Significance of the inability to reproduce pulmonary function test results.

In 1985, 864 patternmakers participated in a voluntary union-sponsored health screening program that included an evaluation of respiratory symtomatology and dysfunction. Pulmonary function test (PFT) measurements included a minimum of three readings of forced expiratory volume at 1 second (FEV1). A "reliable" test was one where the two best volumes were within 5%. Medical history and respiratory symptoms were assessed on a standardized questionnaire. Fifty-nine of the 864 tested were unable to reproduce their best FEV1 result. Although these 59 case subjects had significantly lower PFT results than the other 805 tested (P < .01), the mean values for FEV1 and forced vital capacity for the case subjects were greater than 90% of predicted values. The case subjects were more likely to experience wheezing and dyspnea and have a history of emphysema than the rest of the group screened (n = 805). They also had a higher mean age and more years in the trade. Twenty-one of the 59 case subjects were among the 602 who participated in a similar health screening program offered 3 years later. To minimize the effects of age and smoking status on PFT performance, these 21 case subjects were each matched on age and smoking with two comparison subjects who had reliable tests. At follow-up, the 21 case subjects and 41 comparison subjects both had a decline in ventilatory capacity that was significantly greater than would be expected by advancing age alone. A number of methodological issues that impact the interpretation of these data are discussed.