A study of byssinosis and other respiratory symptoms in 2,528 flax workers aged 35 years and over in Northern Ireland is reported. This represented 82-5 % of the total available population.Only 3 % of workers were not seen because of absence or a refusal to co-operate. Workers were interviewed using a questionnaire based on the Questionnaire on Respiratory Symptoms (Medical Research Council, 1960a) with additional questions relating to respiratory symptoms at work.Byssinosis was found in workers in all stages of the industry, though its prevalence was highest in flax preparers; wet spinners and wet polishers did not appear to be at serious risk of developing the condition. When the effects of other relevant factors had been allowed for, e.g., age, duration of employment, and smoking habits, differences between the prevalence in the two sexes were found to be very small. The associations between byssinosis and the age of workers and their durations of employment in flax-preparing occupations were complex, and it was thought that a selective discharge of affected workers before the study might, in part at least, explain the absence of marked associations between these variables.Marked associations were found between both chronic bronchitis and exertional dyspnoea and the type of occupation in the mill. Workers in the early preparing occupations had a considerably higher prevalence of these conditions than expected on a null hypothesis. There were also marked associations between byssinosis and bronchitis, and between byssinosis and dyspnoea. The possible importance of these associations with regard to the aetiology of byssinosis is discussed, and it is suggested that byssinosis represents an acute, specific effect of certain textile dusts on the respiratory system, superimposed on a non-specific chronic bronchitic process.

No abstract

No abstract

Pulmonary pressures, total and timed vital capacity, functional residual capacity and carbon monoxide uptake have been correlated with temperature, humidity, suspended particulate matter and gaseous acid in a 3-month prospective survey of 10 cardiorespiratory cripples. Results have been analyzed by a sequential multiple regression technique, and significant relationships demonstrated between pulmonary function measurements and time, humidity and suspended particulate matter. The immediate response to suspended particulate matter was consistent with bronchoconstriction, although the most significant features were a transient depression of pulmonary pressures and carbon monoxide uptake. The prolonged depression of carbon monoxide uptake, starting 4 days after an intense episode of pollution, did not show itself as a significant long-lag coefficient for the survey as a whole; the existence of a threshold dose of suspended particulate to produce structural damage is postulated. Submitted on February 20, 1959

It has been suggested that the data already reported should have been analysed by a different method from that which we adopted. The suggested method, which is similar to a certain type of splitplot analysis (e.g. Cochran and Cox, 1957), differs from the method we used in that more than one source of variation is allowed for; e.g. variation between different subjects, variation in the change during the day for a given man. We agree that the split-plot analysis of variance might have been preferable because it provides a more valid comparison of:(a) The average FEV(1) or FVC between smokers and non-smokers, and of (b) The average decline during the day of FEV (1) or FVC between smokers and non-smokers. We considered this approach before publishing our results but decided against it because of heterogeneity of error between smokers and non-smokers. However, we now realize that no significant heterogeneity exists.We have re-analysed our data using the split-plot technique which shows that the comparisons between smokers and non-smokers could have been made for all days combined instead of for each day separately as in our Table IV. Thus our data may be summarized by the following Table: FEV ( The analysis shows that average FEV(1) and FVC were not significantly greater in non-smokers than in smokers as we originally concluded. The new analysis also shows that the declines in FEV(1) and FVC during the day were not significantly greater in smokers than in non-smokers.