Spirometry is often performed annually to monitor worker lung health. Excessive FEV1 loss in an individual employee can be a useful indicator of the development of lung disease, while differences in FEV1 declines between exposure groups may suggest continuing workplace hazards. However, the operating characteristics of the test are highly dependent on test performance and quality standards. We evaluated factors affecting FEV1 using 21821 test results from 1884 workers who participated in the annual spirometry screening program at a single chemical plant between 1973 and 2003. Testing procedures and each of the four models of volume spirometers used by the program met ATS standards. The influence of multiple factors on repeated measurements of FEV1 was examined using a mixed model. The FEV1 level was significantly associated with birth cohort and spirometer type, as well as age, gender, race, body weight, and tobacco smoking. FEV1 loss averaged 24.8, 17.9, 16.7, and 9.8 ml/yr for white and black males and females, respectively, after controlling for birht year, age at the first test, weight, pack-years of smoking, and spirometer type. Testing using different spirometer types gave FEV1 results differing by as much as 90 ml, over 3 to 9 times the observed annual declines. The accurate classification of individuals or worker groups as demonstrating either excessive or normal declines requires attention to multiple factors which have been recognized to affect lung function. Based upon these results, future analyses of job and exposure information should account for these factors, in order to facilitate recognition and prioritization of potential workplace hazards.
Proceedings of the American Thoracic Society. 2005 ATS International Conference, May 20-25, 2005, San Diego, California