Research Priorities for Airborne PARTICULATE Matter in the United States
Despite substantial progress in reducing air pollution
JONATH A N SA MET JOHNS HOPKINS UNIV ERSIT Y R AY MOND WASSEL K. JOHN HOLMES EILEEN A BT K ULBIR BAKSHI NATIONAL RESEARCH COUNCIL BOARD ON ENV IRONMENTAL STUDIES A ND TOX ICOLOGY FOR THE COMMITTEE ON RESEARCH PRIORITIES FOR AIRBORNE PARTICUL ATE MATTER
over the past 30 years, understood health concern that requires further study.
T
he dramatic London fog of 1952 still serves as a reminder of the public health risks associated with air pollution (1). That episode, which caused thousands of deaths, resulted from a pollution mixture with high concentrations of particles and SOx, which were emitted from residential coal burning, industrial processes, vehicles, and other sources. In the ensuing decades, researchers in epidemiology and toxicology have elucidated many of the risks posed by the principal combustion pollutants (particulate matter [PM], SOx, NOx, and carbon monoxide) and by ozone and PM formed in the atmosphere from emitted pollutants. In many developed countries, concentrations of these major pollutants declined to various degrees over the past decades. Many people anticipated that the health risks of air pollution would lessen in parallel (2). However, beginning in the early 1990s, a new wave of epidemiological research indicated that air pollution was still adversely affecting the public’s health despite the substantial progress made over the past 30 years in reducing air pollution (3). The new evidence indicated that PM might be the pollutant most closely linked to the adverse effects. © 2005 American Chemical Society
In this article, we provide a brief overview of the work of an independent National Research Council (NRC) committee on PM. We highlight the committee’s process for developing a research agenda as well as its findings on research progress during its deliberations. We also reflect on the committee as a potential model to provide guidance on a broad research area in which findings may have significant policy implications.
More data, more questions Most of the studies in the 1990s that evaluated associations between changes in health effects (such as deaths or hospitalization counts) and changes in exposure indicators, such as ambient PM concentrations, used time-series analyses. These time-series studies, which showed associations of health outcomes with pollutant concentrations on a short-term basis of one to several days, lacked insights into the extent of the overall public health burden or the extent to which these short-term associations were indicative of significant shortening of life. This uncertainty was reduced, to some extent, by longer-term epidemiological studies, which showed that exposure to air pollution was also associated with increased mortality on time scales of years rather than days. The results of two prospective cohort (group follow-up) studies were particularly prominent—the Harvard Six Cities Study (4) and the American Cancer Society’s Cancer Prevention Study II (5). However, the poorly understood mechanisms of injury by PM and other pollutants at ambient concentrations in U.S. cities were a further barrier to interpreting this evidence.
PHOTODISC
particulates remain a poorly
Nonetheless, in 1997, the U.S. EPA found the evidence to be sufficiently compelling to warrant the promulgation of new National Ambient Air Quality Standards (NAAQS) for fine PM with an aerodynamic diameter of 2.5 µm (PM2.5), while retaining the standards for PM10 (6). The estimated health benefits of meeting these new standards were substantial—avoiding ~15,000 premature deaths per year (7). However, gaps in the scientific foundation for decision making were evident; these helped to make the new PM2.5 standard highly controversial and eventually led to litigation against EPA that was decided in the Supreme Court (8). The court issued a unanimous decision in favor of EPA. The scientific gaps ranged from sources of PM through health effects. For example, the contribution of PM in outdoor air to total personal exposure had not been well characterized, and questions remained about patterns of deposition and clearance from the lung of particles with differing characteristics. Researchers had hypothesized about mechanisms of injury, but little supporting evidence was available and the toxicity-determining characteristics of particles had not been described. Information needed for implementing the PM2.5 standards was incomplete; inventories of PM and PM-precursor emission sources and knowledge of atmospheric processes involved in the transport and transformation of PM also required elaboration. In addition, the new standards were based on the mass of PM2.5, so they continued to treat PM2.5 as though any associated risk to health was similar regardless of sources, chemical composition, and physical characteristics, such as size, shape, and surface area. However, neither PM2.5 nor PM10 are single chemicals. Rather, these particles consist of a rich mixture of materials from various sources. Particles that can be inhaled into the respiratory tract span a range of aerodynamic diameters, from molecular clusters with diameters as small as 0.001 µm to large particles of >10 µm. The numbers of airborne particles and their chemical composition can vary within specific particle size fractions across locations and over time, depending on the types of sources, meteorological and topographical conditions, and the concentrations of various gas-phase pollutants. Particles generated by outdoor sources also penetrate into indoor environments. Indoor particle sources, such as cigarette smoking, insects, molds, and cooking may contribute substantially to total exposure to particles. Figure 1 demonstrates some of the complexity for ambient PM2.5. To change from general, massand size-based PM NAAQS to standards that account for particle composition and heterogeneity would require a substantial deepening of the base of evidence on PM characteristics, sources, and risks to health. In fiscal year 1998, Congress recognized the need for further evidence on PM2.5 and began to fund a major, multiyear expansion of research to reduce uncertainties, almost doubling EPA’s requested PM research budget to nearly $50 million. At the same time, Congress also mandated that EPA arrange for
independent guidance from NRC. An NRC committee was directed to identify the most important research priorities to evaluate, set, and implement NAAQS for PM; to develop a conceptual plan for PM research; and, over the next 5 years, to monitor and evaluate research progress toward better understanding the relationship between PM and its public health effects. Although airborne particles have other important effects, such as reduction of atmospheric visibility, the committee was assembled to focus solely on research related to human health effects of PM. NRC’s multidisciplinary committee of experts met 13 times and held 4 workshops from 1998 to 2004. The committee’s composition reflected the range of scientific disciplines involved in PM research and included those with expertise in the administration and management of scientific research and in air pollution control issues in general. The committee produced reports in 1998, 2000, 2002, and 2004 (9–12). The first two focused on the development of a research investment portfolio and plans for monitoring the progress on this portfolio. The third and fourth reports assessed the progress of this research. During its meetings, the committee heard from a wide array of researchers, policy makers, and stakeholders concerning various aspects of airborne particles and health. In parallel, EPA initiated a substantial PM program that largely followed the research agenda recommended by the committee, funding a total of $368 million for fiscal years 1998–2003. EPA representatives provided several updates to the committee. In addition, the committee organized specialty workshops to discuss the evaluation of research progress on specific topics. The committee used a sources-to-effects framework to develop the research agenda, synthesize available information, and evaluate the extent to which scientific uncertainties were reduced. The committee also developed criteria to assess the value of research to policy making, the quality of the research, and the extent to which the research was effectively planned and widely available.
Examining progress and crosscutting issues As part of its early efforts, the committee developed 10 research topics linked to key policy-related scientific uncertainties (see box on p 302A). EPA and PM researchers nationwide embraced all of these highpriority topics. In addition, the committee reviewed research progress related to those topics in detail in its final report, summarizing the gains in scientific knowledge for each topic from 1998 until the middle of 2002, with some updating as particularly relevant contributions were made (12). Because only a few issues are highlighted here, readers are directed to the full report for a thorough discussion. The committee found a mixed rate of progress on specific topics. For topic 1 (including exposures in indoor environments), substantial new evidence supports the conclusion that ambient PM concentrations are a key determinant of variations in personal exposures. Research also demonstrates that people who experience common respiratory abnormalities,
FIGURE 1
Composition of PM2.5 at representative urban and rural locations in North America The concentration of particles and their chemical composition can vary within specific particle size fractions across locations. Toronto (Canada); Washington, D.C.; Atlanta, Ga.; Mexico City; Los Angeles, Calif.; and Fresno, Calif., were the urban sites sampled. Averaging periods and average PM 2.5 mass are indicated. All sites have at least 1 year of sampling, except Mexico City, for which the average was determined for 14 days. More recent short-term measurements from December 1995 and January 1996 at Fresno, Calif., and Kern National Wildlife Refuge, Calif., show lower PM 2.5 mass concentrations but similar composition to the data displayed here. The Colorado Plateau data are the averages of the Interagency Monitoring of Protected Visual Environments (IMPROVE) sites located at Bryce Canyon, Canyonlands, Grand Canyon, Petrified Forest, Mesa Verde, and Zion National Parks. Adapted with permission from Ref. 13. Sulfate Nitrate Ammonium Toronto (1997–1999) Esther (1995–1999) 12.3 µg/m3 Egbert (1994–1999) 4.6 µg/m3 8.9 µg/m3
Black carbon Organic carbon Soil
Abbotsford (1994–1995) 7.8 µg/m3
Other
St. Andrews (1994–1997) 5.3 µg /m3 Fresno, CA (1988–1989) 39.2 µg/m3
Quaker City, OH (1999) 12.4 µg/m3
Kern National Wildlife Refuge, CA (1988–1989) 23.3 µg/m3
Los Angeles (1995–1996) 30.3 µg/m3
Arendtsville, PA (1999) 10.4 µg/m3
Colorado Plateau (1996–1999) 3.0 µ g/m3
Washington, DC (1996–1999) 14.5 µg/ m3 Yorkville, GA (1999–2001) 16.0 µg/m3
Mexico City– Nezahualcoyotl (1997) 55.4 µg/ m3
Mexico City–Pedregal (1997) 24.6 µg/m3
such as chronic obstructive pulmonary disease, have more PM deposited in their respiratory tracts than those people who have normal lungs (topic 6). Research into some of the broader topics has yielded interesting results, including a statistically robust independent effect of particles in the presence of gaseous pollutants (topic 7); increased risk of health effects in susceptible subpopulations, such as older adults (topic 8); and potential mechanisms of injury (topic 9). In general, the greatest measurable gains have been made on the more-specific research topics with a narrower scope, such as topics 1, 6, and 10 (analysis and measurement). For the broader topics, including topics 7, 8, and 9, substantial uncertainties remain.
Atlanta (1999–2001) 19.8 µg/ m3
Characterization of emissions sources (topic 3) and air-quality models (topic 4) require more research for more effective implementation of PM controls. Research needs under topic 5 remain largely untouched, because they will require a sophisticated integration of air-quality and health research. This research will be critical for refining various PM management strategies, including assessing susceptible subpopulation exposures to toxic PM components (topic 2). The committee also considered the integrated nature of PM research and the major issues that cut across multiple disciplines. Five important issues will likely require significant interdisciplinary collaboration, if the PM research portfolio is to be completed. First, data on adverse health outcomes and
susceptible subpopulations associated with particle exposures must be expanded from focusing primarily on effects on the respiratory system to include those on the heart and reproduction. Second, particle toxicity must be differentiated on the basis of variations in particle characteristics. Third, researchers must shift their emphasis from studies directed primarily at answering whether particles cause particular health effects to those that characterize direct exposure–dose–response relationships. Fourth, PM health effects should be considered within the context of the wide array of other ambient air pollutants; this will help researchers assess whether these pollutants act together to increase risk. Fifth, research advances must be considered when PM NAAQS are set and plans to attain the standards are drawn up. These crosscutting issues provided the committee with a basis for identifying key, overarching challenges for the years ahead.
Future challenges Seven remaining scientific challenges rank as high priority for completion of the PM research agenda, according to the committee. Meeting these challenges may require new research approaches and research management strategies, including an emphasis on multidisciplinary collaboration. Completing PM inventory and models. Improved emissions characterization and air-quality model testing and development are critical for implementing
any PM NAAQS. These tools play a significant role in creating state implementation plans, which detail the steps that state and local governments use to attain NAAQS. Research to improve emission inventories and air-quality models will support more targeted emission-control strategies as progress is made in characterizing hazardous PM components. EPA’s PM research program must recognize the critical role that emissions inventories, ambient monitoring, and airquality modeling play in regulatory decision making. The North American air-quality research community has responded through NARSTO (formerly known as the North American Research Strategy for Tropospheric Ozone) by providing an important assessment of the state of knowledge of those topics and defining research objectives (13). Systematically assessing component toxicity. Developing a program to assess the toxicity of different components of the PM mixture has important implications for the direction of research and for PM-control strategies. Answering the key questions concerning hazardous components of PM requires a carefully coordinated, long-term multidisciplinary research effort. Although substantial research has been conducted, the committee concluded that approaches to assess hazardous PM components are too disparate to converge. A more systematic evaluation of PM by characteristics, components, and health outcomes should complement the previously emphasized investigator-initiated approach to research.
Research priorities and questions recommended by the NRC committee Topic 1. Outdoor measures vs actual human exposures What are the quantitative relationships between concentrations of particulate matter (PM) and gaseous copollutants measured at stationary outdoor air-monitoring sites and the contributions of these concentrations to actual personal exposures, especially for subpopulations and individuals? Topic 2. Exposures of susceptible subpopulations to toxic PM components What are the exposures to biologically important constituents and specific characteristics of PM that cause responses in potentially susceptible subpopulations and the general population? Topic 3. Characterization of emission sources What are the size distribution, chemical composition, and mass-emission rates of PM emitted from the collection of primary-particle sources in the United States, and what are the emissions of reactive gases that lead to secondary formation through atmospheric chemical reactions? Topic 4. Air-quality model development and testing What are the linkages between emission sources and ambient concentrations of the biologically important components of PM? Topic 5. Assessment of hazardous PM components What is the role of physiochemical characteristics of PM in eliciting adverse health effects? Source: Reference 12.
Topic 6. Dosimetry—Deposition and fate of particles in the respiratory tract What are the deposition patterns and fate of particles in the respiratory tracts of individuals belonging to presumed susceptible subpopulations? Topic 7. Combined effects of PM and gaseous pollutants How can the effects of PM be disentangled from the effects of other pollutants? How can the effects of long-term exposure to PM and other pollutants be better understood? Topic 8. Susceptible subpopulations What subpopulations are at increased risk of adverse health outcomes from PM? Topic 9. Mechanisms of injury What are the underlying mechanisms (local pulmonary and systemic) that can explain the epidemiological findings of mortality and morbidity associated with exposure to ambient PM? Topic 10. Analysis and measurement To what extent does the choice of statistical methods in the analysis of data from epidemiological studies influence estimates of health risks from exposures to particulate matter? Can existing methods be improved? What is the effect of measurement error and misclassification on estimates of the association between air pollution and health?
Some of the physiochemical characteristics that may influence toxicity are presented in Table 1. However, as the committee pointed out, a single characteristic is unlikely to predict risk for all health effects associated with human exposure to ambient PM. Enhancing air-quality monitoring. Meeting the most important PM research priorities requires shifting the current air monitoring program from a focus on compliance with the NAAQS toward an approach that uses air monitoring data for various purposes: air-quality forecasting, alerts during episodes, exposure characterization, health and atmospheric studies, source identification, and evaluation of the long-term effectiveness of control strategies. This shift in thinking will require greater use of continuous monitors and compound-specific integrated samples at diverse locations. The same resources currently dedicated to the compliance monitoring network could be used to develop a carefully designed air-quality sampling network, which would provide greater regulatory and research benefits. Investigating long-term exposure. Estimates of the disease burden associated with particle exposure are vital for quantitative risk assessment and cost–benefit analysis. Thus, long-term epidemiological studies, such as the Harvard Six Cities Study (4) and the American Cancer Society’s Cancer Prevention Study II (5), will continue to be principal tools for assessing the public-health burden caused by air pollution. However, the limitations of these studies include the difficulty in characterizing past exposures and in tracking residential and personal histories, as well as the need for ongoing information on long-term exposure to air pollution and health. Therefore, research approaches need to be developed on both new and existing cohorts for follow-up studies. Researchers also need better mechanisms for enrolling and tracking cohorts that include exposures and health outcomes and that will allow for ongoing characterization of the health impacts of long-term exposure to air pollution. Improving toxicological approaches. The committee recognizes the need for complementary toxicological and epidemiological research. Various factors have limited the application of toxicological approaches to human health assessment, including the difficulty of replicating actual inhalation exposures to PM and the inability to readily replicate human diseases associated with increased susceptibility in experimental animals. Another substantial challenge involves reducing the uncertainty due to extrapolating results from animal studies, often carried out at high exposure concentrations, to human populations experiencing lower real-world exposures. Appropriately designed toxicological studies and well-characterized particle samples are needed. Migrating to a multipollutant research program. A challenge to completing the committee’s research agenda lies in the artificial separation of research on PM from studies on air pollution in general. This separation mirrors the regulatory approach of setting NAAQS for six criteria pollutants individually without adequate recognition of their
TA B L E 1
Examples of particulate matter characteristics potentially important to health responses Physical characteristics
Size
Coarse, fine, ultrafine
Surface
Surface-to-mass ratio, physical vs functional surface, biological absorption characteristics
Morphology
Spherical, aggregate, fibrous
Mass concentration
Total mass, size-specific mass, airborne particulate matter mass vs filter-derived mass
Number concentration Charge Physical chemistry
Hygroscopicity, lipophilicity, hydrophilicity Bioavailability Acidity
Solubility in biological media, penetration and distribution
Oxidant potential Surface vs core chemistry
Surface reactions, adsorbed materials
Chemical components
Metals
Transition vs other valence states
Carbon
Elemental, temperature-resolved fractions, organic (by class and species), semivolatile (particle and vapor partitioning), adsorbed volatile organic compounds
Biogenic
Antigens, microorganisms, toxins (endotoxin and other), plant and animal debris
Secondary inorganic aerosols
Sulfates, nitrates
“Dusts”
Crustal minerals (crystalline state), street dust (tire, brake, and road wear)
Particulate matter as a Interactions with other pollutants component of air pollution (other than additive) and with other environmental variables (such as weather) Source: Reference 12.
interrelationships in determining risks to health. Although the PM research agenda has appropriately considered the role of other pollutants, it has been directed at PM out of necessity. The committee acknowledged a critical need to shift from the current single-pollutant focus to a multipollutant focus. Such a multiple-air-pollutant program could help produce information relevant to the setting of
standards and the development of air-quality management strategies. Integrating across disciplines. The committee recognized the need for complementary evidence on PM from multiple disciplines and called for interdisciplinary research and the establishment of PM research centers to foster collaboration. Expanding multidisciplinary strategies and programs is essential to more effectively address important uncertainties linking PM sources to health effects and to implement a multipollutant approach for airquality management.
Overview of committee’s contributions The committee seized the opportunity to make substantial scientific contributions to the nation’s PM research during a time of controversy over the new NAAQS for PM2.5. The committee’s efforts resulted in the development of an organizing framework and research portfolio, including cost estimates for EPA and other organizations to fund long-term PM research; fostering of interdisciplinary discussions on the key topics; appointment of an EPA manager of PM research to coordinate the agency’s diverse array of activities; impetus for all funders of PM research to coordinate activities; and an evaluation of research progress at the end of five years. There were also indirect consequences. The committee’s reports motivated interdisciplinary discussions within EPA and encouraged collaborations, particularly between scientists in the areas of ambient monitoring and health effects. The research cost estimates from the committee not only justified continued funding for PM research by EPA but also challenged the agency to follow the committee’s recommendations or to justify departures. Also, the committee’s recommended PM research plan, which integrated short- and long-term phasing of studies over a 13-year period, provided a model for EPA’s 16 research plans now in place, which cover various environmental topics. EPA used the committee’s approach to develop multiyear plans with short- and long-term goals, as well as performance measures to respond to the requirements of the new Government Performance and Results Act. Is there value in maintaining the role played by the committee for future PM research efforts? In its fourth report, the committee answers in the affirmative by calling for continued independent review and oversight of the PM program and the broader multipollutant research program it has advocated. An independent and interdisciplinary effort will be important to ensure that future research investments are sound. Such a role may also be broadened to other complex environmental problems that need interdisciplinary research. Jonathan Samet is a professor and the chair of the department of epidemiology in the Johns Hopkins University’s Bloomberg School of Public Health. Raymond Wassel, K. John Holmes, Eileen Abt, and Kulbir Bakshi are senior staff officers for the NRC’s Board on Environmental Studies and Toxicology.
Acknowledgments We wish to acknowledge the members of the committee who sustained six years of meetings and workshops and wrote four reports. Members were Jonathan Samet (chair) and Ronald White, Johns Hopkins University; Judith Chow, Desert Research Institute; Bart Croes, California Air Resources Board; Robert Forster, University of Pennsylvania School of Medicine; Daniel Greenbaum, Health Effects Institute; Philip Hopke, Clarkson University; Petros Koutrakis, Harvard University School of Public Health; Daniel Krewski, University of Ottawa (Canada); Paul Lioy, Robert Wood Johnson Medical School; Joe Mauderly, Lovelace Respiratory Research Institute; Roger McClellan, Chemical Industry Institute of Toxicology (retired); Gunter Oberdörster, University of Rochester School of Medicine; Rebecca Parkin, George Washington University; Joyce Penner, University of Michigan; Richard Schlesinger, Pace University; Frank Speizer, Harvard University Medical School; Mark Utell, University of Rochester Medical Center; Warren White, University of California, Davis; Ronald Wyzga, Electrical Power Research Institute; and Terry Yosie, American Chemistry Council. Glen Cass served as a member of the committee until his untimely death in 2001. NRC project staff were Raymond Wassel (project director), James Reisa (director, Environmental Studies and Toxicology), Eileen Abt, Kulbir Bakshi, K. John Holmes, Karl Gustavson, Amanda Staudt, Ruth Crossgrove, Rachel Hoffman, Mirsada Karalic-Loncarevic, Emily Brady, and Sammy Bardley.
References (1) U.K. Ministry of Health. Mortality and Morbidity during the London Fog of December 1952; Reports on Public Health and Medical Subjects No. 95; London, UK, 1954 (abstract available at www.bopcris.ac.uk/bopall/ ref 9517.html). (2) National Research Council Committee on Air Quality Management in the United States. Air Quality Management in the United States; National Academies Press: Washington, DC, 2004. (3) Dockery, D. W.; Pope, C. A. Particles in Our Air: Concentrations and Health Effects; Harvard University Press: Cambridge, MA, 1996. (4) Dockery, D. W.; et al. An Association between Air Pollution and Mortality in Six U.S. Cities. N. Engl. J. Med. 1993, 329, 1753–1759. (5) Pope, C. A., III; et al. Particulate Air Pollution as a Predictor of Mortality in a Prospective Study of U.S. Adults. Am. J. Resp. Crit. Care 1995, 151 (3 Pt 1) 669–674. (6) EPA. National Ambient Air Quality Standards for Particulate Matter: Final Rule. Fed. Regist. 1997, 62, 38,651– 38,760. (7) Criteria Air Pollutants: Particulate Matter. Health Effects Research, Progress Report; EPA Office of Research and Development, National Health and Environmental Effects Laboratory: Research Triangle Park, NC, 1997. (8) Whitman v. American Trucking Assns. 2001, 531 U.S. 457, 464, 475–476. (9) NRC Committee on Research Priorities for Airborne Particulate Matter. Research Priorities for Airborne Particulate Matter: I. Immediate Priorities and a Long-Range Research Portfolio; National Academies Press: Washington, DC, 1998. (10) NRC Committee on Research Priorities for Airborne Particulate Matter. Research Priorities for Airborne Particulate Matter: II. Evaluating Research Progress and Updating the Portfolio; National Academies Press: Washington, DC, 1999. (11) NRC Committee on Research Priorities for Airborne Particulate Matter. Research Priorities for Airborne Particulate Matter: III. Early Research Progress; National Academies Press: Washington, DC, 2001. (12) NRC Committee on Research Priorities for Airborne Particulate Matter. Research Priorities for Airborne Particulate Matter: IV. Continuing Research Progress; National Academies Press: Washington, DC, 2004. (13) NARSTO. Particulate Matter Assessment for Policy Makers: A NARSTO Assessment ; McMurry, P., Shepherd, M., Vickery, J., Eds.; Cambridge University Press: Cambridge, U.K., 2004.
