The big picture

• May 01, 2000

Cheryl is miserable as she drives her compact low emissions vehicle LEV through a residential area of Albuquerque, NM. She is distracted by her pollution-related respiratory illness, groggy from the medication it necessitates and disheartened by knowing that there will be no relief until the current air pollution episode ends. Ironically, the March air is clear, the temperature inversion and oxyfuel season is past, a breeze is blowing in from the pristine mesas to the west and the view of Mt. Taylor 80 miles away is unblemished. Finding herself behind one of the city's few remaining old diesel buses (most having been replaced by cleaner diesels), she hardly notices the black cloud streaming over her car as the stoplight changes. Other than observing it momentarily dims the blue sky above, she gives little thought to this pollution source.

The pollution source affecting Cheryl and thousands of her fellow Albuquerqueans has nothing to do with engines, smokestacks, refineries or any other sources that scientists, regulators, Congress, environmental activists and the popular media typically associate with air pollution. The culprit, pollen grains, are airborne particles larger than 10 micrometers (µm) in diameter (PM₁₀). Pollen protein has not been declared a "criteria" or "hazardous" air pollutant and trees are not an environmental pollution source targeted by the U.S. Environmental Protection Agency (EPA). After all, isn't pollen natural — part of our ecosystem and necessary for plant survival? Cheryl's greener acquaintances proudly accept the moniker "tree hugger" for their defense of nature's flora and even she staunchly defends an offending mulberry in her own front yard from her husband's pruning.

Some communities such as Albuquerque have instituted restrictions on the planting of certain trees, including mulberries. However, because trees last even longer than diesel engines, Cheryl and her fellow sufferers will gain little relief in their lifetimes. Despite a 90 percent drop in diesel emissions over the past 10 years, Cheryl wheezes, sneezes, yawns and drives on, resigned to her fate.

We attempt to understand and manage the relationship between air quality and health by focusing on one pollutant, pollutant class or source at a time, despite the fact that nobody breathes only one pollutant, or pollutants from one source, at a time. This approach is complicated by the facts that diverse pollutants can have similar health effects, such as aggravation of asthma, and concentrations of many pollutants increase and decrease together during episodes caused more by meteorology than by emissions changes. The more we learn, the less likely it seems that any health effect associated with air pollution is solely caused by one pollutant or source.

Missing pollutants

Our approach tends to ignore many air contaminants altogether as we focus on links between a few classes of human-made pollutants and health. "Natural" contaminants, including plant pollens in the Southwest, organic vapors from trees in the Southeast, windblown sea salt in coastal cities and resuspended allergenic animal and plant material everywhere are seldom invoked in air pollution debates. Neither are many of the less-studied human-made air contaminants and products of atmospheric reactions. Because epidemiological studies largely derive their "exposure" measures from environmental or occupational air monitoring studies performed for regulatory purposes, linking health effects to a limited range of pollutants and sources tends to be a self-fulfilling prophecy. Past laboratory studies also tended to focus on a limited range of popularized pollutants and sources since funding for these studies is largely driven by cyclical regulatory debates.

We have a very limited knowledge of how natural and human-made air contaminants may act together to impair health. Take our friend Cheryl, for example. Recent studies in the United States and Japan (such as Diaz-Sanchez, et al., Allergy 52:52-56, 1997) suggest the possibility that the bus on the street and the tree in her yard could both be connected to her dilemma. Organic compounds on diesel soot may amplify allergic reactions to pollen proteins. If this is so, similar combustion products from other sources are likely to have the same effect.

The truth is, we have only a sketchy understanding of how an ever-changing mixture of thousands of natural and human-made air contaminants may contribute to the many respiratory and cardiac disorders whose incidences are statistically correlated with indicators of air pollution. This issue has been dubbed the "pollution mixtures" problem — and is one that's been largely kept in the closet.

We have difficulty dealing with complex mixtures of air pollutants. Our single-pollutant focus has created a self-perpetuating knowledge gap that is becoming increasingly apparent. We tend to regulate — and thus debate, popularize and fund research for — one pollutant or source at a time in a "revolving door" manner.

For example, a successful proposal for research on environmental air pollution will focus on particulate matter or "PM." A decade ago, ozone was the key to funding. In view of the debates surrounding EPA's promulgation of the new fine PM (PM_2.5) standard in 1997, Congress boosted funding for PM research in EPA's budget and asked the National Academy of Sciences to help guide the agency's booming PM research program (see National Research Council, Research Priorities for Airborne Particulate Matter, National Academy Press, 1998). EPA shifted its internal and external air pollution research from other pollutants to particles, funded five new university-based PM research centers and began expanding its particle measurement network to meet both regulatory and research needs. Other agencies, such as the U.S. Department of Energy and the National Institute of Environmental Health Sciences, have also developed particle initiatives. PM meetings, workshops and symposia occur more rapidly than rese archers can produce new information.

Comprehensive but unrealistic

Make no mistake; most of the current research on particles is good research per se. Indeed, this is the most comprehensive, most broadly coordinated, most heavily planned and certainly the most heavily reviewed attack on a class of air pollutants in history. We have undoubtedly learned a great deal about the health risks from airborne particles. The quality of the research is not the problem. The problem is that we really shouldn't have a particle research program at all; we should have an air pollution research program. We must approach the air quality-health relationship in a manner that more realistically considers how the diverse air contaminants in outdoor and indoor environments act together with many other environmental and personal factors to influence health.

Although some would prefer to ignore it, the air pollution mixtures issue is moving to center stage. Statistical efforts to resolve particle-related risks are forcing scientists to acknowledge their modest ability to disentangle the contributions of various air pollutants to the adverse health outcomes seen in populations. Legal challenges to regulatory standards, regardless of origin or outcome, are also focusing attention on our modest ability to ascribe causality to specific pollutants and identify safe levels (American Trucking Associations, et al., vs. U.S. EPA, No. 97-1440 consolidated, U.S. Court of Appeals, District of Columbia, May 14, 1999).

Change is evident even in EPA's own deliberations. EPA's 1996 Criteria Document for PM, developed as part of the PM standard review (EPA/600/P-95/001aF), largely dealt with other air pollutants as "confounders" to be swept away with statistical brooms in order to isolate the effects of particles. In contrast, the first draft of the new PM Criteria Documentnot yet finalized (EPA 600/P-99/002a) has sections dealing explicitly with "co-pollutant" effects among particles and other air contaminants and with relationships between outdoor and indoor exposures. Some epidemiologists are now suggesting that effects formerly attributed largely or solely to PM might more realistically be attributed to clusters of primary and secondary pollutants resulting from combustion processes.

There is growing recognition that we need alternate approaches to studying and managing the health impacts of complex air contaminants, but alternatives are not straightforward. Scientists can't possibly conduct experiments on every combination of pollutants. Studies determining whether combinations of two or three pollutants have greater or less than additive effects can explore specific interactions, but fall far short of real-world complexity. More complex mixtures can be synthesized in the laboratory, but enormous effort would be needed to determine the roles of the full spectrum of pollutants and their interactions through simple addition-subtraction studies. More broadly-applicable approaches need to be developed, as opposed to performing more addition-subtraction studies. Our challenge lies more in thinking than in doing.

New research initiatives

Two new initiatives serve as examples of different approaches to the issue. Both are founded on the view held by many epidemiologists, statisticians and data modelers that more detailed exposure and health response data can disentangle the roles of different air pollutants and sources, and their combinations. The first example is EPA's new "Supersites" program (not to be confused with Superfund) under which research-grade air pollution measurements will be made in several locations that have different pollutant mixes and meteorology. This program will provide more detailed data on the composition and sources of urban air contaminants. If accompanied by adequate personal exposure and health outcome studies, this program should improve our knowledge of exposure-health relationships. However, the exposure and health studies were neither mandated nor funded by the program, nor was full site-to-site consistency in measurements required.

The second example, the National Environmental Respiratory Center, at the Lovelace Respiratory Research Institute in Albuquerque, is taking a laboratory-based approach to the same problem. This program aims to put statisticians and data modelers to the test by using laboratory studies to generate detailed data on pollutant composition vs. health response. Congress provided seed money for the Center in the last three EPA appropriations, with the intent that other stakeholders, including federal agencies, states, industry associations, individual companies, environmental and health advocacy groups and individuals provide the rest. Responsibility for the strategic guidance of the center has been placed in the hands of a diverse, blue-ribbon advisory panel that cuts across academia, government, industry and disciplinary bounds.

The center's research strategy ensures both short-term and long-term pay-off by combining practical and theoretical goals. Several complex atmospheres with different, but overlapping composition, will be studied using identical sets of health assays in order to create a data matrix of composition versus response. The real-world, source-based complex pollution atmospheres use largely, but not solely, from combustion sources. Identically-designed studies of diesel (both contemporary and outdated) and gasoline (both on-road and off-road) emissions, wood smoke (hardwood and softwood), cooking fumes (meat and vegetable), tobacco smoke and simulated power plant emissions will provide useful head-to-head-comparisons while the total matrix of composition vs. health data is being developed. The health responses evaluated will encompass airway irritation and inflammation, allergic responses and asthma, resistance to infection, lung and heart function and cancer potential.

Preliminary work is almost completed and the study of the first atmosphere, contemporary diesel emissions, will begin late this spring. The development of the data matrix is expected to take five years, with the first reports from the individual atmospheres appearing next year.

No doubt about it, the air pollution mixtures problem is coming into focus. The two research approaches described above will help move us forward, but additional approaches should be explored. Most, if indeed not all, of our air pollution research should be framed around the central issue of mixtures rather than focused on single pollutants or sources. However difficult it may be to shift our perspective, if we don't begin to do so, we could be stuck for a long time in our single pollutant-single source "do loop."

Health Effects Institute: Particulate Air Pollution and Daily Mortality Analysis of the Effects of Weather and Multiple Air Pollutants The Phase I.B Report of the Particle Epidemiology Evaluation Project — http://204.127.236.248/Pubs/peepib.htm

Click here to post comments about this topic, and read what others have to say.

This article originally appeared in the 05/01/2000 issue of Environmental Protection.