Improve chemical exposure standards.
Air quality (Analysis)
Tamura, Todd
Pub Date:
Name: Issues in Science and Technology Publisher: National Academy of Sciences Audience: Academic Format: Magazine/Journal Subject: Science and technology Copyright: COPYRIGHT 2011 National Academy of Sciences ISSN: 0748-5492
Date: Fall, 2011 Source Volume: 28 Source Issue: 1
Geographic Scope: United States Geographic Code: 1USA United States

Accession Number:
Full Text:
I certainly agree with Gwen Ottinger and Rachel Zurer ("Drowning in Data", Issues, Spring 2011) and the followup letter by Sarah A.Vogel (Issues, Summer 2011, p. 16) that it is a fundamental challenge to translate chemical concentration data into information that is meaningful to the public. However, both of these pieces include some incorrect information and miss key aspects of the issues.


First, Ottinger refers to various ambient air standards and the fact that there is "no consensus on what constitute [s] a safe or permissible level." A casual reader could get the impression that all of the health-effects research that has been done to date has basically been useless in terms of relating con-centrations to health effects, which would be misleading. Although there is some variability in the underlying technical information, much of the variability in standards is because standards have different purposes and/or are applied to different situations. For example, contrary to what Ottinger and Zurer say, the federal Clean Air Act does not set ambient air standards for volatile organic compounds (VOCs); however, Clean Air Act regulations do set stack concentration limits for VOCs, for some industrial sources. Not surprisingly, if you are comparing stack concentration limits to ambient concentration limits, you are going to expect orders of magnitude differences, but these differences are not due to a lack of scientific consensus. They are due in part to the fact that they are applied to different locations (inside an exhaust stack versus in the ambient air) and in part because they are not both based on what is "safe." For example, some industrial equipment standards are based on what is achievable with available control technologies, which may be more or less than what someone deems to be safe. In addition, some air standards are not standards for what are safe levels in the ambient air, but are conservative standards used for issuing air pollution permits. In other words, the standards are for purposes of comparing the worst-case effects of a facility to a person standing at the facility's fence line for an extended period of time.

Second, it needs to be recognized that while identifying safe levels can and should be based on scientific information, there is also some political judgment involved. Although there are some health effects with thresholds, others do not have clear thresholds. For example, the default assumption for cancer risk is that the only concentration that corresponds to zero risk is zero. In addition, there are questions about how to address scientific uncertainty, how to extrapolate animal data to humans and account for the most susceptible humans, and how to extrapolate data that were taken at very high doses in order to produce a measurable effect down at the low doses. Vogel states that "what is safe for a 180-pound healthy man is not safe for a newborn, but our safety standards for industrial chemicals, except for pesticides, treat all humans alike." But this is incorrect, because many of the health-based air standards are in fact designed to be protective of the most sensitive individuals and do take children explicitly into account (OSHA standards are one obvious exception, because they are applicable to workplace conditions experienced by adults). For cancer risk, some areas of the country like using a 1-in-a-million lifetime cancer risk benchmark, but there is absolutely no technical basis for this standard. If people are made aware of the fact that calculations based on conservatively derived risk factors of lifetime cancer risk associated with urban ambient air quality are typically a couple of orders of magnitude higher than 1-in-a-million (but in many cases have been decreasing over the last several decades), that people's exposures indoors and in their cars are in most cases significantly higher than if they were simply exposed to ambient air quality, and that the American Cancer Society's calculations of Americans' actual lifetime risk of contracting cancer is closer to 300,000 to 500,000 in a million, they tend to feel that the 1-in-a-million standard is very or overly protective; however, if you simply put a 1-in-a-million standard in front of somebody, say it is health-based, and show concentrations that are above or even only slightly below it, they are likely to be much more alarmed. Vo-gel's recommendation to obtain better information about real-life exposure scenarios is a good one because this information could be used to both establish the extent to which exposures are resulting from ambient air versus more localized exposures (not just industrial facilities, which appear to be the targets of the authors, but also situations such as poorly ventilated cooking, travel on busy streets, etc.) and help provide context to people as to what their current exposure levels are.

Third, there needs to be an understanding that epidemiology (the solution identified by Ottinger and Zurer and Vogel) can have significant limitations. Although it can show correlations, it does not show causality, and there is often a multitude of confounding correlating factors. In addition, it can effectively detect only correlations that are epidemics; for example, a 10% or more effect on the population, or at best maybe 1%. Although epidemiology may be useful for evaluating some of the highest exposures, many people would argue that regulatory standards should be set more stringently than what epidemiology is capable of de-tecting, as many of them currently are.

But perhaps most important, the fourth thing that needs to be recognized is that there are and always have been countless chemicals in the air at some concentration or another. We have always been and always will be, in Vogel's words, "silently exposed to chemicals," some of which are man-made and some of which are not, all of which are potentially dangerous at some level, and the number of variables that could be studied are endless. Therefore, there is a need for both prioritization (at a multimedia, comprehensive level) and science, by which I mean the organization of scientific information regarding health effects, and not just the existence and continued execution of scientific studies on individual chemicals or situations.

Todd Tamura Tamura Environmental, Inc. Petaluma, California
Gale Copyright:
Copyright 2011 Gale, Cengage Learning. All rights reserved.