In recent years, employees at the three headquarters buildings of the U.S. Environmental Protection Agency (EPA) in the Washington, D.C. area have expressed concerns about air quality and work environment discomforts. As part of a large-scale study of health and comfort concerns, environmental monitoring was carried out in March 1989 at approximately 100 sites (rooms) within these buildings. Employees in the vicinity of the monitors were administered a brief questionnaire to elicit information regarding their work environment, comfort levels, odors noticed, health symptoms, mood states, and perceptions of overall air quality. Statistical analyses were carried out for the 191 males and the 192 females for whom both questionnaire and monitoring data were available. The analyses entailed estimation of linear regression and logistic regression models aimed at testing for associations between the employees' responses and the environmental measurements, which included temperature, humidity, carbon dioxide, and particlate concentrations (100 sites), and various microbiologic and volatile organic compound concentrations (subset of 56 sites). Principal component analyses were used to develop some of the outcome and explanatory variables used in the models. In the paper, the authors describe the study design, the study limitations, the statistical models and methods, and the results and implications of the data analysis.

"For presentation at Healthy Buildings - IAQ 91, sponsored by ASHRAE and CIB, Washington, DC, September 4-8, 1991." "EPA/600/D-91/125." Microfiche.