Workers in poorly ventilated offices are twice as likely to report the symptoms of sick building syndrome (SBS) as are employees in a well-ventilated environment, a new Cornell study finds.

The researchers say they find no link, however, between SBS complaints and almost three dozen potential irritants studied, or between the syndrome and age, education, gender, general stress, positive or negative feelings or a variety of other psychological factors. They did find mild links to a variety of physical workplace problems, including sensitivity to odors, feelings of being overworked, migraines, allergies and, surprisingly, musculoskeletal problems, which indicates that ergonomic factors play a role in the syndrome.

"These results strongly suggest that symptom reports are not primarily psychological in origin, which some researchers have suggested," says ergonomist Alan Hedge, director of the Human Factors Laboratory in Cornell's College of Human Ecology and co-author of the report. "Both the workers with very few symptoms and those with more intense symptoms show a clear pattern of increased problems by the end of the day, suggesting that something is making the workers who are more sensitive feel sick," says Hedge. "They are not simply grumblers -- though nothing we've looked at so far seems to be the sole culprit."

The study is among the first to find that a relatively small buildup of carbon dioxide from human respiration -- an indicator of poor ventilation -- is related to SBS. It is also the first study to compare employees in similar work environments with no or few symptoms with those with many symptoms by asking them to keep a daily diary for one week.

Hedge and research associate William Erickson first tested four, multistory state office buildings in Trenton, N.J., hourly, for two to three consecutive days, measuring nearly 36 potential worker irritants, including light levels, temperature, carbon monoxide, carbon dioxide, relative humidity, dust mass, carpet dust, dust mite allergens, suspended particulate counts, nicotine and formaldehyde.

They then collected 1,508 questionnaires from workers in the buildings concerning their perceptions of ambient conditions, job stresses, work-related SBS symptoms as well as personal information.

No one irritant was linked to a particular symptom, even though workers in the study showed a clear pattern of feeling worse by the end of each day. But Hedge and Erickson found that the odds of workers reporting specific SBS symptoms were substantially higher when the carbon dioxide levels were above 650 parts per million.

"This suggests that SBS symptoms may be associated with building ventilation performance," Hedge says.

Hedge has been studying SBS for more than 10 years. In 1993 he reported that in a study of 1,324 workers from nine buildings, SBS symptoms were linked to the amount of man-made mineral fibers in settled office dust and not to tobacco smoke. In 1996 he reported that the brighter the office lights, the more often workers reported problems related to lethargy, tiredness and headaches.

In this latter study, SBS seemed to be linked to such nonenvironmental variables as heavy computer use, gender (women report more problems), job stress, lower job satisfaction and advancing age. Several of these findings, however, were not supported by the latest study.

The new study, Sick Building Syndrome and Office Ergonomics: A Targeted Work Environment Analysis, was funded by the Center for Indoor Air Research, Linthicum, Md.