During the first four years of our Core's existence, 80 papers have been published by Core members (or are in press). Several recent Center publications indicate progress toward coherence of epidemiologic evidence on the question of which pollutants are the most important risk factors for human respiratory disease. Additional publications indicate progress in laboratory investigations to corroborate the epidemiologic evidence of specific pollutant-health relationships, and to understand their biological mechanisms. From the Children's Health Study, the two-part papers by Peters et al. (1999), reporting initial cross-sectional analyses, indicated that both symptoms and low lung function were associated more consistently with NO2, fine particulate matter, or airborne strong acid (all highly correlated with one another) than with O 3. Further cross-sectional analysis in the paper by McConnell et al. (1999) indicated that chronic phlegm and bronchitis were more prevalent in asthmatics from areas with high PM or NO2 , although asthma prevalence was not associated with any measured pollutant. Subsequent longitudinal analyses, reported in the paper by Gauderman et al. (in press), associated PM/NO2/acid, but not O3, with reduced lung function growth rates.

A separate daily time-series analysis of hospital admissions and air pollution in metropolitan Los Angeles, reported in the paper by Linn et al. (in press), associated asthma admissions (and also cardiovascular disease admissions) with PM, NO2 , or CO, but not with O3. Thus, a broad range of new epidemiologic evidence from the Center points toward "primary" pollutants, rather than secondary photochemical oxidants, as the principal source of health risk in Southern California air.

On the laboratory side, the series of papers by Diaz-Sanchez and coworkers points to one possible explanation, in showing that upper-airway challenge with a realistic dose of fine particles from diesel exhaust can markedly exacerbate the inflammatory response to an aeroallergen inhaled concurrently. This has led to new ideas about how to improve individual exposure assessment such as looking at proximity to roadways where exposure to primary pollutants occurs. Traffic density data plus GIS mapping is a high-priority initiative being pursued as this application is being submitted. The paper from Dr. Glen Cass' group (Miguel, et al. 1999) is relevant to this issue in that the research shows that allergenic material in road dust becomes airborne from vehicular traffic. This entire issue brings together elements from throughout our Center. The Exposure Assessment Research Core with its interest in particle characterization and quantification, the Study Design and Statistical Methodology Research Core with multi-stage models and measurement error expertise, and the Analytical Facility Core all need to work with this Core to understand disease relationships. This line of inquiry also has relevance to both the Adult and Children's Cancer Research Cores as ambient and in-house exposures resulting from mobile sources are likely to play some role in cancer etiology.

The mechanistic models described by Dr. Gilliland in section 4.1.4 and his publication (Gilliland et al, 1999) are very likely to reveal genetic polymorphisms associated with susceptibility to these pollutant effects. His preliminary data are already showing asthma risks associated with certain polymorphisms. The paper by Gong et al. (in press) demonstrates that it is feasible to expose volunteers to concentrated Southern California ambient PM and test a wide range of health responses; this approach will allow a range of realistic yet controlled exposure-effect studies which potentially can link the broader epidemiologic findings with the mechanism-oriented laboratory research. These examples demonstrate the cutting-edge quality of this Core's research and also shows how the Center structure and mechanisms for collaboration facilitate complex, interdisciplinary research.