During the past five years, researchers have started to use cross-sectional surveys of multiple office buildings to identify factors that are statistically associated with health symptoms. This new methodology is yielding valuable information on the causes of these symptoms. The California Healthy Building Study (CHBS) is one of these recent cross-sectional surveys. It is the first survey of this type performed in the U.S. and is a project of the Center's Indoor Environment Program. The researchers are myself, Al Hodgson, Joan Daisey, David Faulkner, and Matty Nematollahi - all with the Indoor Environment Program; Mark Mendell, National Institute for Occupational Safety and Health; and Janet Macher, California Department of Health Services. During the study's initial phases, the researchers gathered background data on health symptom prevalences and indoor air quality in typical ("non-sick") buildings and tested several hypotheses about the associations between the symptoms and features of the buildings, their indoor environments, and jobs performed. The study's long-term goal is to understand how to create "healthy" office buildings whose occupants have fewer work-related symptoms and higher productivity.

Twelve San Francisco-area office buildings were selected without regard for occupant complaints. To better correlate symptoms with method of ventilation, we divided the buildings into three groups: naturally ventilated, mechanically ventilated with operable windows and no air conditioning, and mechanically ventilated with air conditioning and sealed windows. Questionnaires were completed by 880 occupants, who reported their health symptoms and provided demographic and job data. Indoor and outdoor concentrations of CO2, CO, volatile organic compounds (VOCs), fungi, and bacteria were measured along with indoor temperatures and humidities.

Average symptom prevalences for the entire study population plus the minimum and maximum prevalences in individual buildings. For this figure, a work-related symptom is defined as a symptom that occurred often or always during the previous year and improved when the occupant was away from the building.

The next step was to look for correlations between symptom prevalences and the characteristics of the individual, job, workspace, building, and indoor environment. A few results are worth singling out. For example, buildings that used mechanical ventilation without air conditioning and those using mechanical ventilation with air conditioning had a higher prevalence of all symptoms except headaches compared to buildings with natural ventilation. The association between air conditioning and higher frequency of symptoms is consistent with the results of European surveys. The CHBS is the first study to include a group of buildings with mechanical supply and exhaust ventilation but operable windows and no air conditioning. Elevated levels of symptoms in these buildings are surprising since the building type is not commonly associated with health complaints. One possible explanation is that mechanical ventilation systems are themselves sources of pollutants such as bioaerosols, fibers, and VOCs.

Job-related or workspace factors also correlated with increased prevalences of one or more symptom groups. For example, our finding that the use of carbonless copy paper is associated with increased symptoms agrees with the findings of a Danish study. Organic chemicals in this type of paper may be the cause, and inhalation of vaporized compounds or physical contact with the paper may be the exposure route.

European surveys and the CHBS also agree that increased symptoms and carpets are associated. Carpets could be a source of increased symptoms because they release VOCs or fibers or because microbiological material such as fungi and dust mites find them perfect habitats. In the CHBS, release of VOCs from carpets was probably not the cause of symptoms because the carpets were generally old.

So far, no associations between symptoms and environmental parameters measured in the study have been identified. Most other surveys have also failed to verify a connection between symptoms and indoor air pollutants, but several indicate that the frequency of symptoms increases with temperature. Connections between pollutants and symptoms would not be identified if the study measured the wrong pollutants or if the measurements took place at the wrong times and locations to represent the occupants' exposures adequately. Follow-up studies are underway in the same buildings to investigate the environmental causes of the symptoms. Eventually, the study team will conduct experimental interventions such as increases in ventilation rates or improved office-cleaning practices to evaluate ways of improving the health of office workers.

--William Fisk

William Fisk
Indoor Environment Program
(510) 486-5910; (510) 486-4089 fax