Airtoxics Workshop - comments roger mclellan

The Health Effects Institute
"A Partnership of the U.S. Environmental Protection Agency and Industry"

Comments on Mobile Source Air Toxic Workshop
Washington, D.C., February 8, 2000

Roger O. McClellan, DVM, DABT, DABVT
1111 Cuatro Cerros SE, Albuquerque, NM 87123
Tel: 505-296-7083; Fax: 505-296-9573; e-mail: roger.o.mcclellan@worldnet.att.net

    These comments summarize my views as a participant in the Health Effects Institute sponsored Mobile Source Air Toxics Workshop held in Washington, DC on February 8, 2000. I applaud HEI for taking the initiative in organizing the workshop and the various participants for their presentations. In particular, I commend the state and local agency participants for their well-organized and coordinated presentations.

Overall Perspective:

    The genesis of the currently anticipated rule-making on mobile source Air Toxics resides with the Clean Air Act Amendments of 1990 (CAAA-1990) and the frustration of the public and the Congress in the U.S. EPA’s failure to move expeditiously in addressing the issue of potential human health risks of hazardous air pollutants or air toxics. Key elements of the CAAA-1990 were intended to "force" the agency to move ahead on these important issues – both developing new relevant information and using this information for rule-making.

    Recognizing that nearly a decade has passed since passage of the CAAA-1990, it is disappointing how little new information has been developed to fill critical gaps in our knowledge of the human health risks of air toxics arising from mobile and other sources. At every step in the link from emissions to ambient levels to personal exposure to estimates of human health risks, there are serious deficiencies in knowledge. These deficiencies preclude the development of any science-based estimates of human health risks of air toxics. As a result it is not clear what level of action, if any is required to reduce air toxics and minimize human health risks. The human health risks of air toxics could be significant -- and, thus, expenditure to reduce emissions of air toxics or precursor could be a "good buy" in improving human health. On the other hand, it is quite possible that the human health risks of air toxics are at "de minimus" levels when considered relative to other risk factors and expenditures to reduce air toxics emissions may be a "poor buy" in terms of improving human health.

    The present lack of knowledge for action is deplorable recognizing that the gaps have been identified for more than a decade. There is a critical need to create a strategy for identifying information needs, conducting research that will fill the knowledge gaps and using this new information to develop science-based estimates of human health impacts of air toxics from mobile and other sources and to place these risks in perspective relative to other risk factors.

Specific Needs

    There are major uncertainties in estimating emissions, ambient concentrations, personal exposures, and the human health risks of air toxics. The uncertainties were not adequately identified at this Workshop. To the contrary, many presentations gave estimates of risk characterized to three significant digits presented in a manner that would lead this listener to believe these estimates were actual measurements of risk. Not a single presenter placed their estimates of the health risks of air toxics in perspective relative to morbidity and mortality from all sources including other risk factors. This approach championing a "single risk factor" to the exclusion of a broader human health risk minimization orientation will almost certainly result in mis-allocation of societal resources.

    There is a critical need for an integrated approach to the air toxics human health issue within a linked framework that considers the following elements:

A.    emissions of air toxics and precursors. Considerable uncertainty exists with regard to emissions from on-road versus off-road versus other sources.

B.    ambient concentrations as a function of season, time of day and geographic location. For most geographic areas the data available on ambient concentrations is extremely meager. The efforts of this South Coast Air Quality Management District to acquire data for the Los Angeles Basin is applauded. Similar efforts are needed in many other locations. The failure of the U.S. EPA to more adequately integrate air toxics measurements into its recent initiative to measure PM_2.5 at multiple sites is deplorable. This is yet another example of the agency’s "silo" or "tunnel" vision approach to addressing issues. The agency remains "chained" to the regulatory lamp post in acquiring new information on ambient air quality rather than taking a broader orientation that would give priority to acquisition of information for science-based decision making.

C.    personal exposures: Information on personal exposure to air toxics is seriously limited. The agency’s continued use of CO as a surrogate for air toxics is a useful "screening" tool but may be flawed in giving realistic estimates of personal exposure to many air toxics.

D.    toxicity potency of air toxics: The estimates of toxic potency used for many of the air toxics are highly uncertain. In some cases, the agency and other presenters used quantitative estimates of potency even though the scientific community has clearly indicated that the current science will not support such estimates, for example, diesel exhaust particulate matter. It is disheartening how little progress has been made during the last decade in improving the potency estimates. The EPA has failed to provide leadership for developing science-based estimates and the science-based agencies (such as the National Institute of Environmental Health Science) and the scientific community at large has seemed to be indifferent to the issues.

E.    Shape of the exposure– response relationship: In every instance that an estimate of toxic potency was given, the values were based on linear models assuming that even one molecule of the air toxic has an associated health risk that can be calculated. Without question, some of the air toxics have non-linear exposure-response (threshold) relationships that should be considered in developing estimates of population risk for low levels of exposure. Examples are formaldehyde and acetaldehyde.

F.    Population at risk and population risk: The population sizes have generally been appropriately estimated. This is probably the most certain element of the total risk analysis. However, it does little good to know the size of the population if reliable estimates of personal exposure are not available. For the screening type analysis conducted to date it is probably not necessary to attempt to ascribe risks to specific sub-populations of varied sensitivity. Indeed, the information is not at hand to make such projections.

A serious deficiency in all the analyses, as noted earlier, is the failure to place the projected risks of air toxics in perspective relative to actuarial estimations of morbidity and mortality for cancer, respiratory disease and other health end-points from all causes. Taken in isolation, the highly uncertain estimates of the health impacts of air toxics appear large. When viewed in the context of total morbidity and mortality from all causes and the opportunity for reducing risks from other risk factors (cigarette smoking, diet, etc) one gains a very different view of the situation.

G.    risk targets: Additional policy debate is needed on the appropriate risk targets to be used for air toxics as well as other sources of risk. The targets of 1 in 10⁶,or 1 in 10⁵, or 1 in 10⁴ lifetime risk have no clear policy basis and have largely been formalized by court decisions. Viewed against a 1 in 4 lifetime risk for cancer mortality for the total population it is not clear that these targets should be used as bright line values as was done repeatedly at this workshop.

Summary
    In my professional judgement, the information presented at this workshop is not sufficient for regulatory decisions concerning human health risks of air toxics. The information presented gave highly uncertain estimates of human health risks attributable to air toxics although the uncertainty was not quantified. The risk estimates may have sufficient validity for use in "technology-forcing" decisions although even such use should be done with caution recalling that all actions are not without "down-side" risks. A recent example of such down-side risks of a well-intentioned action is the use of methyl tert-butyl ether to reduce the risks of mobile source air pollution. The information presented at this workshop and the dismal progress over the last decade in improving the information base emphasizes the need for some organization to take the leadership in creating a national strategy for conducting research that will reduce the uncertainties in estimating the human health risks of air toxics (hazardous air pollutants) from all sources. Perhaps the Congress will need to mandate such a program similar to what has been done for particulate matter and involve the National Academy of Sciences/ National Resource Council in an oversight role. The Health Effects Institute could also play a valuable role in identifying major uncertainties and creating a research strategy to reduce them. A targeted program is needed to develop relevant information on health risks, additional information that focuses only on characterizing hazard or understanding mechanisms of action at high levels of exposure will be of limited value.

Go back to the Agenda

The Health Effects Institute "A Partnership of the U.S. Environmental Protection Agency and Industry"

The Health Effects Institute
"A Partnership of the U.S. Environmental Protection Agency and Industry"