Allergic diseases are the 6th leading cause of chronic disease in the United States., affecting an estimated 50 million individuals. Among these individuals approximately 35.9 million suffer from allergic rhinitis, commonly referred to as hay fever, and another ten million are afflicted with allergic asthma, including three million children. The National Institutes of Health estimated the total cost of allergic disease exceed $12 billion in the United States. In order to effectively manage this health issue quantification of the effects of allergens on allergic symptoms is necessary as well as accurate reporting of allergen levels. The goal of this project is to establish dose-response relationships between 17 different pollen and mold spore species and allergic symptoms and spatially describe aeroallergen concentrations in the northeast. Health data will come from two sources; (1) a cohort of asthmatic children in Connecticut and Southern Massachusetts documenting daily asthma exacerbations, and (2) hospital and emergency room admissions in New York state.
A detailed spatial and temporal description of aeroallergen concentrations is necessary to accurately assess the relationship between pollen and mold spore counts with allergic symptoms. Presently, the aeroallergen monitoring network is extremely sparse. In New England there are two certified monitoring stations; Waterbury, CT and Chelmsford/Salem, MA (the Chelmsford station has recently moved to Salem, MA). New York, bordering New England, has stations at Albany, Brooklyn, and Armonk. Newark, NJ has a monitoring station within 12 miles of the Brooklyn site. Therefore, pollen and mold counts that are reported for a particular area are actually being reported from limited locations that may be a great distance from an individual’s location. One of the objectives of this study is to estimate aeroallergen concentrations for the entire study area from data collected at 6 monitoring stations. This will be accomplished by using spatial statistics techniques, such as kriging, and considering phenology, weather conditions, topography, vegetative cover and land-use.
The association between this aeroallergen exposure data and allergic symptoms will be analyzed using a generalized estimating equation and Poisson regression for the cohort and hospital data, respectively. Symptoms will be examined either daily or weekly depending on how accurately aeroallergen concentrations can be determined. Criteria air pollutants, weather conditions and demographic information will be controlled for in the models.