Abstracts: ISEE 20th Annual Conference, Pasadena, California, October 12–16, 2008: Contributed Abstracts
The National Heart, Lung, Blood Institutes’ Cardiovascular Health Study (CHS), initiated in 1989, is a population-based longitudinal (prospective) study of coronary heart disease and stroke among elderly residents of four U.S. counties. In an CHS ancillary study to assess the impacts of air pollution and other environmental exposures on clinical measures of cardiovascular and respiratory health, an exposure assessment questionnaire (EAQ) was administered to participants in three Field Centers (FC).
To define between-community differences in potential exposure modifiers and confounders, and to enhance retrospective assessment of individual-level long-term (years) exposures to air pollution (particulate matter (PM10, PM2.5), ozone (O3), nitrogen dioxide (NO2), carbon monoxide (CO), and sulfur dioxide (SO2).
The main EAQ was developed for CHS participants to self-administer at home and complete within 30 minutes. Questions ascertained residence information, housing characteristics, and patterns of usual activities. In 2000, with the exception of those who were known by local CHS FC staff to be too ill, or who had declined participation in ancillary studies, all active participants in Sacramento County, California, Forsyth County (Winston-Salem area), North Carolina, and Allegheny County (Pittsburgh), Pennsylvania were invited to participate. The questionnaire, a cover letter from the CHS FC principal investigator, and a postage-paid return envelope was mailed to each current participant (n∼3000) in each of two seasonal periods (winter/spring, summer/fall). Non-respondents were contacted by telephone and at that time asked a subset of the questions contained in the main questionnaire.
Although not optimum, a reasonably good response rate was achieved with the winter administration: 54% (n= 574) in Forsyth, 52% (n=658) in Sacramento, and 47% (n=509) in Allegheny. An additional 10% from each County completed the telephone-administered questionnaire. Approximately 80% of winter respondents also completed the summer questionnaire. Statistically significant between-community differences in a number of exposure-modifying factors were observed. For example, distance of primary residence to a roadway differed (P < 0.01): In Pittsburgh, Forsyth and Sacramento Counties, 54%, 30%, and 22% of the respondents, respectively, indicated their residence was either in immediate proximity to a busy street or within a block of a busy street. Proximity to a busy street was not related to income bracket. The proportion of homes with gas appliances, which are an indoor source of both NO2 and PM, differed (P < 0.01) among the three communities: the proportion of participants with a gas stove or oven was greatest in Allegheny (46%), while Forsyth had the lowest (11%). The total amount of time a gas stove was used, and frequency of use, did not differ by gender within a given community. Relatively few respondents (<8%) reported that they, someone living with them, or someone who visited them regularly, smoked inside their residence. Activity patterns, including the type, time of day and location of activities also varied by community, by gender and income bracket.
Discussion & Conclusion:
The observed between-community and within-community differences in factors that can modify exposures to pollutants have important implications for both analytic strategies and interpretation of results from studies involving multiple communities.