A Statistical Assessment of the Scientific Evidence Relating Cotton Dust Exposure to Chronic Lung Disease

Abstract

A well-recognized adverse respiratory effect of occupational exposure to cotton dust in the textile industry is termed byssinosis. This term as commonly used embraces a gradation of symptoms resulting from cotton dust exposure that ranges from occasional chest tightness on the first day of the working week (after 36-48 hours off work) through chest tightness and/or difficulty in breathing on one of several days in the working week, to chest tightness and evidence of permanent respiratory disability and reduced ventilatory capacity. Table 1 presents the clinical classification of byssinosis developed by Schilling et al. (1963). Implicit in the Schilling classification is the distinction between the acute symptom of chest tightness the first day back at work and evidence of permanent loss of lung function (or chronic lung disease). This article discusses the epidemiologic evidence of the relationship between cotton dust exposure and chronic lung disease. However, before discussing this evidence, a summary of what is known about the acute response is presented as background. The acute response is manifested by work-week periodicity of chest tightness as determined by the worker's response to a set of standardized questions regarding the presence of work-related chest tightness on Mondays following a weekend off from work. As such, it is a subjective measure of acute bronchoconstriction. Bronchoconstriction can also be measured objectively by the difference in one-second, forced expiratory volume (FEVy) before and after the work shift. FEVI is the volume of air expelled from the lungs during the first second of a maximal forced expiration following a maximal inspiration. A non-smoking white male who is 40 years old and 5 feet 8 inches tall has a predicted value of FEV1 equal to 3.7 + .5 liters. FEVy is the most reliable index of clinically significant airways obstruction. A decline in FEV1 over the working shift beyond measurement error represents acute bronchoconstriction. The correlation between across shift decline on the first day back at work (a Monday fall) and byssinosis is mixed. For example, Imbus and Suh (1973) found an association, but Berry et al. (1973) found a relationship in those processing coarse cotton but not in those processing medium grade cotton. It is generally accepted that the prevalence of the acute response to cotton dust as determined by a questionnaire is influenced by level and length of exposure and depends on the job within the textile manufacturing process. Smoking also influences the prevalence of the acute response. The results of epidemiologic studies regarding the determinants of the Monday fall in FEVy are not as clear. For example, Merchant et al. (1973) found that the level of cotton dust exposure influenced Monday fall only in smokers; however, smoking itself did not affect Monday fall. Berry et al. (1973) and Jones et al. (1979) found a relationship with level of exposure but not with smoking or length of exposure. A possible reason that post-shift decline in FEV, is not strongly related to either chest tightness or other influencing variables is the measurement error variability in FEV1. A true drop across the work shift of 150 ml in FEVy would be considered evidence of acute bronchoconstriction. However, since the measurement error standard deviation in FEVy is approximately 125 ml, a true drop of 0 ml results in an observed drop in excess of 150 ml about 20% of the time. Conversely, a true drop of 200 ml would result in an observed drop less than 150 ml about 40% of the time. Thus, misclassification may be substantial. Turning to the possibility of a chronic response to cotton dust exposure, note that the Schilling byssinosis grading scheme implies a progression from periodic chest tightness on Mondays to permanent lung-function loss. However, there is no evidence for this from prospective studies. Parkes (1982) makes the following observations: