Research Protocol
	Status of the Program

General Research Plan

The Center's research strategy is to analyze a composition-concentration-health response database to determine contributions of individual physical-chemical air contaminant species to the health effects of mixtures of air pollutants from common sources. The database is being constructed by combining results from laboratory inhalation studies of common source emissions. The use of a common protocol for all studies enables the aggregation of results into a single database that encompasses several mixtures of different, but overlapping composition. This plan addresses the lack of an epidemiological or laboratory database allowing statistical analysis of the roles of individual air contaminants and their combinations in the different effects of inhaling complex pollution mixtures. A secondary, but important, goal is to make direct, detailed, contemporary comparisons among the health effects of common source emissions of public health and regulatory concern. Results of individual studies and comparisons among source emissions are being published while the database is under construction.

The initial series of studies will provide useful information, but will also provide a test of the utility of expanding the exposures to include additional air contaminants. There are innumerable atmospheric reaction products, pollens, molds, endotoxins, infectious agents, and other natural and man-made environmental air contaminants whose potential effects, or interactions with other pollutants, need to be understood and placed in context. There may also be important interactions between environmental air pollutants and other exposures in the workplace and home. Many toxicants, irritants, and sensitizers encountered in the “personal environment” could contribute to susceptibility to air pollutants in the outdoor environment.

Progress

The construction of the database is underway. Exposures to simulated on-road contemporary (circa 2000) diesel engine emissions and hardwood (oak) smoke are completed. Exposures to simulated on-road gasoline engine emissions (circa 1996 engines using current national average fuel) are underway. Exposures to paved road dust and simulated “downwind” coal combustion emissions are planned, and the exposures are being developed. As the database is being built, strategies for its analysis are being identified. The status of the research program is described in more detail elsewhere.

RESEARCH PROTOCOL

The general experimental design is described in the Research Strategy section. The complete text of the hardwood smoke study protocol is available as an example of a detailed study protocol. Protocols for other studies are identical except for methods for generating the exposure atmospheres, designation of the primary dilution indicator pollutant used to set exposure concentrations, and the target concentration levels of the indicator. The general protocol is described below.

Exposures

Animals are exposed to the atmospheres by inhalation in whole-body chambers for 6 hours/day, 7 days/week. The length of exposure varies among the health assays from 3 days to 6 months. The study of most atmospheres include five treatment groups, one exposed to each of four concentrations (log dilutions) of the test atmosphere and one clean air negative control group. The gasoline emission exposure includes three levels and control, with both filtered and unfiltered atmospheres at the highest level. Because the exposure-response characteristics of the health effects are of interest, it is desirable to select concentrations producing measurable effects at a minimum of two exposure levels. However, it is also desirable to avoid extreme exposures that bear no relationship to plausible human exposures. It is not a problem, and may be an advantage, if the lowest exposure concentration proves to be a "no observed adverse effects" level. The exposure concentrations and the specific pollutant used to set concentrations (e.g., particle concentration) may differ among the atmospheres. In general, the highest exposure concentration will mimic extreme, but still plausible, human exposures, and the lowest concentration will fall within the range of non-occupational environmental exposures. In practice, the lower bound of exposure may be limited by the ability to control and measure low concentrations against the background of air contaminants produced by the animals themselves.

Analysis of Exposure Atmospheres

The composition of the exposure atmospheres is characterized at the greatest practical level of detail to allow analyses of the relationships between health effects and specific physical-chemical species and classes of air contaminants. Identical analyses are applied to all atmospheres, even though not all chemical species will be present in each atmosphere. This approach parallels the use of identical health assays for all atmospheres, regardless of predicted outcome.

The analyses include several hundred individual analytes and parameters, falling within the classes listed in the table below. Particle analyses include mass and number concentrations, size distributions, morphology, size-specific chemistry, and the mass fraction of adsorbed compounds. Mass concentrations of gases, volatile hydrocarbons, and semi-volatile organics are measured. Particle-bound and vapor-phase semi-volatile organic compounds are speciated in detail by chemical class and individual species. Inorganic ions and elements are also measured.

Some analytes are measured daily. Detailed speciation is typically conducted twice per exposure chamber per study and averaged results (if the two measurements agree) are used to describe the exposures.

Health Assays

A standardized suite of “core” health assays is used in studies of all atmospheres. The assays were selected on the basis of recommendations from a 1999 workshop. The criteria for selection included: 1) coverage of respiratory and cardiac health outcomes suggested by population studies; 2) accepted interpretive value; 3) accepted use for hazard assessment; and 4) availability of established methods. More exploratory assays are also applied, but are not required to be consistent from study to study.

STATUS OF THE PROGRAM

The NERC program was begun on April 1, 1998 with receipt of initial EPA funding. During the first five years (through March 30, 2003 ), the Center was organized, the External Scientific Advisory Committee was established, five pilot studies were conducted and published, the research strategy was formulated, the core study protocol was developed, multiple experimental design workshops were conducted, a quality assurance management plan and a data policy were developed, the first two core exposures (diesel emissions and hardwood smoke) were specified and completed, the third exposure (gasoline emissions) was specified and preparations for exposure were begun, the basic parameters for the coal emissions exposure were determined, and several papers were published.

Current activities and results are described in the 2005 annual Center report, available on the "Center News" page. The third exposure, simulated on-road gasoline engine emissions, is underway and will be completed in January 2006. Work in preparation for the fourth and fifth exposures is currently underway. The fourth exposure, simulated "downwind" coal combustion emissions, will begin in late summer 2006. The general parameters of the target exposure atmosphere have been defined, and trials are underway to achieve the proper composition using Powder River Basin low-sulfur sub-bituminous and Central Appalachian low-sulfur bituminous coals. The coal for the main study will be selected on the basis of results of the generation trials. It is expected that the fifth exposure, resuspended paved road dust, will follow the coal emissions exposure. In view of the lack of information on the range of composition and toxicity of road dust, a substantial pilot study effort is underway. Dusts have been collected from active traffic surfaces of paved roads in multiple locations in several cities and rural areas. The PM10-2.5 and PM2.5 fractions are being separated and analyzed chemically, and toxicity is being compared using in vitro assays. This work will provide a basis for selecting exposure material for the main study.

Results are published as data sets (i.e., data from a set of health assays or atmospheric analysis) are completed, audited, and analyzed. Multiple papers result from each exposure study, and each is published when ready, regardless of the status of other data from the study. Data are made available to the public via this web site for information and further analysis after the papers are published.

More detailed information on the program's status and accomplishments can be found at the following locations:

•  2004 Annual Progress Report

•  Final Report to EPA for First 5 Years of Center Funding

•  Publications

•  Center Quality Management Plan