Republican lawmakers criticized the Environmental Protection Agency (EPA) yesterday for problems with the toxicological assessments it uses for evaluating chemical safety. "The program's credibility is threatened when it continually puts forth assessments that fail to address fundamental issues raised by reviewers," said Representative Paul Broun (R-GA), chair of the House science committee, at a hearing  on the topic. But democrats dismissed the attacks as motivated by a desire to delay regulations. On Tuesday, the agency announced plans to improve the process.
When EPA is deciding how to regulate a chemical, it turns to its Integrated Risk Information System (IRIS ). The database contains EPA scientists' assessments of the toxicity of some 540 chemicals. This information helps inform regulations, such as drinking water or air-quality standards. The assessments can be politically controversial, such as with formaldehyde, which is widely used by the chemical industry.
The issue with IRIS isn't entirely ideological. Over the past decade, committees convened by the National Academy of Sciences' (NAS') National Research Council have identified shortcomings of IRIS assessments. "Problems with clarity and transparency of the methods appear to be a repeating theme over the years," said Jonathan Samet, an epidemiologist at the University of Southern California in Los Angeles. The same issues were present in the draft assessment of formaldehyde, which a NAS committee reviewed  this April. The NAS panel, chaired by Samet, made general recommendations for improving IRIS draft assessments, such as including more complete discussion of methods and rationales for picking and weighing certain studies.
Paul Anastas, EPA's science adviser, described changes to the IRIS program, announced  earlier this week, to address the NAS panel's recommendations. "IRIS documents will be shorter, clearer, more concise, and transparent," he told the committee. In addition, the agency will add a peer-review process in the preliminary stages of development to get scientific input on major assessments, and it is creating a scientific advisory committee for IRIS. These improvements were lauded by all.
The political divide comes with the question of how much additional peer review—and how much scientific certainty—is needed before IRIS assessments are ready for use in making regulations. Calvin Dooley, head of the American Chemistry Council, an industry lobby in Washington, D.C., suggested to the committee that all IRIS assessments should be reviewed by NAS until the reforms are clearly working. But Rena Steinzor, president of the Center for Progressive Reform in Washington, D.C., balked at Dooley's proposal. "Repeated rounds of redundant peer review allows the chemical industry to slow progress to a crawl," she said.
The slow pace of IRIS assessments, although not a focus of the hearing or the NAS review, has been a long-standing complaint from public health advocates. The chronic delays led the U.S. Government Accountability Office in 2009 to declare  IRIS in need of broad transformation.
Since then, the agency has made a substantial amount of progress in accelerating IRIS assessments, according to information (pdf ) released by EPA on Tuesday. Most assessments now take 23 months or less; in December 2007, assessments had typically been under way for more than 5 years. In the past 2 years, EPA has completed 16 assessments, exceeding the number of those finished in the 4 years before.
The improvements have come in part from streamlining  that EPA implemented in 2009 and from added resources. In fiscal year 2010, the IRIS budget increased by 50% (now about $15 million) and the number of staff by 25%.