

"Our scientists do a very good job of evaluating the science" when they prepare or update an IRIS entry, Gray says. "What I want to do is push us to incorporate the state-of-the-art analytic tools that are there to help us quantify that uncertainty." This, he says, will lead to better decisions and guide EPA's research agenda, as well as "help us just be more humble about our science, making sure we are recognizing this uncertainty."

But Greer of NRDC and others are skeptical about Gray's plan to expand uncertainty analysis and express risk as a range, especially in IRIS.

Providing a range as an estimate for risk "completely undermines the usefulness of the IRIS database," Greer says. Ranges of uncertainty, in some cases, could extend across orders of magnitude, she says. For instance, an IRIS entry expressed in a range could suggest that a cleanup standard for a chemical should fall somewhere between 5 and 5,000 ppm, she says.

Such ranges can befuddle regulators, Greer says. Regulators in other parts of EPA, in state pollution control programs, or even in other countries have less expertise in selecting an appropriate risk number, she says. Greer, who has a doctorate in environmental toxicology and has worked with risk estimates for years, says she too struggles with risk ranges.

The current "point estimates," the single numbers now found in IRIS entries, represent the best professional judgment of EPA managers who are intimately familiar with the risk data on a substance. These people are experts on the chemicals they study and have no stake in the regulatory outcome from an IRIS entry, Greer says.

By introducing a range of possible numbers and eliminating point estimates, EPA will have "taken the scientific expertise out of the IRIS entry," Greer says.

Ranges will also allow decisionmakers to hide economic considerations in decisions that are supposed to be risk-based, Greer charges. For instance, with a 5- to 5,000-ppm range, regulators can determine that a 5-ppm cleanup standard is prohibitively expensive. They may instead opt for a 500-ppm standard, which is equally valid given the range. They won't have to explain that economics swayed the decision, she says. Currently, regulators may make similar decisions but must explain clearly that their policy choices include cost considerations, she says.


Plus, risk ranges open the door to increased negotiations with industry over determining the best cleanup or pollution standard, as well as more legal challenges, she says. EPA attorneys "must be going nuts" because a regulatory decision based on a range of risk could be attacked in court as arbitrary regardless of the final standard the agency selects, she says.

But Gray says formally quantifying uncertainty will help guide policymakers as they make decisions, just as information on uncertainty guides financial markets. "We're just more uncertain about some things than others," he says.

Meanwhile, Gray sees two overarching challenges in his work at EPA. One is ensuring that science plays a key role in the agency's decisions. "Especially with an administrator who is a scientist, the agency's leadership is receptive to that," Gray says, noting that EPA Administrator Stephen L. Johnson is by training a toxicologist.

A second challenge, Gray says, is "helping scientists understand that they're not the only part of a decision. Decisions that are made in the agency depend on the legal constraints and on a variety of other competing issues.

"It's a recognition of the importance of science, what science can and cannot tell you in decisionmaking, but also helping scientists recognize and remember that a number of other factors legitimately go into decisions that are made," he says.

Even with this reality in the milieu of government policy, Gray's love for the world of science seeps out. With a grin he says, "Science changes. That's what makes it exciting and fun."


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.