5
The Challenges Ahead for Particulate Matter Research

INTRODUCTION

In this chapter, the committee identifies key, overarching scientific challenges for the years ahead in completing the research portfolio on particulate matter (PM). Chapter 6 offers the committee’s recommendations on strategies to most efficiently manage research on PM and address the key information gaps while meeting the scientific challenges laid out in this chapter.

The committee has identified seven overarching scientific challenges for the years ahead in air pollution research; these challenges need to be met for improving the scientific basis for regulation and public health protection and for cost-efficient control:

  • Completing the PM emissions inventory and PM air quality models necessary for NAAQS implementation and for informing health research.

  • Developing a systematic program to assess the toxicity of different components of the PM mixture.

  • Enhancing air quality monitoring for research.

  • Planning and implementing new studies of the effects of longterm exposure.

  • Improving the relevance of toxicological approaches.

  • Moving beyond PM to a multipollutant approach.

  • Integrating disciplines.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress 5 The Challenges Ahead for Particulate Matter Research INTRODUCTION In this chapter, the committee identifies key, overarching scientific challenges for the years ahead in completing the research portfolio on particulate matter (PM). Chapter 6 offers the committee’s recommendations on strategies to most efficiently manage research on PM and address the key information gaps while meeting the scientific challenges laid out in this chapter. The committee has identified seven overarching scientific challenges for the years ahead in air pollution research; these challenges need to be met for improving the scientific basis for regulation and public health protection and for cost-efficient control: Completing the PM emissions inventory and PM air quality models necessary for NAAQS implementation and for informing health research. Developing a systematic program to assess the toxicity of different components of the PM mixture. Enhancing air quality monitoring for research. Planning and implementing new studies of the effects of longterm exposure. Improving the relevance of toxicological approaches. Moving beyond PM to a multipollutant approach. Integrating disciplines.

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress The regulatory, public health, and scientific contexts are set by the comparatively low concentrations of PM now measured in most of the United States. At these concentrations, the effects found in observational studies are small in comparison to those found in the past, particularly effects observed during dramatic air pollution episodes, such as the 1952 London Fog. The current risks are also small in comparison with those associated with some causes of the same outcomes, for example, active smoking and risk for ischemic heart disease events. Consequently, epidemiologists face the challenge of detecting a relatively weak signal of effect and of ensuring that the effects found do not reflect factors other than air pollution, for example, meteorological conditions and personal use of tobacco products. The pollutant concentrations of interest also have implications for the design of toxicological experiments and interpretation of their findings. Although it is feasible for epidemiologists to investigate large populations to detect effects, toxicologists typically use exposure levels that are far higher than those experienced by the population to detect measurable effects. Exposures far higher than those experienced by the population generally might be needed to induce measurable effects. Consequently, there might be uncertainty as to the relevance of biological findings in model systems. Mechanisms operative at high concentrations might not be applicable at lower concentrations. Completing the Particulate Matter Emissions Inventory and Particulate Matter Air Quality Models Necessary for NAAQS Implementation and Informing Health Research Although the committee recognizes that its objective is to provide independent guidance for planning and monitoring a long-term PM research program, the committee has long acknowledged that this research program should also provide the tools necessary for the implementation of current and possible future PM NAAQSs. In particular, the committee views improved emissions characterization and air-quality model testing and development as critical for rapidly-approaching deadlines for state implementation plans (SIPs). As shown in Table 2-1 of Chapter 2, the states will be developing and submitting their implementation plans over the next 5 years. The committee’s second and third reports offered an agenda for research related to these topics. As described in more detail below, EPA should develop a comprehensive prioritized plan for systematically translat-

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress ing new source-test methods and activity data into a completed, comprehensive national emissions inventory. In addition, source-based air quality models earmarked for regulatory application will be operationally useful only to the extent that the data needed to support them are routinely available. Although EPA alone might not have adequate resources to fully evaluate these models, it can help to shape efforts involving other entities with substantial field monitoring programs to enhance the value of the resulting data for model evaluation. Although some positive steps have been taken by EPA toward further developing the emissions inventories and models needed for the SIPs, the committee sees a need for faster progress in meeting the critical scientific and technical challenges identified in this report. These inventories and models are needed not only for implementation but for subsequent development of control approaches that target those sources of particles linked to the greatest risk to health. Emission inventories are not only critical for air quality management but also for health research that is directed toward sources rather than toward PM generally. The broad base of need for a complete PM source inventory is not sufficiently appreciated in the committee’s judgment. Previous committee reports identified the need for more comprehensive measurements of particle-size distributions, chemical composition, and precursor gases from major stationary, area, mobile, and natural sources. The committee still considers these measurements to be needed in the near term. The measures need to be made with representation of ambient atmospheric conditions and of sources beyond on-road vehicles that contribute major fractions of ambient PM. Methods also need to be developed and applied to better quantify PM and precursor emission rates from in-use engines operating in nonroad environments. The committee is concerned that the challenge of developing methods for this purpose and making the measurements is neither sufficiently acknowledged nor addressed. There are two components to the development of emissions inventories. The first is emission-source characterization, which requires accurate measurements of the mass emission rates, composition, and size distribution from a representative sample of a particular source type. Because of the importance of secondary formation of PM, emission-source characterization also requires emission rates of reactive precursor gases (SO2, NOx, ammonia, and volatile and semivolatile organic compounds). To develop a robust and informative source characterization, accurate testing is needed for a sufficient number of units of a particular source category under the full range of possible operating conditions. The second component of emissions inventory development involves applying source-testing information (mass emission

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress rates, particle-size distributions, and chemical compositions) to estimates of the number of individual units and their use patterns to produce the emissions inventory (see Box 3-1 in Chapter 3). Research is also needed to gather information relevant to this second component. To control ambient PM concentrations through emissions reductions, the relationship between emissions and ambient concentrations needs to be characterized. An adequate characterization of this relationship is complicated by geographic variability in the natural emissions that contribute to background PM concentrations and in meteorological processes that affect atmospheric processes and also patterns of human exposures. In its second report, the committee addressed the need for research on these topics, with the anticipation that EPA could use monitoring data collected under the PM monitoring programs to greatly improve characterization of the relationships between emissions and ambient PM concentrations. The committee has previously commented that the three activities of emissions tracking, air quality modeling, and ambient monitoring are parallel, complementary and reinforcing. (See topic 4 in Chapter 3 and Appendix C for a discussion of needed improvements in air quality models.) DEVELOPING A SYSTEMATIC PROGRAM TO ASSESS THE TOXICITY OF DIFFERENT COMPONENTS OF THE PARTICULATE MATTER MIXTURE To answer the key questions concerning the hazardous components of PM (topic 5), a carefully coordinated, long-term multidisciplinary research effort will be required that goes well beyond the work now under way. Although substantial relevant research has been carried out on this topic, the committee’s review showed a collection of evidence with little convergence. The key “lesson learned” is, in fact, the need to reconsider the research strategy for addressing the assessment of hazardous PM components. This topic has proved particularly challenging because of the many aspects of particles that might plausibly determine toxicity and the strong possibility that different characteristics of particles could be relevant to different health outcomes. Additionally, in addressing the topic, coherent and converging evidence is needed from both toxicological and epidemiological research that addresses specific components and health outcomes in parallel. The barriers to implementing such needed integrated research are scientific, administrative, and perhaps cultural within the research community, and the costs will likely substantially exceed those originally estimated

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress for the topic of hazardous PM components by the committee. With regard to the scientific approach, models have been lacking for simultaneously pursuing hypotheses in the laboratory and in population contexts. Recently however, several approaches have been taken that might prove to be useful models. For example, the epidemiological studies in the Utah Valley were followed, albeit a decade later, by animal and human toxicological investigation using previously collected particles from the same area (Pope 1989, 1991). As another example, it is possible to use particle concentrators to expose animals for toxicological studies in the same locations where epidemiological studies are in progress. At present, investigators in Southern California are characterizing roadside particles in detail, exposing animals at roadsides in mobile chambers, and carrying out related epidemiological research. A similarly integrated program of research on ultrafine particles has also been implemented. An additional barrier is the need for population “laboratories” in which hypotheses related to toxicity of PM can be tested. In general, few research groups have either the financial or the technical resources to implement monitoring for assessment of exposure that will be sufficiently detailed to be the foundation for epidemiological research on risks to health and characteristics of PM. As discussed subsequently, the U.S. Environmental Protection Agency (EPA) needs to establish these monitoring sites; the limited success of the agency’s coordination of the Supersites Program to provide a sufficient platform for epidemiological research provides an example of the need for coordinated planning involving the health research and monitoring communities to develop sustained multicharacteristic monitoring necessary to inform future population studies. Fortunately, other funding agencies stepped forward and supported research at the Supersites in Fresno, California; Atlanta, Georgia; St. Louis, Missouri; and Baltimore, Maryland. It is also important to gather information about actual exposures of susceptible subpopulations and the general public to particle components of biological relevance. Finally, the scope of the research task, as described by a matrix with particle characteristics as one dimension and health outcomes as the other, calls for an integrated and more programmatic approach to assessing hazardous PM components. The committee’s prior reports addressed the need for integration and interaction in the conduct of its research agenda; the topic of hazardous PM components clearly needs greater integration and interaction than has occurred to date, and the committee strongly recommends that research on this topic should be managed in a more programmatic fashion. Although individual research groups have made advances on specific aspects of particles and health risks, given the heterogeneity of

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress approaches and the highly selective nature of the hypotheses pursued, the assessment of hazardous PM components will not be addressed with sufficient certainty without taking a new approach. This new approach is especially important because the research being conducted by different groups is being done without an overarching framework, and the various groups are using different techniques to identify the toxicity of specific components of the PM mixture. Research on hazardous PM components becomes even more challenging when placing the PM mixture in the context of other air pollution components, a scientific challenge identified in more detail below. A plan is needed that approaches the matrix of particle characteristics by health outcomes in an organized and tiered fashion, screening across the matrix with common approaches so that priorities can then be set for a second stage of more focused investigation. Beyond the scientific challenges of developing such an approach, a new management approach will also be necessary (see Chapter 6). ENHANCING AIR QUALITY MONITORING FOR RESEARCH Meeting the key research priorities identified in Chapters 3 and 4, especially identification of hazardous components of the PM mixture (topic 5) and the relative role of PM and other copollutants (topic 7), will require an air quality monitoring network designed with sufficient spatial, temporal, physical, and chemical detail to test atmospheric models and to best approximate population exposure rather than solely assessing compliance with the NAAQS. Progress in using epidemiological approaches to understand the toxicity associated with physical and chemical characteristics of PM will depend on the availability of modeling and monitoring data. Such data will be useful in examining relationships among personal exposures to particle components of biological relevance and corresponding ambient concentrations for susceptible subpopulations and the general public. Monitoring programs in place will provide data that should support exploration of some hypotheses. However, as noted in previous committee reports, these programs are often designed with little input from health and atmospheric scientists, and the programs often do not measure all the PM characteristics of interest and the gaseous pollutants. The gaseous pollutants contribute to the overall burden of pollution-associated morbidity and mortality, and data on their concentrations may be useful for health studies as well as for testing models. Moreover, the frequency of available measurements can be inadequate and weaken some analyses.

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress Some progress has been made in addressing these challenges, as EPA has implemented the new PM monitoring network. Most notably, the advent of the new PM speciation network, with every-3-day speciation at more than 50 locations, offers opportunities for new research approaches. Limitations can be identified in these initiatives, however; the most notable of which is the slow pace of implementing daily speciation monitoring at the 10 planned sites (at this writing, none of the sites is fully operational, and only five sites have even begun feasibility testing with collocated continuous sulfate, nitrate, and carbon measurements). Another limitation is the slow pace of developing and implementing monitoring programs for ultrafine particles and measurements of soluble metals and organic species. As some recent epidemiological studies suggest, short-term peak exposures (on the order of a few hours) over time and at different locations might be more highly associated with health responses than measures of average pollution concentrations. Thus, continuous measurements might be needed in some locations. In looking forward, the monitoring paradigm needs to shift increasingly from assessment of compliance with national standards toward serving multiple purposes, such as air quality forecasting, episode alerts, exposure characterization in populations at high risk, health studies, atmospheric process studies, evaluation of source zones of influence, and evaluation of long-term effectiveness of control strategies. This shift implies less use of federal reference methods (FRMs) at urban locations and greater use of in situ continuous monitors and compound-specific integrated samples at locations representing background, boundary, transport corridor, regional, urban, and neighborhood spatial distributions. Such an enhanced network should have the following characteristics: Use continuous measures of appropriate indicators with real time access. These measures could be used by local air quality authorities to issue advice and alerts to the public and support the application and improvement of forecasting methods that would permit better public health planning. Public access to these data would permit people to make informed decisions about activities that might affect their personal health. They would allow more precise relationships to be established between the timing of increased concentrations and specific health outcomes. These data would permit better relationships to be established between ambient concentrations and source contributions, thereby better focusing emission-reduction strategies. Represent less uniform micro- and middle-scale exposures. Many people are exposed to vehicle exhaust near heavily traveled roadways

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress for at least a portion of the day, and some potential health indicators (for example, black carbon and particle number count) are much lower when measured at a few tens of meters from roadways, as required by compliance siting criteria. A dearth of regional-scale measurement locations also limits the ability to determine the effect of transport from other areas on many air quality planning regions. Encourage the completion of development of continuous monitors for indicators other than mass concentrations. Much progress has been made in developing continuous instruments for specific chemical components and size distribution at the research level rather than the applications level. A sufficient number of locations requiring these measurements are needed to support an economical instrument production volume and to determine their utility and applicability. A carefully designed network serving multiple purposes could be implemented by broadening the design of the compliance monitoring work to meet the needs of regulation and research alike. In fact, for both purposes, the nation’s monitoring network should be useful for estimating population exposures. It would stimulate new methods for relating ambient concentrations to health outcomes, better estimate human exposure, refine the understanding of particles in the atmosphere, and allow people to make their own decisions about where and when they will perform certain activities that might adversely affect their health. Several aspects of a new national monitoring strategy are being developed by the EPA’s Office of Air Quality Planning and Standards through its proposed national monitoring strategy (known as NCore). This effort is focused on a broad range of pollutants and not only PM and has the potential to move the monitoring system toward a multipollutant approach. However, to date, development of this strategy has involved primarily federal, state, and local air pollution agencies. A need still exists for a broader involvement of air quality and health researchers to optimize national networks for the purposes specified above within the existing resource constraints. INVESTIGATING THE HEALTH EFFECTS OF LONG-TERM EXPOSURE TO AIR POLLUTION Epidemiological Approaches The striking findings of the Harvard Six Cities Study (Dockery et al. 1993), which linked chronic exposure to increased mortality, provided a

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress strong impetus for reevaluating the PM NAAQS, particularly after their confirmation in the 1995 publication based in the American Cancer Society’s Cancer Prevention Study 2 (CPS 2) (Pope et al. 1995). The findings on increased mortality associated with longer-term exposures to higher concentrations of particles suggested that the associations observed in the time-series studies did not reflect only a slight advancement of the timing of death for frail individuals. The findings of the two studies were confirmed with an extensive reanalysis (Krewski et al. 2000) and on further follow-up of the CPS 2 cohort (Pope et al. 2002). Findings from several other cohort studies have also been reported (Abbey et al.1999; Lipfert et al, 2000; Hoek et al. 2002). Although these cohorts have provided critical evidence for long-term effects, evidence from further follow-up of these two U.S. cohorts alone will have little use for decisionmaking. The cohorts were established decades ago, and some critical data items, including residence history and potential confounding and modifying factors, have not been comprehensively updated. Consequently, an increasing degree of exposure misclassification can be anticipated as the participants move from their original residences. And, most important, characterization of current air quality cannot recreate the complex air environments in which the individuals and populations lived and worked in the many years for which data are not available. Long-term studies are likely to remain central, however, in assessing the public health burden caused by air pollution. For quantitative risk assessment and cost-benefit analysis, estimates of the disease burden associated with exposure to particles are needed. These estimates could come from a new generation of studies with more complete information on short- and long-term exposures to PM, its components, and exposures to other pollutants. Recognizing both the limitations of these studies and the need for ongoing information on long-term exposure to air pollution and health, the committee recommends that research approaches continue to be developed on the basis of existing and new cohorts. Mechanisms are needed for enrollment and tracking of cohorts over time to provide an ongoing characterization of any impact on health of long-term exposure to air pollution. Without substantial commitment of personnel and funds, studies, such as the Six Cities Study and the CPS 2 cohorts, cannot be readily and feasibly undertaken. Rather, such studies might be based on cohorts routinely enrolled for other purposes, for example, investigating cardiovascular diseases (Atherosclerosis Risk in Communities [ARIC 2004] and the Cardiovascular Health Study [CHS 2003]), Medicare participants, and cohorts assembled by the National Center for Health Statistics. However, even such studies will require substantial funding, and their value must be

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress compared with data collection specifically designed as long-term studies of health effects of air pollution. Medicare has a large cohort under follow-up that is maintained with replacement sampling. The Veterans’ Administration also has a large cohort under follow-up. In addition, there might be other opportunities for adding a component related to air pollution and health; the anticipated National Children’s Study (2004) is one example. That study might provide insights into air pollution and childhood asthma or lung development, for example. New cohort studies of persons having informative patterns of exposure or heightened susceptibility may also be warranted. Studies of effects of long-term exposure to PM, based on residence location and other information, need to include large numbers of participants and to incorporate exposure estimates. With information on residence location, the EPA’s monitoring data, captured in the Air Quality System (AQS) database (EPA 2004), could be used to estimate exposures. However, these data might not be optimal for health studies, and additional data collection or model data would be needed to better capture population exposure (see Chapter 6). For example, the spatial detail within communities might be better captured with focused monitoring and use of population exposure models. As the AQS data are increased from the new speciation sites and other data-collection efforts, it should become possible to develop estimates for exposures beyond particle mass alone. It is critically important that future monitoring strategies go beyond currently regulated pollutants to allow the testing of a broader range of epidemiological hypotheses. An additional concern in any cohort study is the availability of information on potential confounding and modifying factors. Life styles and the associated frequency of chronic diseases, particularly heart and lung diseases, are variable across the country. There is a potential for a varying profile of susceptibility to PM across the country and for confounding as well. Some approaches based on population-level data can be identified that might be used to characterize potential confounding and modifying factors. Population-level data are available on tobacco sales, although they are a poor surrogate for actual smoking rates within the cohorts; available data on prevalence of tobacco use and mortality provide an index of the underlying rates of chronic heart and lung disease, particularly coronary heart disease and chronic obstructive pulmonary disease. Population sampling might be done to augment those data resources. However, such population-level data are inherently imperfect measures of individual-level exposures. Some health-system-based cohorts, such as Medicare, include information on diagnoses leading to outpatient visits and hospitalizations. Those data could be used to identify susceptible groups.

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress The development of new approaches to carrying out these cohort studies will be challenging and time-consuming and should be supported by EPA or other agencies. In 2001 and again in 2003, EPA sought new cohorts for studies of long-term effects through its Science to Achieve Results (STAR) grant mechanism, but it should also support an ongoing planning effort. Although a request has been initiated by EPA to establish a long term cohort to follow up cardiovascular events, it is important for EPA to recognize the need for continued and substantial financial support necessary for these types of studies. At the same time, it will be important for EPA to continue to support additional alternative approaches. The spectrum of human heath effects has expanded over the past several years (see Table 5-1). Because each of these effects has the potential to result in substantial economic and social consequences, as well as significant health impairment, it is important that continued work be undertaken to quantify as much as possible the degree to which PM contributes to these conditions. Toxicological Approaches Toxicological approaches have proved to be especially challenging to use in addressing the research topic of assessing hazardous PM components. Further exploration is needed of the role that chronic exposure studies of animals (those encompassing most or all of their life span) can play in predicting human health effects from long-term exposure to PM or enhancing understanding of the mechanisms and exposure-response relationships of the effects. The technology exists to expose rodents (or other species) chronically by repeated inhalation to PM or complex atmospheres containing PM for periods up to the full life span. Many such studies have been conducted in the past, but most were high-dose carcinogenicity bioassays based on study designs that might not be useful for predicting long-term health hazards from current concentrations of PM. Practical considerations limit study group sizes to a few hundred animals in chronic exposure studies, resulting in much smaller cohorts than considered adequate for epidemiological studies. In general, the outcome measures of such studies (life-span shortening and histological, hematological, serum chemistry, body and organ weight, respiratory function, and bronchoalveolar lavage assays) have not demonstrated effects at PM exposure concentrations even well above environmental concentrations. For example, the largest bioassay of inhaled diesel emissions involved exposure of only 220 rats and 360 mice per group and did not demonstrate life-span shortening, cancer, or significant noncancer effects from near lifetime exposures to PM emissions

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress quately simulated the pathogenesis and expression of the human health outcome of concern and that group sizes were sufficient to detect low-incidence effects that might be relevant in a public health context. The higher the level of confidence that a susceptible subpopulation was being adequately represented by a responsive animal model, the smaller the group size that might be acceptable. Given these conditions, the most contemporary, detailed, and sensitive biological measurements could be incorporated into a chronic exposure study. Until uncertainties about the specific outcomes of long-term human exposures and the fidelity of animal models for human responses to PM are reduced and until hypotheses about the pathogenesis of chronic PM effects are refined, it might not be appropriate to launch long-term animal studies. A more likely application of toxicological research to understanding the consequences of long-term human exposures to PM lies in the use of studies incorporating repeated exposures ranging from several days to a few months in length to improve the understanding of the mechanisms of effects of such exposures. For example, hypotheses about the effects of chronic PM inhalation on interference with cellular repair mechanisms, antioxidant protection, or immune function might be addressed by intermediate-term, repeated exposures of animals (or, to a limited extent, even by carefully designed clinical studies). Such studies could provide a needed progression beyond evaluation of acute high-dose phenomena to determining the cumulative effects, amplification of effects, progression of effects, or adaptations associated with repeated exposures. Intermediate-term studies could help determine the need for the most appropriate design of future long-term studies. IMPROVED TOXICOLOGICAL APPROACHES The committee recognized the need for complementary epidemiological and toxicological evidence in relation to several of the topics, particularly assessing hazardous PM components, combined effects of PM and gaseous pollutants, and susceptible subpopulations. Toxicological approaches have been limited by the difficulty of replicating real-world inhalation exposures to PM in terms of chemistry, by the frequent use of relatively high doses and instillation rather than inhalation of particles in animal studies, and by the inability to readily replicate the human diseases associated with increased susceptibility in animal models. Toxicological

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress approaches have also proved especially challenging to use in addressing the research topic of assessing hazardous PM components (topic 5). Separating the potential effects of particle size from those of particle chemistry is difficult, because particles of different size can have different chemical characteristics and different rates or routes of clearance, which affect response. Studies of appropriate design are needed, as are well-characterized particle samples for experimental exposures. Other aspects of the design of toxicological studies will also be relevant in studies directed at assessing hazardous PM components and combined effects of PM and gaseous pollutants. To the extent possible, toxicological studies should include exposures at concentrations that are similar to those in the ambient air. If impossible, then exposure- or dose-response relationships need to be characterized down to concentrations that come as close as possible to ambient concentrations. Achieving that level of comparability will provide some assurance that mechanisms of injury in the toxicological studies are likely to be the same as those operating under the usual conditions of human exposure. Such assurance would allow for more confident extrapolation of toxicological findings relevant to the topics of assessing hazardous PM components and the combined effects of PM and gaseous pollutants. In addition, particles should be delivered in a physiologically relevant manner, that is, by inhalation. Alternative modes, such as instillation, do not result in deposition and clearance processes that fully mimic those occurring in inhalation exposure. Intratracheal instillation and other nonphysiological dosing methods have their place for certain exploratory and comparative purposes, but results need to be validated by inhalation to be considered conclusive. The committee commented in previous reports on the need for biologically relevant animal models for the chronic heart and lung diseases considered to increase susceptibility to PM in humans. Various models have been developed, attempting to mimic asthma, chronic lung disease, ischemic heart disease, and hypertension. These models can be very useful for exploring specific health hazards and defining specific steps in the pathogenetic chain, but typically fall short of reproducing the full set of exposure-dose-response relationships, susceptibility factors, co-exposure factors, and disease expressions occurring in humans expressing disease after repeated exposure. Categories of potential mechanisms include the following: Autonomic nervous system responses. Physiological responses.

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress Immunological responses. Irritant and inflammatory responses. Growth and development. Cellular responses. Research on the mechanisms of air pollution effects is proceeding simultaneously with the advancement of cellular-molecular research tools and the understanding of underlying biological mechanisms. For example, gene micro-array techniques are being used in studies of air pollutants even though determination of the most important genes, the roles of the genes, and the best way to evaluate the huge amount of resulting data is still being resolved (see Leikauf et al. 2001). Similar situations exist for interpreting changes in protein products of gene activation, cellular receptor signaling, and the metabolic products of PM-associated compounds and the biological cascades they stimulate. Such studies often have great spin-off benefit by using inhaled toxicants as tools to advance the understanding of biological processes; however, such studies are also often fraught with difficulties in interpreting the results and in determining whether observed responses should be considered adverse. It is appropriate that investigators use the best tools at their disposal to understand the mechanisms of adverse effects from PM and variations in susceptibility and that exposures to pollutant species be used as perturbing agents to study biological response mechanisms. However, until the correspondence between cellular and molecular phenomena and health effects of PM and other air pollutants is well-under-stood, caution should be used in interpreting cellular and molecular changes as representing adverse effects of PM. FROM A PARTICULATE MATTER RESEARCH PROGRAM TO A MULTIPOLLUTANT RESEARCH PROGRAM One further challenge to completing the committee’s research agenda lies in the scientifically artificial separation of research on PM from research on air pollution generally. This separation follows the regulatory approach of setting ambient standards for the six criteria pollutants and emission standards without adequate recognition of their interrelationships in the atmosphere and in determining risk to health. Given the need to develop the evidence base for a particular NAAQS, research has too often been driven on a schedule reflecting the cycle of NAAQS review and a

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress scope restricted to single pollutants rather than air pollution mixtures.1 The committee’s topic 7 identified the need to consider other pollutants along with PM, but more as copollutants than as part of a complex mixture, reflecting the possibly artificial assumption that the criteria pollutants act independently to cause adverse health effects. The focus on individual pollutants does not square with the underlying science. Atmospheric scientists are always mindful of the dynamic nature of the atmosphere, the myriad chemical reactions taking place, and the variable partitioning of substances between gas and particulate phases. Scientists monitoring the ambient environment or monitoring personal exposures must select sampling methods that provide the best possible characterization of the ever-changing atmosphere. The complex nature of the atmosphere and the potential role of multiple pollutants, and perhaps their combinations, in increasing the occurrence of any given health outcome requires that the investigator also consider the potential roles of pollutants other than the specific pollutant that is the target of current regulatory scrutiny. Laboratory scientists also need to consider the complexity of real-world emission sources and exposures, even when taking a reductionist approach to studying a specific pathway for a particular pollutant related to a given disease. That is obvious when it is recognized that all humans live in and breathe complex, varying atmospheres of particles and gases arising from multiple sources as primary emissions and from secondary transforma- 1   In this section, several different terms are used in discussing linkages between atmospheric constituents and human health. The term single air pollutants refers to individual criteria pollutants, such as PM, ozone, or other individual chemical constituents. As will be discussed, much of the past air pollution research has focused on single pollutants. The term multiple pollutants refers to multiple gaseous and PM constituents within a total atmosphere. The term is used fully recognizing that the ambient atmosphere is dynamic in both its spatial and temporal dimensions. The committee advocates at the conclusion of this section that EPA’s PM program over time evolve to a multiple pollutant program that includes the traditional pollutants currently classified for regulatory purposes as criteria pollutants, and hazardous air pollutants (including air toxics) and nonclassified atmospheric constituents. The term one atmosphere has been increasingly used in air quality management discussions to refer to an approach that considers in an integrated manner all the atmospheric constituents when air quality management decisions are made on any single pollutant. Because the term “one atmosphere” has not achieved a widely accepted definition, we use the term “multiple pollutant program” in this report.

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress tions occurring in the atmosphere. The subdivision into criteria pollutants (PM, ozone, oxides of nitrogen, sulfur dioxide, carbon monoxide, and lead), hazardous air pollutants (such as 1,3-butadiene, benzene, and coke oven emissions) and other unclassified chemicals is a reflection of the legislative history rather than a logical scientific taxonomy. Although some pollutants, such as lead, produce specific health effects, the health effects that are associated with most air pollutants typically represent increases in the incidence of chronic diseases that are of common occurrence and have multiple etiologies, such as cardiovascular and respiratory disorders and cancers of multiple organs, or represent exacerbations of these diseases. Moreover, as the concentrations of regulated pollutants in the ambient air continue to fall, even approaching natural background concentrations, the likelihood that specific health outcomes can be ascribed solely or primarily to single pollutants or pollutant classes is likely to be diminished. In addition, among the thousands of compounds in the air, some of them cannot be easily categorized. For example, semivolatiles can exist in either gaseous or particulate form. In contrast to this broader more inclusive view of multiple pollutants, the current expanded PM research program is an extension of the EPA research program on criteria air pollutants that has been operative for many years. EPA’s research on criteria pollutants has been carried out primarily through EPA’s intramural laboratories with a much smaller extramural effort carried out under STAR grants, contracts, and cooperative agreements. To a substantial extent, the research program has been loosely linked to the schedule for revising criteria documents and review of NAAQS. Typically, a few years in advance of the development of a criteria document, the level of research on the criteria pollutant under consideration would be increased to fill perceived critical data gaps. As the deadline for inclusion of new published material in the criteria document approached, the level of research on the criteria pollutant under consideration was reduced, and effort was redirected toward the next criteria pollutant to be reviewed. This pattern was typically maintained for all six of the criteria pollutants. For PM, the epidemiological findings of associations between several PM indicators, especially PM2.5, and risks to health resulted in an increase in EPA funding of PM research. The intensity of the debate over the PM2.5 NAAQS led Congress to appropriate a further increase in funding for PM research and to direct EPA to contract with the National Research Council to form this committee. The resulting research program is novel in multiple

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress ways: (1) its strong strategic orientation, (2) its multidisciplinary character, (3) the breadth of the research activities (within the research paradigm outlined by the committee), (4) the involvement of agency laboratories and extramural institutions, including creation of five extramural academic PM research centers and eight PM monitoring Supersites, and (5) level of funding, (6) continuation of a sustained research effort for more than 5 years, and (7) progress made in reducing some of the uncertainties concerning the health effects of PM within certain aspects of the overall research agenda. While attention was focused on PM, the level of research funding for other criteria pollutants was reduced. Although the PM program has begun to consider other pollutants, it has of necessity focused on PM. Some other air contaminants, largely other criteria pollutants, have been considered in the role of copollutants for their potential impact on the effects of PM. Epidemiologists have largely considered copollutants by determining the extent to which their addition to multipollutant models for data analysis diminished the statistical strength of the association between PM and health outcomes. Even this approach is necessarily limited by the lack of widespread availability of data on environmental concentrations of air contaminants other than the few criteria pollutants whose measurements are mandated for compliance purposes. A few laboratory studies of factorial design have evaluated the effects of PM with and without another pollutant (such as ozone), but factorial designs are not suited for more than three pollutants and thus fall far short of the complexity of environmental exposures; moreover, as stressed earlier, PM itself varies widely in composition. Some studies involving exposures to concentrated ambient PM have also included gaseous pollutants but generally at their original ambient concentration, but the concentration of gaseous agents can be increased if desired. A few laboratory studies of animals and humans exposed to engine emissions have included groups exposed to filtered emissions, thus showing the relative effects of the PM and non-PM fractions. There is an opportunity and a critical need to shift the focus of the EPA program from a single pollutant, PM, to a multipollutant orientation. Because of the momentum that the PM research program has generated over the past 6 years, now is an opportune time to begin orienting EPA’s air quality research program toward a broader scope that specifically considers all components of the atmosphere—PM and the other criteria pollutants, hazardous pollutants, and the other nonclassified components of the atmosphere. The committee envisions a transformation from a PM-focused research program to a multiple air pollutant program (MAPP).

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress The committee’s MAPP concept recognizes that multiple sources contribute multiple pollutants (particles and gases) that are subject to atmospheric transport and transformation. This mixture of pollutants, which varies in composition by time and location, serves as the source of exposures to human populations and ecosystems, resulting in a wide range of adverse effects in humans and ecosystems. The adverse effects that result from the mixtures of multiple pollutants may be attributed to single pollutants or, to a variable and largely unknown degree, combinations of pollutants that together make up an infinitely variable atmosphere. In part, the scientific emphasis on isolating effects of single pollutants follows from the artificial regulatory separation of individual components of the pollution mixture. The committee envisions the MAPP concept to incorporate as a guiding principle the paradigm advocated by this committee for research on PM (see Figure 1-1 in Chapter 1): source ambient atmosphere personal exposure tissue dose health response. The committee urges that single air pollutants be considered and addressed in the comprehensive context of the range of multiple pollutant ambient air environments that people actually experience. The committee recognizes that shifting to the MAPP concept will require the development of new scientific approaches to evaluate the linkage of multiple sources of pollutants and multiple pollutants to health effects. This broadened approach should lead to epidemiological study designs and analytical methods that better address the health risks of the components of mixtures and to enhanced toxicological study designs that elucidate biological mechanisms. The improved scientific understanding of the role of pollutants as components of a complex mixture will provide the science base essential for informing future regulatory actions and related control strategies. The move from a pollutant-by-pollutant orientation to a multiple pollutant orientation is viewed as an evolutionary rather than revolutionary change. Indeed, the multiple pollutant orientation was key to past advances in the scientific understanding of the linkages between primary emissions of PM, volatile organics, sulfur oxide, and nitrogen oxides; atmospheric transformation to ozone and secondary particulates, including sulfates and nitrates; and personal exposure to these mixtures with associated health effects. An understanding of these linkages undergirds the one-atmosphere principle that is emerging to guide air quality management strategies (NRC 2003). A logical next step in evolving to a MAPP is to develop and implement an integrated research program that includes PM and the other criteria

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress pollutants, hazardous air pollutants, and currently nonclassified atmospheric constituents. The successes as well as the lessons learned from creating and implementing the PM program over the past 6 years provide building blocks for the broader initiative. Substantial expertise has been mobilized in the intramural laboratories of EPA and in research laboratories of universities and other institutions across the United States. A continually improving science base for PM and other atmospheric constituents is essential for ensuring continued progress in improving total air quality and reducing air-pollution-related health impacts. Some continued movement toward a multipollutant approach will undoubtedly result from research on hazardous PM components and on combined effects of PM and gaseous pollutants. However, there is a clear need to apply the strategic multidisciplinary orientation that has proved useful for conducting research on PM to the broader study of other criteria and hazardous pollutants. Methods and models developed by the future PM research program are likely to be useful for studying multiple pollutants. A science-based multipollutant approach can be useful for the development of information relevant to setting standards and developing air quality management strategies. One very important benefit of this approach is the likelihood that the resulting information will aid the understanding of the relative importance of the various pollutants (and thus sources) and their interactions in causing adverse health effects. The approach should be beneficial in optimizing the cost-effectiveness and the “health-effectiveness” of future air quality management strategies. Indeed, it is conceivable that a multipollutant approach to reviewing scientific uncertainties; developing research strategies; conducting research; and analyzing, interpreting, and reporting research findings will extend to developing more integrated criteria documents. That could lead to integrated development of standards, implementation plans, and control strategies that have a stronger science base than that achieved by past single pollutant approaches. INTEGRATING ACROSS THE DISCIPLINES The need for complementary evidence on PM from toxicological, exposure, epidemiological, and atmospheric approaches was recognized early by the committee, which called for interdisciplinary research and proposed the PM centers as one mechanism for fostering collaboration across disciplines. Although there has been greater cross-disciplinary

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress integration of some PM research topics, to a large extent the coordination of epidemiological, toxicological, exposure, and atmospheric research has received more discussion than implementation to date. Expanding multidisciplinary strategies and programs will be essential for implementing the committee’s MAPP approach. The concept of integration across disciplines is both logical and appealing, but it has been difficult in practice to identify research models appropriate for coordinated research or to implement focused interactions between the fields. Exchanges of information between field and laboratory scientists and between scientists studying people directly and those using nonhuman research models are more frequent now than in the past. This improved communication, occurring both within research centers and at scientific meetings, has provided opportunities for the different fields to work in less isolation than a decade ago and to build greater knowledge of other disciplines’ research principles and methods; some researchers have taken good advantage of these opportunities. However, research efforts in which epidemiological and toxicological tools are merged in coordinated, preplanned research strategies to answer specific questions remain infrequent. There are examples of the “hand-off” of research issues between epidemiology, exposure research, and toxicology, although none resulted from preplanned coordination of efforts. Perhaps the best example is the demonstration by laboratory studies that the soluble iron content of ambient PM from Provo, Utah, was related to PM toxicity (Ghio et al. 1999) and that these differences corresponded to differences in population health outcomes measured during the times that the ambient particle compositions in Provo varied (Pope 1996). As a second example, the epidemiological evidence for PM-related cardiovascular morbidity and mortality has caused toxicologists to strive to reproduce presumed associations between exposure and effects and to explore underlying mechanisms, such as alterations in cardiac electrophysiology and blood-clotting factors. Such examples, although providing useful information, have not been sufficiently common. In investigating the health effects of airborne particles, epidemiology researchers have often treated PM as a single agent. Research on airborne particles has made clear the simplifying nature of that assumption, as the chemical and physical complexity of particles in the atmosphere across places and time has been described. To understand the health risks of the PM mixture and the likely differential toxicity of different components of that mixture outlined in the committee’s Research Topic 5, and then to increasingly place that understanding in the MAPP context

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress and to determine the most efficacious methods for reducing emissions and risk will require an entirely new level of collaboration among the disciplines to integrate actual exposures with effects. This integration has begun with efforts by the atmospheric community, through NARSTO,2 to reach out to the health community and is also evident in some health research that involves increasingly detailed characterization of the PM mixture to which animal and human subjects are exposed. However, substantially improved integration of epidemiological and toxicological approaches, incorporating improved metrics of atmospheric and personal exposure, will be required to advance the knowledge of PM health effects. For example, laboratory studies of nonhuman biological systems can be designed to explore the basis of causality and describe dose-response relationships, mechanisms of response and susceptibility, markers of exposure and effect, key PM components, and effects of copollutants at a level of detail and precision not possible in the population. As another example, research on emissions characterization and air quality model testing and development could be better integrated. Thus, even better coordination between receptor modeling, grid-based modeling, and emissions research is needed. The specific question is whether the source profiles used in receptor models are consistent with current inventories or whether they indicate the presence of gaps. If so, another question is whether improvements in the inventories lead to better grid-based model predictions. To some extent, the challenges of integrating disciplines will always be there—differences in the cultures and terminology of different communities of scientists and, to some extent, institutions that conduct the different types of research are inherent, and difficult to overcome. However, the likelihood of success will greatly be enhanced when atmospheric, exposure, epidemiological, and toxicological research tools can be integrated proactively into combined, interactive research strategies to answer specific questions rather than proceeding in parallel to address similar general issues. There are hopeful signs of such efforts, such as the Fresno Asthmatic Children’s Environment Study (FACES 2004), which California, EPA, and others are implementing around the Fresno (California) Supersite, and some of the efforts of the PM centers. Much more extensive efforts are necessary, however, to ensure that the full suite of issues related to the health 2   NARSTO, formerly known as the North American Research Strategy for Tropospheric Ozone, is a multiple stakeholder body organized in 1994 with financial support from the public and private sectors to sponsor public- and private-sector policy-related research on tropospheric ozone and PM.

OCR for page 127
Research Priorities for Airborne Particulate Matter: IV - Continuing Research Progress effects and sources of the complex mixture of pollutants in urban air are better understood over the long term. At a minimum, these efforts should include Active collaborative research design. A shift by funding agencies toward giving higher priority to research implemented by truly multidisciplinary teams. Adequate research funding for projects to allow the active involvement of a full team, including senior investigators from multiple disciplines, if needed. Fellowships or sabbaticals that will enable scientists to spend time with groups outside their disciplines. Redoubled efforts of appropriate professional societies to hold joint workshops and meetings and to publish proceedings. Ultimately these efforts will need to result in fully multidisciplinary review, integration, and synthesis of the science by EPA in the criteria document and staff paper processes. SUMMARY AND CONCLUSIONS In 1998, the committee recognized that meeting its research agenda would require a substantial investment as well as the development of new research approaches to address complex scientific questions. In reviewing work carried out since that report, the committee has identified seven scientific challenges that should be a focus of further work to complete the PM research agenda. Of course, there are other challenges, but they are not as critical to moving forward on the full agenda. The next chapter gives the committee’s guidance on strategies to meet these challenges.