Grass seed field smoke and its impact on respiratory health.

Introduction

Despite 30 years of protests by local citizens claiming that the burning of thousands of acres of grass in Eastern Washington and Northern Idaho has negatively affected their health, grass field burning continues. Recently, pressure to alter the practice has grown as a result of a convergence of concerns among several diverse political, social, and economic groups. These groups are concerned about the potential health impacts of grass smoke, as well as about economic impacts on tourism. They include an area Chamber of Commerce, the Spokane County Medical Association, the American Lung Association, the Washington Environmental Council, an extremely active Spokane citizens' group, and the Spokane County Air Pollution Control Authority (SCAPCA) (1-5). Given the numbers of opponents, why does the practice of grass field burning continue? In part, the burning continues because no research has been undertaken to identify its potential health impacts. Furthermore, Kentucky bluegrass growers have not found any alternative methods of stimulating grass seed production that are as effective as field burning. Without studies identifying the health hazards of grass smoke, and with grass seed sales constituting a $75 million industry in the inland Northwest, economics have taken precedence over health concerns. In fact, the right to burn grass seed fields was written into a 1991 revision of the Washington State Clean Air Act (6).

In early discussions of the health effects of grass field burning, the president of the Inland Grass Growers Association (IGGA) stated that grass smoke was 80 percent steam and presented no health hazards to area residents, unlike wood smoke, which contains harmful airborne particulates. Opponents contend that the smoke produced by burning grass results in harmful airborne particulates and that grass burning is a virtually unregulated practice, unlike wood burning (7).

Airborne particulates are measured by diameter in microns (i.e., [PM.sub.2.5] refers to particles equal to or less than 2.5 microns [[micro]meter] in diameter). Particles 10 [[micro]meter] in diameter or more ([PM.sub.10]) do not remain airborne for long and are usually filtered by the nasal passages. The smaller the size of the particulates (especially 5 [[micro]meter] and under), the longer the particulates circulate in the air, and the deeper they penetrate into the alveolar cells of the lungs.

In addition to particle size, health concerns relate to the chemicals that are usually associated with the process of incomplete combustion (8,9). It has been shown that the concentration of polycyclic aromatic hydrocarbons (PAHs) in suspended particulates is inversely correlated to particle size and that 47 to 70 percent of PAH in smoke is contained in the [PM.sub.2.0] fraction (10,11).

The importance of air pollution in the pathogenesis of bronchial asthma and other pulmonary diseases has been of interest for decades. Considerable epidemiological evidence implicates fine-particulate air pollution, especially [PM.sub.10] and [PM.sub.2.5], as a trigger that exacerbates respiratory conditions in some individuals with asthma and chronic obstructive pulmonary diseases (12-16). The biological mechanism for these effects is as yet unknown, but [PM.sub.10] particulate matter has been demonstrated to have free radical activity and proinflammatory effects in lung tissue in vivo and in vitro, as well as immunosuppressive properties (17,18).

The pro-inflammatory effects appear to negatively affect individuals suffering from a variety of respiratory conditions such as asthma, chronic bronchitis, cystic fibrosis, and emphysema. Individuals whose respiratory conditions include a reactive airway disease (RAD) component appear to be most negatively affected. RAD is often diagnosed as chemically induced asthma and may account for up to 30 percent of diagnosed asthma cases. Chemically induced asthma occurs in response to triggers such as chemical irritants present in the smoke produced by the open burning of biomass (19-24).

Open burning of biomass is a technique commonly used worldwide both for the disposal of crop and forest debris and for land preparation. This agricultural practice is known to produce significant amounts of volatile organic compounds (VOCs) such as phenols, as well as PAH either in a gaseous aerosol or adsorbed onto inert particulate matter (19,25). Currently, farmers in Eastern Washington and Northern Idaho burn thousands of acres of biomass, particularly Kentucky bluegrass.

In the early 1990s, SCAPCA introduced a cap on burning acreage for Spokane County and reduced the number of permissible burn days. These regulations did not, however, alter the numbers of acres of grassland burned in the county The growers successfully argued against stronger limits by pointing out that previous studies on the negative impacts of smoke have been done on wood smoke, not grass smoke, and are therefore not applicable. At public hearings, growers dismissed individual claims of damage as anecdotal and unreliable. They argued that those who claimed to be negatively affected by grass smoke were already hypersensitized and that the true impacts of burning thus could not be clearly defined. In cases in which clear evidence of worsening health occurred, such as in extensive hospitalization of cystic fibrosis patients during burns, they attributed the sudden increase in disease severity to the unpredictable and progressive nature of the underlying disease (7,26). By the summer of 1995, local clean-air citizen groups, a small group of physicians, and even SCAPCA had met with little success in their quest to eliminate or restrict the practice of grass field burning.

TABLE 1

A Comparison of Spokane County Population Data and Survey
Demographics for 1995

                             1995 Census Data   Survey Demographics

Population                        401,205              1,850
White                          93.4 percent         95 percent
Black/African American          1.4 percent        0.6 percent
Hispanic                        1.9 percent        1.5 percent
Asian/Pacific Islander          1.0 percent        1.8 percent
Native American                 0.1 percent        1.5 percent
Median Age                           44                 42
Poverty 18 years and under           NA           20.8 percent

Starting late in 1995, a series of events occurred that eventually led to more stringent limitations on grass burning in Washington state. The Spokesman Review became the battleground between the grass growers and their association and the area citizens' group. Numerous articles, letters to the editor, and editorial comments were published, both pro and con. In 1996, the Natural Resources Defense Council reported that particulate levels in Spokane County placed its air quality levels among the 10 worst in the country (22). Early in 1996, the Spokane County Medical Society and the Washington Thoracic Society issued statements charging that grass seed field burning represented a public health threat. Local business owners and university researchers added their voices to the fray as well. Once again the testimony of the doctors was dismissed by the grass growers as anecdotal. Nevertheless, when combined with epidemiological research citing correlations between increased morbidity levels and increases in [PM.sub.2.5], the medical testimony forced politicians to acknowledge that a problem did exist. In April 1996, the Washington Department of Ecology (DOE) issued an emergency order stating that grass field burning posed a significant health hazard. The DOE order required that burning be phased out over a three-year period (7).

TABLE 2

Preliminary Identification of Phenols and PAHs in Smoke from the
Burning of Kentucky Bluegrass

PAHs                         Phenols

Acenaphthene                 4'-Methoxy Actophenone
Anthracene                   2,6 Dimethoxy Phenol "Syringol"
Benzo[Alpha]pyrene           4-Methoxyphenol
Chrysene                     2-Methylphenol
Pyrene                       Phenol
Phenanthrene

In 1997, DOE reworded the original order from a complete phase-out to a partial burn reduction over a two-year period. The revised rule reduced burning by 66 percent of the acreage burned in 1995, with an exemption for areas with steep slopes (27). Two legal attempts by farmers to overturn this order failed. The farmers are now suing DOE, arguing that the methods used by DOE to institute the reduction were flawed. Some growers, having seen the writing on the wall, have shifted their operations a few miles to the east, into nearby Northern Idaho or onto reservation lands. So, although the battle appears to be going in favor of health proponents in Washington state, it is far from over; smoke from the Idaho and reservation fields still blankets the region (28).

This study hypothesized that the high levels of particulates (as quantified by [PM.sub.10] and [PM.sub.2.5]) would result in higher levels of respiratory diseases, increased hospitalization rates for asthmatics, and an increase in medication purchases during the burn season. As the exact causative agents of the expected medical effects remain unknown, this study further hypothesized that the smoke produced by the incomplete combustion of organic matter (grass) contained chemical irritants.

Materials and Methods

Respiratory disease rates among adults were assessed in Spokane County. The database used to assess the disease rate in the adult population was obtained from an early 1995 Spokane County Regional Health Department survey. The survey results were evaluated to establish the percentage of individuals in the county affected by asthma, chronic bronchitis, and emphysema. Before this 1995 survey, neither the state nor the county had assessed specific respiratory disease rates. In an attempt to better understand what the data might mean, the results were compared with national data. The 1995 national disease data for asthma, chronic bronchitis, and emphysema are unavailable, so the 1995 disease estimates for asthma were derived by estimating the rate of increase from 1992 disease rates to 1994 disease rates (29-31). The final estimate was derived only from 1993 and 1994 data as the World Health Organization's International Classification of Diseases numeric asthma code has changed. The codes for emphysema and chronic bronchitis have remained the same, so the 1995 estimates for these diseases were based on data from 1990 through 1994. The authors found that these rates were very stable.

The SF12 and SF36 questionnaires, distributed by the Medical Outcomes Trust, were used as the county survey instrument. This survey has been thoroughly tested for reliability and validity; one study specifically examined the survey's asthma questions for validity (32). The survey was randomly administered throughout the county and was completed by 71 percent (1,850) of the subjects. The respondents' demographic profiles matched those of the county almost identically (Table 1). A Spokane County Regional Health Department representative felt that the survey results could be applied analogously to children (33). Prior research has shown, however, that adult rates are lower than those of children, so the authors attempted to identify the asthma rate for children in the county (34).

The authors conducted phone surveys of the area public schools and preschools. When re-calling sites to recheck the survey data, the authors found that the data obtained from many of the schools and daycare centers was not reliable. The authors were able to establish asthma rates for low-income children attending Head Start. These programs had excellent health collection instruments and processes for obtaining reliable data on the children in their programs. As that rate can be accurately applied only to low-income children (20.8 percent) residing in Spokane County, it was not possible to develop a rate for all of the children in the county (35).

Hospitalization rates for asthma were established with data from the Comprehensive Hospital Abstract Reporting System (CHARS). These data are collected by the state health department from all state-licensed hospitals except federal institutions. Records are kept by hospitalization, not by individual, and do not include Washington State residents who are cared for in another state. The data contain several fields into which the International Classification of Diseases asthma code can be entered (36).

An additional data set used in this study consisted of two years of asthma-related drug purchase records drawn from the pharmacy database of Group Health Northwest (GHNW). GHNW is a managed care organization with approximately 160,000 members. It contracts with individuals, private and public businesses, the federal government (the military and Medicare), and the state (Medicaid and two special sliding-scale programs for individuals who would otherwise be uninsured) (37). The county survey revealed that 10 percent of the population remains without medical insurance and thus would not be included in this database.

The analysis and quantitation of the compounds was carried out with a Hewlett Packard gas chromatograph/mass spectrometer (GC/MS) (HP5890II/HP5971A) controlled by an HPCHEM software program. Individual phenols and PAH were determined by correlations with a National Bureau of Standards mass spectrometry library and by comparison with authentic and commercially available phenols (from Sigma) and PAH (from Supelco).

Data Collection

The county survey contained two questions that pertained directly to respiratory health. The first question asked if the respondent had been diagnosed with asthma. The purpose of this question was to ascertain the childhood asthma rate. The second question asked if the respondent had been diagnosed with emphysema or chronic bronchitis.

Pharmacists from GHNW identified four drugs (Beclovent, Intal, Azmacort, and Ventolin) as strongly associated with the treatment of asthma. GHNW's pharmacy provided the study with two years of purchase data for these four drugs. The correlation between the purchase patterns for these drugs and [PM.sub.10] and [PM.sub.2.5] levels were examined statistically. First, the daily number of drugs purchased was compared with the [PM.sub.10] and [PM.sub.2.5] levels in the air, then nonstandard weekend purchase totals were separately compared with [PM.sub.2.5] levels [ILLUSTRATION FOR FIGURES 1 AND 2 OMITTED]. In 1994 and 1995, grass fields were burned on 12 nonconsecutive days. These burn days occurred over a period of about one to two months; the number of acres burned varied, as did the location, wind speed, and direction. These facts, along with other confounding variables, made more detailed statistical analysis unreliable.

The hourly and daily [PM.sub.10] and [PM.sub.2.5] data were collected by SCAPCA with tapered-element oscillating microbalances (TEOMs) (Rupprecht & Patashnick series 1400a) to continuously monitor the particulate loading in the air. These instruments determine the mass loading on a Teflon-coated, quartz-fiber filter (15 millimeters in diameter). A U.S. EPA-equivalent method was used for mass analysis of [PM.sub.10]. For the [PM.sub.2.5] measurements, a calibrated cyclonic separator is positioned upstream of the flow splitter.

Two sampling methods were used to concentrate the smoke. In the initial approach, smoke was collected during the burning of Kentucky bluegrass fields in southern Spokane County in the fall of 1991. The air samples containing smoke were drawn over charcoal (130 mg) in glass tubes (from SKC) by battery-driven personal sampling pumps (also from SKC) at a rate of 5.0 liters per hour. The samplers were placed at distances ranging from i to 8 kilometers from the combustion sites for periods ranging from 24 to 48 hours.

A later study that involved a controlled open burn of Kentucky bluegrass field residues used a different sampling technique. Smoke from this controlled burn was sampled at a rate of 5 liters per minute (L/min) with a Lane County sampler situated i meter from the fire. The smoke was trapped with a polyurethane foam (PUF) plug from University Research Glassware. Twenty-five-minute samples were taken of all of the smoke and of smoke with a particle size 2.5 [[micro]meter] or less ([PM.sub.2.5]).

In the controlled burns, the charcoal, with the smoke compounds adsorbed, was transferred from the glass tubes to an empty stainless steel tube (25 cm long by 3 mm in internal diameter). The charcoal was immobilized in the tube by acid-washed silanised glass wool. The steel tube was placed in the thermal desorption trap position in a purge and trap (HP 7965) sampler. Using an HP software program, the trap was heated for 4 minutes at 300[degrees]C, during which time helium (at 30 pounds per square inch) was passed through the trap. The thermally desorbed compounds were automatically transferred from the charcoal to an RTx capillary column (from Restex) in a gas chromatograph (HP5890 II). The capillary column was 105 m long and had an internal diameter of 0.32 mm. The PUF plugs used in the sampling of the controlled burns were heated to 80 C for 15 minutes in 25 mL of a polar organic solvent, acetonitrile, and then filtered. This extract was reduced in volume to 0.2 mL by evaporation under pressure. Finally, 2 [[micro]liter] of this solution was injected into the gas chromatograph with a DB-624 capillary column (from J&W Scientific) 30 m long by 0.32 mm in internal diameter.

Results

Chronic Bronchitis and Emphysema

The results of the county survey indicated that 7.4 percent of the population in Spokane has chronic bronchitis or emphysema (1995); the 1995 estimated national rate is 6.2 percent, a rate that has remained constant since 1990 (34). This difference was significant (p = .013) and will result in 4,013 excess disease cases in the county. Although smoking is strongly associated with these diseases, a Regional Health Department representative stated that the percentage of smokers (23.9 percent) does not exceed national levels (25 percent) (33).

Asthma

After identifying the percentage increase in asthma from 1993 to 1994 (0.7 percent), the authors added this figure to the 1994 data and estimated that 5.8 percent of the nation's adults had asthma in 1995 (30,31). A Chi square analysis of the actual rates of adult asthma in Spokane County in relation to the expected rates resulted in a p value of less than .01. To assess the number of potential excess asthma deaths in the county, the total number of 1994 asthma deaths, 2.1 per 100,000 population, was compared with the overall asthma rate (5,610 per 100,000 population) to establish the percentage of asthmatics who died in 1994 (37). By dividing the number of deaths per hundred thousand into the number of asthmatics per hundred thousand, it was possible to calculate the expected percentage of deaths among asthmatics: 0.0374 percent. This figure was then multiplied by the number of excess disease cases among adults (15,341) in the county. The results revealed that expected excess deaths per year would be 5.74.

Death Rates for Asthma

A 1993 study of death certificates suggests that the base figure used for calculating the numbers of asthma deaths per year may be too low. That study found that asthma may be greatly under-reported as a cause of death; 16 percent of the deaths among asthmatics were caused by asthma but were miscoded (38). If this figure is confirmed in future studies, the actual number of deaths caused by asthma in the general population will turn out to be higher than the 2.1 per 100,000 figure reported by Sly and O'Donnell (37).

Hospitalization Admission Rates for Asthma

A review of the CHARS data for 1995 showed that hospitalizations for asthma in Spokane County exceed the state levels. The first data are recorded in several formats and are reported as the number of cases per hundred thousand population (e.g., 86.1 per 100,000). The first format is for unduplicated admissions in which asthma is the primary diagnosis. (Unduplicated admissions do not count the individual again if he or she is readmitted during the year with the same primary diagnosis.) Spokane's rate was 91.7 admissions per 100,000; the state average was 86.1. Duplicated admissions with asthma as the primary diagnosis for Spokane were 111.9 per 100,000 in 1995; the state average was 102.3. This study also examined hospitalization through duplicated admissions in which asthma showed up in any of the diagnosis fields. For example, if a patient was admitted for heart surgery but needed an asthma treatment during the course of the admission, asthma would then show in one of the diagnosis fields. The rate for Spokane County for 1995 was 371.7 per 100,000; the average rate for the state was 301.4.

Pharmaceutical Use Patterns

An analysis of daily purchase data during the burn season showed a positive but not significant correlation between increases in [PM.sub.2.5] levels and drug purchases. No correlation was found between changes in particulate levels and drug purchase patterns during the weeks before grass burning. The weekend drug purchases were then compared to PM levels.

An analysis of mean nonscheduled weekend asthma-related drug purchases during burn season produced correlation coefficients of 0.94 (n = 4) in 1994 and 0.96 (n = 6) in 1995; these coefficients, when compared with [PM.sub.2.5] levels, yielded a significance level of p [less than] .01 [ILLUSTRATION FOR FIGURES 1 AND 2 OMITTED]. [PM.sub.10] levels and 1995 purchase patterns returned a positive but not significant correlation of 0.67. An analysis of weekend purchase patterns and [PM.sub.2.5] levels one month before the burning period did not indicate any clear correlation between purchase patterns and either [PM.sub.10] and [PM.sub.2.5] levels. These analyses returned correlation values of 0.1 and -0.2 respectively.

The chemical analysis of the smoke samples revealed at least five different phenols in the smoke produced by combustion of Kentucky bluegrass residues in controlled open burning. Six PAHs were identified as components in air samples collected 1 to 5 kilometers from where the uncontrolled open burning of grass seed fields was conducted (Table 2).

Discussion

Lack of existing data limited this study in a number of ways. The first limitation stems from the fact that no previous attempt has been made to accurately ascertain disease rates at the local and state levels. This lack of state and local data limits the authors' ability to identify disease trends. Although national respiratory disease rates for 1995 have not yet been calculated, it was possible to estimate 1995 rates by examining the rates from previous years. Asthma rates for 1995 were estimated from 1993 and 1994 data.

Another limitation stems from the fact that purchase patterns do not necessarily reflect use patterns. Most inhalers are designed to last at least one month, and many long-time asthmatics tend to stock up. Also, some doctors acknowledged premedicating sensitive patients in anticipation of adverse reactions to the smoke. Given these facts, the authors speculated that purchase patterns would not reflect the actual use patterns of all asthmatic individuals. Discussions with two GHNW pharmacists led the authors to conclude that weekend drug purchase data might provide insights into some of the health effects of increases in [PM.sub.2.5]. The pharmacists felt, based on their experiences, that many weekend drug purchases occur because of unplanned or emergency needs and newly diagnosed cases. The authors hypothesized that if grass field smoke were not affecting health, they should find no change in the purchase levels for emergency or unplanned needs or for newly diagnosed conditions during periods of increased particulate levels caused by grass burning.

Although many variables may contribute to the respiratory rates in Spokane County, the data described in this paper suggest that the county and the region face a number of health-related problems that appear to be associated with the high levels of fine particulates ([PM.sub.2.5]) produced by grass field burning. For individuals with chemically induced asthma (RAD) triggered by grass seed field burning, the regional nature of the practice continues to pose a health risk. A 1996 study found that 10 percent of the residents in eastern Washington and northern Idaho reported needing to purchase more medical care and supplies during field burning and eight percent were forced to leave the area entirely (21). For most of these individuals, asthmatic symptoms rarely result in death. For cystic fibrosis patients with RAD, however, the cumulative effects of exposures eventually can lead to respiratory failure and death (20).

An explanation for the way grass smoke appears to act as a trigger can be found in the Material Safety Data Sheets (MSDS). The MSDS state that all of the phenolic compounds that were identified have acute effects and "may be harmful by inhalation, ingestion, or skin absorption, may cause eye and skin irritation" (39). In addition, some of these phenols are "irritating to mucous membranes and the upper respiratory tract; depending on the intensity and duration of the exposure, the effects may vary from mild irritation to severe destruction of tissue" (39). The PAHs found are generally considered to be carcinogenic, although some (e.g., pyrene and acenaphthene) are also listed as being irritants to mucous membranes and eye, skin, and lung tissue (40).

The high county rates of adult asthma and hospitalization of asthmatics, as well as the drug purchase patterns, suggest an association between the [PM.sub.2.5] levels produced by grass field burning and morbidity. It would appear that the fine particulates produced by combustion and the chemical irritants in the smoke have combined to negatively affect the health of individuals suffering from RAD. Of course, further study will be necessary to ascertain if any other factors are acting in concert with the smoke to produce these results.

Some individuals believe that the newly proposed [PM.sub.2.5] particulate standards suggested by the U.S. Environmental Protection Agency (U.S. EPA) will solve agricultural burning problems of this type. The proposed U.S. EPA daily average standard for [PM.sub.2.5] is 65 micrograms per cubic meter ([[micro]gram]/[m.sup.3]). Farmers tend, however, to restrict their burning activities to a few hours each day. This pattern would only produce transient increases in particulate levels. Since [PM.sub.2.5] is considered to contribute 30 to 90 percent of [PM.sub.10], the authors believe that average daily [PM.sub.2.5] levels will remain below the new daily standard (39). For example, in 1995, a review of hourly [PM.sub.10] data during open field burning near Spokane produced extremely high hourly [PM.sub.10] values of between 145 and 246 [[micro]gram]/[m.sup.3], but only resulted in a [PM.sub.10] 24-hour average of 60 pg/[m.sup.3]. Calculated at 30 to 90 percent of [PM.sub.10], the average daily [PM.sub.2.5] level remains below the new standard. Thus, it would appear that the new standard, as currently proposed, will have no effect on the regulation of open field burning. It is therefore imperative that environmental health researchers evaluate the health impacts of agricultural burning even when particulate standards are not exceeded.

Acknowledgements

The authors thank Nancy Birch, Ph.D., for her assistance with the statistical interpretations and doctors McCarthy, Chestnut, and Whitehouse for their willingness to commit to their patients' continued health. Assistance in gathering and understanding the data used in this study was provided by Torney Smith of the Spokane County Health Department; pharmacists at Group Health Northwest; Ron Edgar of the Spokane County Air Pollution Control Authority; Spokane's Headstart Programs; and Patricia Hoffman of Save Our Summers.

REFERENCES

1. Marcella, N. (1996), "Economic Impacts of Smoke," Symposium on Grass Seed Field Burning, Olympia, Wash.: Washington State Department of Ecology.

2. Whitehouse, A. (1996), "Health Impacts of Smoke," Symposium on Grass Seed Field Burning, Olympia, Wash.: Washington State Department of Ecology.

3. Bucklin, Y. (1996), "Impact of Smoke on Citizens," Symposium on Grass Seed Field Burning, Olympia, Wash.: Washington State Department of Ecology.

4. Hoffman, P. (1996), "Alternatives and Economics," Symposium on Grass Seed Field Burning, Olympia, Wash.: Washington State Department of Ecology.

5. Rudbeck, M. (1996), "Economic Impacts of Smoke, Kootenai County, ID," Symposium on Grass Seed Field Burning, Olympia, Wash.: Washington State Department of Ecology.

6. Washington State Clean Air Act (Amended May 1991), RCW 70.94.650.

7. Roberts, R. and R. Zinke (1996), "A Case Study in Power and Backroom Politics: Washington State Senate Bills 5609 and 5898," Presented at the Western Social Science Association 38th Annual Conference, April 17-20.

8. Peters, A., D.W. Dockery, J. Heinrich, and H.E. Wichmann (1997), "Short-Term Effects of Particulate Air Pollution on Respiratory Morbidity in Asthmatic Children," European Respiratory Journal, 10(4):872-879.

9. Choudhory, A.H., M.E. Gordian, and S.S. Morris (1997), "Associations Between Respiratory Illness and PM 10 Air Pollution," Archives of Environmental Health, 52(2):113-117.

10. Kim, C.S., and T.C. Kang (1997), "Comparative Measurement of Lung Deposition of Inhaled Fine Particles in Normal Subjects and Patients with Obstructive Airway Disease," American Journal of Respiratory and Critical Care Medicine, 155 (3):899-905.

11. Lipsett, M., S. Hurley, and B. Ostro (1997), "Air Pollution and Emergency Room Visits for Asthma in Santa Clara County California," Environmental Health Perspectives, 105 (2):216-222.

12. Li, X.Y., P.S. Glimour, K. Donaldson, and W. MacNee (1996), "Free Radical Activity and Pro-Inflammatory Effects of Particulate Air Pollution (PM10) in Vivo and in Vitro," Thorax, 51(12):1216-1222.

13. Albright, J.F. and R.A. Goldstein (1996), "Airborne Pollutants and the Immune System," Otolaryngology and Head and Neck Surgery, 114(2):232-238.

14. Glovsky, M.M., G.R. Cass, and A.G. Miguel (1997), "Particulate Air Pollution: Possible Relevance in Asthma," Allergy and Asthma Proceedings, 18(3): 163-166.

15. Purdom, P.W. (1980), "Industrial Hygiene," Chapter 8 in Environmental Health, 2nd ed., New York: Academic Press, pp. 462-465.

16. Dockery, D.W., M.E. Fay, B.G. Ferris, A.C. Pope, F.E. Speizer, J.D. Spengler, J.H. Ware, and X. Xu (1993), "An Association Between Air Pollution and Mortality in Six U.S. Cities," New England Journal of Medicine, 329:1753-1759.

17. Ando, M., K. Tamura, and K. Katagiri (1991), "Study on Suspended Particulate Matter and Polycyclic Aromatic Hydrocarbons in Indoor and Outdoor Air," International Archives of Occupational and Environmental Health, 63(4):297-301.

18. Yang, W.M., J.J. Yang, and Y.M. Zhao (1994), "Deposition of Air Particles with Polycyclic Aromatic Hydrocarbons (PAH) in Human Respiratory Tract," Chung Hua Yu Fang I Hsueh Tsa Chih, 28(3): 151-153.

19. Jenkins, B.M., A.D. Jones, S.Q. Turn, and R.B. Williams (1996), "Emission Factors for Polycyclic Aromatic Hydrocarbons from Biomass Burning," Environment, Science, and Technology, 30(8):2462-2469.

20. McCarthy, M. (1996), "Health Impacts of Smoke," Symposium on Grass Seed Field Burning, Olympia, Wash.: Washington State Department of Ecology.

21. Wandschneider, P., and R.D. Scott (1996), Contingent Valuation Study for Reducing Smoke from the Burning of Bluegrass Fields: Results of a Phone Survey, Pullman Wash.: Department of Agricultural Economics, Washington State University.

22. Shprentz, D. (1996), Breath-Taking: Premature Mortality Due to Particulate Air Pollution in 239 American Cities, New York: Natural Resources Defense Council, Inc.

23. Hahn, D., and J.W. Beasely (1994), "Diagnosed and Possible Undiagnosed Asthma: A Wisconsin Research Network (WREN) Study, Wisconsin Research Network (WREN) Asthma Prevalence Study Group," Journal of Family Practice, 38(4):373-379.

24. Taggart, V.S., and R. Fulwood (1993), "Youth Health Report Card - Asthma," Preventive Medicine, 22:579-84.

25. Larson, T.V., and J.Q. Koenig (1993), A Summary of the Emissions Characterization and Noncancer Respiratory Effects of Wood Smoke, Research Triangle Park, N.C.: U.S. EPA Office of Air Quality Planning and Standards.

26. Audience comments at the Symposium on Grass Seed Field Burning (1996), Olympia, Wash.: Washington State Department of Ecology.

27. WAC 173.43.0 (1997), Agricultural Burning, Olympia Wash.: Department of Ecology.

28. Agricultural Burning Practices and Research Task Force (September 26, 1997), Meeting 31, Spokane Wash.: Department of Ecology.

29. Adams, P.F., and V. Benson (1993), "Current Estimates from the National Health Interview Survey, 1990-1992," In Vital and Health Statistics, Bethesda, Md.: National Center for Health Statistics.

30. "Current Estimates From the National Health Interview Survey, 1993" (1994), In Vital and Health Statistics, Bethesda, Md.: Centers for Disease Control and Prevention, National Center for Health Statistics.

31. Adams, P.F., and M.A. Marano (1995), "Current Estimates From the National Health Interview Survey, 1994," In Vital and Health Statistics, Bethesda, Md.: Centers for Disease Control and Prevention, National Center for Health Statistics.

32. Bousquet J., J. Knani, H. Dhivert, et al. (1994), "Quality-of-Life in Asthma: Internal Consistency and Validity of the SF-36 Questionnaire," American Journal of Respiratory and Critical Care Medicine, 149:371-375.

33. Smith, T., District Assessment Center Manager with Spokane Regional Health (June 1, 1996), Personal communication.

34. U.S. Bureau of the Census (March 1995), Current Population Survey.

35. Wilson, S., Systems Compliance Specialist of Membership Accounting at Group Health Northwest (December 8, 1997), Personal communication.

36. O'Connor, C., Office of Epidemiology, Washington State Department of Health (December 3, 1997), 1995 Comprehensive Hospital Abstract and Reporting System Asthma Hospitalizations: Spokane and Washington State, Personal communication.

37. Sly, R.M., and R. O'Donnell (1997), "Stabilization of Asthma Mortality," Annals of Allergy, Asthma, and Immunology, 78(4):347-54.

38. Hunt, L.W., E.J. O'Connel, W.M. O'Fallon, C.E. Reed, M.D. Silverstein, and J.W. Yunginger (1993), "Accuracy of Death Certificate in a Population-based Study of Asthmatic Patients," Journal of the American Medical Association, 269:1947-1952.

39. Aldrich-Sigma-Fluka Material Safety Data Sheets (April 1996), CD version.

40. Sax, N.I. (1984), Dangerous Properties of Industrial Materials, 6th ed., New York: Van Nostrand Reinhold Co., Inc.

Corresponding author: Dr. Roe A. Roberts, Assistant Professor, Eastern Washington University, Health Services Administration N., 668 Riverpoint Blvd., Room 327, Mailstop #3, Spokane, WA 99202.