Prevalence of spirometric abnormalities in a representative sample of the population of Michigan

Abstract

A random sample of a large industrial state in the United States (Michigan) was studied via respiratory questionnaire, medical history, physical examination, and spirometry: 976 white adults performed a battery of spirometric tests (FVC, FEV1, FEV1/FVC and FEF25-75%). Using predictive equations derived from normal subjects in the survey and defining spirometric abnormality by the lower 95% confidence interval, prevalences of abnormality were greater in those who had a smoking history. When equations derived from normal non-smokers were used, as has been the general practice, 21.3% of men and 18.5% of women with a positive smoking history had an abnormal FEV1 and 26.7% and 17.8% an abnormal FEF25-75%, respectively. Although FVC is thought to be affected little by smoking, mean values were lower and prevalences of abnormal values higher in smokers. Prevalences are also shown using an adjustment for the effect of smoking and using conventional (percent of predicted value) definitions of spirometric abnormality. The prevalences of spirometric impairments in this cross section of a large state should be useful for comparison with other populations under study in North America.