Research Priorities for Airborne Particulate Matter: III. Early Research Progress (2001)

This is the third in a series of reports by the Committee on Research Priorities for Airborne Particulate Matter. The committee was convened by the National Research Council (NRC) in January 1998 at the request of the U.S. Environmental Protection Agency (EPA) following directions from Congress in EPA's fiscal year 1998 appropriations report. The congressional request for this study arose from the uncertainties in the scientific evidence used by EPA in reaching the July 1997 decision to establish new National Ambient Air Quality Standards (NAAQS) for particulate matter (PM) less than 2.5 µm in aerodynamic diameter (PM_2.5). Anticipating the next scheduled review of the standards in 2002 and every 5 years thereafter, Congress appropriated substantial funds for a major research program to reduce the scientific uncertainties. It also directed EPA to arrange for this independent NRC study to provide guidance for planning the research program and then monitoring research progress for the 5 years of 1998-2002.

The committee's first report, Research Priorities for Airborne Particulate Matter: I. Immediate Priorities and a Long-Range Research Portfolio, was released in 1998. It proposed a conceptual framework for a na-

tional program of PM research, identified 10 high-priority research topics linked to key scientific uncertainties relevant to setting policy, and presented a 13-year, integrated “research investment portfolio” including recommended short- and long-term scheduling and estimated costs of the research. In developing its research recommendations, the committee did not undertake to judge the adequacy of the scientific foundation for EPA's 1997 decision to establish the new PM standards, recognizing that such policy considerations extend beyond the realm of science, and the committee was neither charged nor constituted to undertake a review of the standards.

In its second report, Research Priorities for Airborne Particulate Matter: II. Evaluating Research Progress and Updating the Portfolio, released in 1999, the committee described its plans for monitoring the progress of the research. In addition, the research recommendations from the committee's first report were updated, and two of the recommended research topics were substantially revised.

Congress and EPA have given strong support to the recommendations presented in the committee's first two reports, as indicated in Table ES-1, which shows funding levels budgeted during fiscal years 1998-2001 for EPA's PM research and related technical work. In contrast, the fiscal year 1997 funding level for PM research and related technical work was about $21 million. EPA's Office of Research and Development (ORD) has shifted its PM research program to respond

to the research priorities identified by this committee in its first two reports. The overall effort involves inhouse research studies at EPA laboratories and centers, EPA funding of university-based research centers and investigator-initiated competitive research grants, and enhanced collaboration with other agencies and organizations. The majority of the funding has been used to support inhouse studies in EPA. Substantial research is also being funded by other governmental and nongovernmental agencies.

In preparing this, the third report, the committee began the task of assessing the progress made in implementing the PM research program. The committee has matched the PM research projects sponsored by EPA and other institutions against the committee's recommended research portfolio to assess the extent to which ongoing research is addressing the major issues that decisionmakers will consider as they review the scientific evidence relevant to the PM NAAQS. For each of its 10 priorities, the committee then evaluated current and planned research activities with six criteria: scientific value, decisionmaking value, timing and feasibility, multidisciplinary interaction, integration and planning, and accessibility of information. The first three criteria were applied to the specific priorities while the last three were considered as overarching and applying to the full program.

The committee reviewed the progress made in implementing PM research from 1998 (the year in which the research portfolio was first recommended by the committee) until the middle of 2000. Because this period represents the initial stage of the PM research program, the committee's assessment necessarily focused more on current and planned research projects than on published results. Therefore, the committee's evaluation should be viewed as an interim review of the PM research program which will be updated and modified in the committee's final report in 2002.

In its first report, the committee recommended that information be obtained on relationships between total personal exposures to particles (the exposure measured by placing a monitor directly on a person) and outdoor concentrations of PM. Specifically, the committee recommended that groups of 10-40 persons should be studied over time to examine the relationship between their personal exposures to particles and the corresponding outdoor concentrations. There was also a recommendation that these studies include not only healthy persons, but persons considered to be “susceptible” ¹to PM.

Considerable research is under way on the relationship between ambient particle concentrations and personal exposures. While more time than expected has been taken to initiate these studies, many of the elements of research topic 1 are being addressed. Collectively, the studies are assessing general population exposures to PM and gaseous copollutants (such as ozone) and potentially susceptible subpopulations (such as asthmatics). Personal exposures are correlated with outdoor ambient exposures, but the correlations differ among people depending on their activities and other factors. The current studies are expected to expand greatly the database on personal exposures and the influencing factors, such as indoor PM concentrations, human activities, and characteristics of the places where people spend time (microenvironments).

Preliminary results from the studies have contributed to understanding of the relationship between personal exposures and outdoor concentrations and have added data from potentially susceptible groups. However, the pattern of findings is still based on a small number of studies, and replication of the results will be needed from current or recently completed studies in other cities before firm conclusions can be drawn.

The committee previously recommended that after obtaining and interpreting results of studies from research topic 1, studies be conducted to examine human exposures to specific chemical constituents of PM considered relevant to health effects. The recommendation included exposures to both the general population and potentially susceptible subpopulations. Methods of measuring some characteristics of the particles to which people are exposed are available and being field-tested for this purpose, including various physical properties (such as particle number and size) and chemical characteristics (such as presence of sulfate, nitrate, and carbon and other specific elements). In addition, methods for measuring personal exposures to some gaseous pollutants—such as ozone, nitrogen dioxide, and sulfur dioxide—are available.

The results of the studies in research topic 1 should facilitate the design of cost-effective protocols for future population-exposure studies related to priority 2 that focus on PM components considered in determining toxicity. These studies will be directed at properties of particles found to be important in toxicological and epidemiological studies. Because the research needed to determine PM toxicants is still in progress, the committee expects that research activities related to priority 2 will not begin for a few years.

The third research topic recommended by the committee was measurement of the size distribution and chemical composition of PM emissions from key sources. In addition, characterization of the emission rates of reactive gases that can form particles on reaction in the atmosphere is needed, including emission data on sulfur oxides, nitrogen oxides, volatile organic compounds, and ammonia.

Although the scientific merit of the work under way to develop new source-testing methods is high, potentially greater benefits for decisionmaking would emerge from more-complete and more-accu-

rate knowledge of particle emissions by size and composition. To obtain the needed data, EPA should expand its source-testing program substantially over the next several years. At the same time, EPA should develop a comprehensive plan for systematically using the emissions test results to create a comprehensive emission inventory for the nation based on standardized tests of pollution sources. There is still ample opportunity to plan and conduct such a program over the next 5 years, and the results will be needed to develop cost-effective programs for reducing PM exposures.

Source-oriented models are used to estimate the concentration of airborne particles at specific locations. Receptor-oriented models look backwards from a location towards the sources of particles. Research topic 4 calls for research to develop, test, and evaluate source-oriented and receptor-oriented models that represent the linkages between emission sources and ambient concentrations of PM. Comprehensive source-oriented models for PM are still under development. However, before such models are ready for regulatory applications, they require more-certain emission inventories (as described under the previous research topic) and an improved knowledge of the chemical and physical processes that determine the size distribution and chemical composition of ambient particles. To enhance receptor-oriented models we need better mathematical tools for identifying informative patterns in the spatial and temporal variability of particle concentration and composition. To be used with confidence, the results from receptor models and source models should be cross-compared and also validated against observations from intensive field-measurement campaigns.

Current studies are expected to make substantial contributions to our understanding of chemical and physical processes that determine the size distribution and chemical composition of ambient particles. However, additional research is needed to develop, test, and evaluate source-oriented and receptor-oriented models. In general, there has

not been sufficient effort to evaluate and compare the models developed by EPA and others. There has been inadequate attention to organizing and carrying out field studies to provide the data needed for thorough evaluation of existing models.

The ambient-PM monitoring program should provide an essential foundation on which intensive field-measurement campaigns can be built. However, little effort is being given by EPA to leveraging its investments in the PM monitoring program in order to provide field data for model evaluation.

The committee's previous reports outlined an agenda to improve understanding of the role that specific PM characteristics (such as particle-size distribution, shape, and chemical constituents) play in determining toxicity for the health outcomes associated with PM exposures. Research was also recommended to develop experimental models, including PM surrogates, for use in toxicity studies. As the committee's recommendations are followed, epidemiological research will assume increasing importance because of the need to establish linkages in the population between PM sources and health effects as a basis for setting standards and implementing control strategies.

To date, new work on this subject has been based largely on toxicological approaches. The question of whether biological responses to PM are nonspecific—that is, are due merely to inhalation of any particle —or depend on specific PM properties is a critical focus of current research. There is considerable research evaluating physiochemical properties of PM in relation to biological effects. However, the effort has generally focused on only a few chemical characteristics, with the largest body of work involving metals—which are found at very low concentrations in the atmosphere. Other potentially important PM characteristics (such as particle number, concentration, and surface area, as well as specific particle constituents, including bioaerosols and organic compounds) have received less attention.

Current work is beginning to address the issue of exposure and dose measures other than mass concentration, although most studies continue to evaluate health effects in relation to total or size-specific mass concentration during exposure. The committee is unable to identify any studies under way that incorporate experimental PM surrogates that can mimic daily, seasonal, and regional variabilities of particle characteristics. Such experimental agents are needed so that research can be focused on specific components. Timely research is also needed to develop animal models that more closely mimic human heart and lung diseases. Study designs should give greater consideration to the relevance of the high doses used in many controlled-exposure studies and the typically low exposure levels to some components of ambient PM.

In epidemiological investigations, because insufficient monitoring data for specific particle characteristics have been available, opportunities for testing hypotheses related to the characteristics have also been somewhat insufficient, and few studies have incorporated substantial monitoring of both particle concentration and other specific characteristics of particles.

Epidemiological studies need to include sufficient exposure assessment to provide findings to guide toxicological studies of PM characteristics. However, because the populations in prospective epidemiological studies are often large, it is generally not feasible to obtain personal exposure measurements for all subjects. Opportunities should be sought to apply research models that combine toxicological and epidemiological research. In addition, consideration should be given to new population-based approaches that could be useful for hypothesis-testing and eventually surveillance. A number of studies in progress should provide information on the risks posed by ultrafine particles.

In its first report, the committee recommended studies to improve scientific understanding of the deposition, translocation, and clear-

ance of particles deposited in the respiratory tract. Although the dosimetric studies under way are at a level of effort meeting the committee's recommendations, there is not yet sufficient focus on the specific information needs described by the committee. To the extent that those research needs are being addressed and in light of the previous lack of dosimetry data for persons who have respiratory abnormalities or in animal models of those conditions, the scientific value of the work is generally high. Nearly all the work builds on previous knowledge in an incremental way and will lead to a more integrated understanding of PM-related health effects.

Only a small portion of the work has addressed characteristics other than age and sex; there has been insufficient work on the impact of respiratory diseases. The committee called for development and validation of mathematical models for predicting deposition and clearance in abnormal lungs. However, there has been only modest advancement in the modeling of dosimetry in potentially susceptible people and very little effort to validate the models. Efforts to improve interspecies-extrapolation models continue in a few laboratories, but there has been little effort to validate the models. There has also been little effort to assess dosimetry of any type in animal models of human respiratory abnormalities. Many potentially important aspects of respiratory abnormalities—such as microdosimetry in tissues and cells, bioavailability of particleborne compounds, translocation and clearance, and handling of diverse particle types—have been addressed very little.

Because ambient PM is part of a pollutant mixture also containing gases, any biological effects attributed to PM in an observational study might also reflect contributions of gaseous pollutants. Research relevant to this topic should include toxicological and clinical studies that examine the effects of gaseous copollutants on the toxicity of PM.

Although research to assess if gaseous copollutants can affect the toxicity of PM has been increasing, the current controlled-exposure

research on this topic is not adequate. Particles can be concentrated in urban air for research in experiments, including exposures of human volunteers. The concentrated particles are referred to as concentrated ambient PM (CAP). The particle concentrators do not concentrate gases; approaches should be sought to augment CAP with concentrated gaseous pollutants or to scrub out specific unconcentrated gases from CAP. Furthermore, there appears to be insufficient toxicological research effort on the role of gases in influencing particle toxicity in comparison to the effort directed at studies of specific components of PM, absent gaseous copollutants. However, the epidemiological research portfolio for this topic is relatively strong; most epidemiological studies of PM include data on gaseous copollutants. Although the array of PM exposure indicators across the studies is broad, most include results of monitoring the principal gaseous pollutants of concern, including ozone, nitrogen dioxide, sulfur dioxide, carbon monoxide, and irritant hydrocarbons. Sample sizes of the studies range from obviously too small (and consequently uninformative) to adequate, so that some studies should have enough data to provide insights into combined effects.

Most research for this priority is directed at acute effects of relatively short-term (modeled or directly measured) exposures to ambient particles. In its second report, the committee recommended that increased efforts be undertaken to conduct epidemiological studies of the effects of long-term exposures to particle constituents, including ultrafine particles. However, effects of chronic exposure are being considered in only a few studies, and the current studies model long-term exposure on the basis of relatively recently measured exposures combined with historical extrapolations. There does not yet appear to be a systematic, sustained plan for implementing human chronic-exposure studies, including examining ultrafine particles.

Several groups in the population have been considered susceptible to air pollution in general and to PM specifically. In each subpopu-

lation, there is likely to be a range of susceptibility. Taken together, current research efforts are expected to provide a rigorous evaluation of PM risks to susceptible subpopulations with chronic diseases, such as asthma, chronic obstructive pulmonary disease, coronary arterial disease, heart failure, and hypertension. The evidence exists primarily in relation to PM mass as an exposure index. A complete understanding of risks in susceptible subpopulations will require research that cuts across several areas, including exposure, dosimetry, toxicity mechanisms, and epidemiology. There is a general need for improved and validated animal models of human disease. In much of the experimental work, the investigators are studying relatively young animals, perhaps in the first fourth of their normal life span. In contrast, a substantial portion of the human diseases of concern occurs in the last quarter of the normal life span.

The major potential biological responses which have been suggested as underlying the reported human health effects from ambient PM exposures include oxidative stress, pulmonary inflammation, airway hyperreactivity, and alterations in the cardiovascular system, such as changes in blood viscosity, heart rate and pattern of beating, and heart rate variability. The issue of mechanistic plausibility has been addressed with a number of approaches: animal models, in vitro systems, and clinical models. In studies examined by the committee, about half of the studies on mechanisms involve animal toxicology, and the other half are roughly evenly divided between clinical and in vitro studies.

It is encouraging that numerous animal models are being used to measure the effects of exposure to PM. However, a substantial number of studies have reported exposing healthy normal animals to particles; this may not be an informative approach for modeling the consequence of exposure of susceptible humans. Even though animal models of cardiac and lung diseases are being used to investigate the

effect of particles, their relevance to humans must be carefully considered. Before they are used, the models need to be well characterized and validated relative to human disease.

Many laboratories can conduct in vitro studies. The current and planned in vitro studies are designed to investigate several components of PM with a number of outcomes, such as inflammatory cytokines, chemokines, release of oxidants, and oxidative stress responses. The issue of age-dependent responses, as well as modulation of responses in cells from susceptible subjects, is also being investigated. A few studies have been designed to assess dose-response relationships. However, the committee is unable to identify any current in vitro studies that have used relevant low doses to test the validity of conclusions for specific mechanisms but instead have used unrealistic high doses. Despite those shortcomings, which need to be rectified, comparative in vitro toxicity studies to establish concepts and elucidate mechanisms of PM toxicity are valuable additions to the database.

The current and planned studies of human volunteers are designed to investigate CAP and several specific characteristics of PM (such as size, acidity, and associated metals) with a number of local and systemic end points. Studies are under way in potentially susceptible populations and are planned in other subgroups with pre-existing diseases. Despite the small number of facilities available for carrying out human exposures, the array of studies under way should provide valuable information on PM toxicity.

In its previous reports, the committee outlined several methodologic issues that need further study. They included the choice of statistical methods for analyzing data obtained from studies, especially epidemiological studies. Because more than one method can be used to analyze data, it is important to understand the extent to which use of alternative methods might influence the analytical results. There is a critical need for analytical methods that will aid in

characterizing exposure-response relationships at low exposures, including the range of concentrations found in ambient environments in the United States.

The recent research appears to address the research gaps and needs identified by the committee. That is especially true of measurement error and harvesting issues. ² This research is new, however, and experience needs to be gained with the methods as they are refined and used to analyze data sets. Data that will allow further testing applications of the methods are either available or being collected. The value of the research will increase as it is applied to more—and in some cases better—data sets. The value will also increase when the various studies, methods, and results can be compared and synthesized.

The committee identified several overarching issues based on application of its three evaluation criteria (that is, multidisciplinary interaction, integration and planning, and accessibility of information) for assessing the implementation of the PM research program.

Institutional and cultural obstacles often discourage attempts to perform research across disciplines, agencies, and institutions (including public, private, and nongovernmental organizations). Such obsta-

cles tend to sustain historical tendencies to conduct research within particular disciplinary or organizational boundaries (for example, toxicology versus epidemiology, EPA versus DOE, or government versus industry).

In viewing such disincentives as they apply to the inter-related tasks of conducting PM research, the committee recommends that a series of incentives be established, or in some cases continued, to orient professional and institutional policies, practices, and behavior in favor of joint planning and information exchange. These incentives should include the following:

• Encouraging federal-agency PM research programs to give greater priority to integrated, multidisciplinary projects.

• Developing a unified, cross-agency federal budget for key PM research initiatives (that is, a “virtual-agency or multiagency budget” for PM research). The committee is aware that the Air Quality Research Subcommittee of the interagency Committee on Environment and Natural Resources (CENR)³ has undertaken such budgetary coordination for PM research. The subcommittee is encouraged to extend this coordination to the greatest extent practicable.

• Conducting regular and frequent multisponsor, multidisciplinary gatherings to build a common community of PM investigators with the goal of determining whether or not the research under development is being adequately integrated for the purpose of achieving the principal goals of the PM research program. A series of triennial colloquia and other meetings have largely met this goal.

The committee has emphasized the critical need for the federal government to provide comprehensive and integrated management of the PM Research Program. Without such management in place, it is likely that many useful individual research projects will not be synthesized into a coordinated, multidisciplinary program to ensure that key PM research questions are answered.

Since the committee's second report, measurable progress has been made in several aspects of integrating and planning the conduct of the PM Research Program:

• EPA established a formal management structure, led by a senior PM program manager within ORD. This has enabled the initial development of multiyear research budgets (an important initiative) and regular reporting of budget priorities and progress toward addressing the committee's research portfolio. In addition, the assistant administrator for ORD had tasked its Board of Scientific Counselors to review the management of the program in detail and provide specific recommendations on how to improve it.

• Across the federal government, the charter of the Air Quality Research Subcommittee of the CENR has been expanded, and efforts are under way, as a critical first step in creating an integrated federal strategy, to create a complete federal inventory of PM research and to make it available through the Particulate Matter Research Activities web site (www.pmra.org). On the basis of the inventory, the subcommittee is now developing a strategy for integrating PM research that is sponsored by federal agencies.

Those efforts have provided the PM research program with the potential to be well integrated and well planned. Although it is too early to assess the effectiveness of current management efforts fully, several aspects of the current management structure pose potential challenges to successful implementation of the PM research program.

ORD has put in place a formal management structure that should help to ensure program success, and there have been initial efforts to integrate health, exposure, and atmospheric research interests into the ambient air monitoring system. EPA's Office of Air and Radiation (OAR) has responsibility for the monitoring program and participates in EPA's PM research management structure. Although it has taken steps to integrate research needs into the implementation of the ambient monitoring program, OAR should strive to develop a stronger management system for its PM technical program which is comparable to the PM management system within ORD. An enhanced management structure for the PM technical program within OAR is needed to ensure full coordination with ORD, to catalyze active involvement of the state air agencies in developing and managing the speciation network (a key source of future data to support health research), and to develop fully the source-emission inventories needed for accurate source apportionment.

Also, past efforts to coordinate federal research have shown that this type of unified management strategy cannot be readily achieved. Sustained efforts on the part of the CENR Air Quality Research Subcommittee will be needed to ensure that these next, more-challenging steps are accomplished.

On balance, many of the elements for successful integration and planning of the PM research program have already been put in place. Successful efforts to address these remaining planning and integration challenges are critical to ensure that the maximum research return is obtained from the sizeable investments currently under way.

The continuing process of planning and applying research in ambient air-quality standard development is part of an established process that includes preparation of EPA's air-quality criteria documents and the staff papers to inform the review of air quality standards. Both documents undertake a synthesis of existing peer-reviewed scientific information and present technical implications relevant to selecting

among the policy options for standard setting. The scientific community and the broader public have long participated in the process of reviewing the adequacy of scientific information and its application in the above documents.

There is a need to extend more effectively the application of research planning and management and scientific review into the standards-implementation process. This need exists because of the many scientific uncertainties associated with implementation, including how to account for regional variations in airborne PM characteristics and long-term changes in PM air quality due to changes in economic activities. Therefore, the committee recommends that EPA develop a research-management strategy to address key uncertainties concerning implementation of air-quality standards for PM.

The committee has found the ongoing research-projects inventory initiated by the EPA and Health Effects Institute (HEI) to be a useful means for interested parties to track research in progress as the body of PM research continues to grow. The database can also be used to identify and plan future research. Efforts should continue to develop and maintain the database, and make it widely known and used by scientists in government, the private sector, and universities, as well as the broader public.

In addition to the PM research inventory database, the committee identified the need for the PM research program to provide information on its planning, budgets, progress, and results to parties that have an interest in PM research. Such parties could include other research organizations, public-health and environmental organizations, industry groups, health professionals, the interested public, and the news media. Information about PM research plans and results should be easily and effectively accessible. And, enhanced efforts should be used to inform and engage interested parties in understanding the multidisciplinary aspects of research results.

Throughout the PM research program, there have been efforts to

enhance accessibility to publicly funded data for scientists and others, as evidenced by the provisions of the “supersite” program and other elements of the air-quality measurement program for central archiving and public data access. In response to legislation, recent revisions to federal grant-making rules have broadened access to the data created by federally funded research.

Although the initial phases of the nation's PM research program have shown promise, and a substantial number of peer-reviewed publications are forthcoming, there is as yet insufficient evidence for the committee to predict the program's ultimate effectiveness. In general, the PM research program, following on the committee' s priorities, is appropriately addressing many of the key uncertainties. However, as discussed in this report, there are a number of critical specific subjects that should be given greater attention. Because research results are coming more slowly than originally expected, managers of the research program may need to adjust the timing of future research activities.

EPA and other organizations have made progress in integrating the full range of research approaches into the implementation of the national PM air-quality measurement system. However, key shortcomings (such as inadequate efforts to provide for data analysis) have limited the agency's ability to plan for and take full advantage of the wealth of new data on air quality likely to emerge from the system. EPA and other organizations will need to support research to ensure that success is achieved.

As the committee continues its evaluation of PM research progress over the next two years, it will seek information to determine the extent to which the research program has provided measurable

growth in knowledge that informs the policy-relevant issues that scientists and decisionmakers must resolve in reviewing the NAAQS for airborne PM. It will also assess the full scope of research results obtained from the individual scientific disciplines in relation to the 10 research priorities; its overall assessment will consider how much understanding has been advanced in the committee's paradigm from sources of PM exposures to health effects. In addition, the committee will consider the degree of acceptance by public and private decisionmakers and interested parties of the quality and relevance of the scientific information obtained. A key determinant will be the extent to which the research findings inform the setting of future NAAQS for PM, namely, identification of the indicator (such as PM _2.5), concentration of the indicator in air, time over which measurements are made or averaged, statistical form of the standard used to determine the allowable number of exceedences, and the effective and efficient implementation of emission-control programs to achieve the NAAQS. In the future, it will also be important to ask whether the PM research and this committee's role offer a model for conducting major research initiatives on other key areas, such as endocrine-disruption effects of chemicals and potential health impacts of genetically engineered organisms.