Factors influencing the occurrence of progressive massive fibrosis in British coalminers. Final report on CEC Contract 7256-34/016/08

This study had two main objectives. The first was to clarify the relationship between exposure to respirable mixed coalmine dust and the incidence of Progressive Massive Fibrosis (PKF) in working miners. The influences of age, colliery of employment, and simple pneumoconiosis (CWP) category were taken into consideration. Subsidiary descriptions of the prevalence of PMF and its progression were also carried out. The second main objective was to identify and describe the relationship to PMF incidence of a wide range of other characteristics, including detailed measures of the coalmining environment, anthropometric data, smoking habits, respiratory symptoms and lung function measurements. A limited evaluation was also carried out of our ability to identify the men most likely to experience an attack of PMF, using a few relevant characteristics which are easy to measure in coalminers generally.Prevalence of PMF: Results in the first part of the study were based on all miners who had ever participated in the Pneumoconiosis Field Research (PFR) of the National Coal Board. In all, 53 382 men were examined during at least one of the approximately five-yearly rounds of medical surveys of working miners at selected research collieries throughout Britain between 1953 and 1977 inclusive. The prevalence of PMF was assessed using clinical readings of chest radiographs by one or more doctors. In general the prevalence rate declined in successive surveys, and was higher in older men. There were marked differences between collieries, related to the geographical location of the pits and to the carbon content of the coal.Incidence of PMF over five-year periods: Studies of the incidence and progression of PMF were necessarily restricted to the 30 534 coalminers (57.2% who attended medical surveys on at least two occasions. Each man’s time under observation was divided into non-overlapping periods called man-intervals, characterised by his attendance at medical survey at the start and end-points of the interval, but not at any time in between. These intervals were the basic units used in the analysis.Results on incidence were based on study intervals where the working miner concerned did not show PKF at the start of the period, according to the clinical classifications of his radiograph. There were 53 054 such intervals of approximately five years’ duration, and in 498 (0.9% of these the men concerned were judged to have experienced an attack of PKF. The attack rate over five-year periods was related to age, CWP category, colliery of employment and the cumulative exposure to respirable dust experienced prior to the observation period (“”previous dust exposure””). Clear relationships were not found between attack rate and exposure experienced during the study interval itself.Dust exposure and incidence of PMF The effect of previous dust exposure was unambiguous. For example, the study intervals were divided into 96 groups by classifying the men concerned into four age-groups at each of the 24 collieries. There were no new occurrences of PKF in 28 of these groups, characterised primarily by younger men at collieries where the carbon content of the coal was lowest. In 65 of the remaining 68 groups, men who were classified as having PKF at the end of the five-year period had, on average, higher dust exposures at the start of the intervals than those who remained free of the disease. Logistic analyses indicated that, in gross terms, a man with twice the previous exposure of his colleague experienced about three times the risk of incidence. When variations in age and colliery were taken into account the relative risk for double the previous exposure fell to 2.5, while in addition each extra 10 years of age was associated approximately with a doubling of the probability of incidence. Differences in risk between collieries varied from lowest to highest by a factor of 20 or more, and were explicable in part by higher risks at collieries with high percent carbon in the coal.CVP category, mixed dust exposure and incidence of PKF: Important differences were found when CWP category at the start of the observation intervals was taken into account. The incidence rate was lowest, at two attacks per 1 000 men at risk over five-year periods, among men without simple pneumoconiosis. There was clear evidence that among these men the risk increased with higher previous exposures by factors comparable to those described above. Curiously, however, no new occurrences were found among men with Category 0 simple pneumoconiosis during the later years of the research.Higher attack rates were found among men with Category 1 or Category 2 CWP, but subsequent incidence of PKF was in general unrelated to the magnitude of the previous exposure which had influenced the occurrence of CWP initially. Finally, among miners with Category 3 CVP the crude incidence rate was one in six, while those who attained Category 3 CWP with relatively low exposures were at greatest risk.Progression of PMF: Results on the progression of PKF were based on 737 study intervals, of which 724 (98.2%) were of approximately five years’ duration. Among the 486 men with Category A PMF initially, 40% progressed, mostly to Category B, and 16% apparently regressed to “”no PMF””; while 17% of the 238 men with Category B initially, progressed to Category C. There were no obvious regional differences in progression rates. Three young men in South Wales apparently progressed from Category A to Category C PKF over five-year periods despite lew exposures relative to other Category A miners, but in general, clearcut associations of PKF progression with age or exposure were not found.The Follow-up study subset: The remainder of the project was based on data from men who were examined both at first and third rounds of medical surveys while working at one of the 24 research collieries, and who again participated in a series of follow-up surveys about 11 years later. Clinical readings of the first and third survey radiographs were used to exclude men with PKF on either occasion. Further exclusions during data verification led to a group of 4 749 men, more than half of whom were ex-miners at the time of the follow-up surveys, and in whom incidence of PKF between third and follow-up surveys was investigated.The follow-up survey radiographs had been classified by a panel of five readers, self-trained in the use of the ILO Classification, and working independently of one another. An attack of PKF was identified if, and only if, a majority of the readers recorded presence of PKF at follow-up survey. On this basis 257 cases of PKF were found.Detailed analyses of the causes of PKF: The 4 749 men were classified into 1 392 sub-groups generated by all possible combinations of colliery, cumulative dust exposure at third survey (in ranges of 25 gh/m3), and according to whether or not the man’s entire coalmining experience was at the research colliery where he was surveyed. Twelve cases were in groups that included only men who had experienced an attack, and so were excluded from the analysis of causes. Each of the remaining 245 cases was matched with up to four controls selected at random from the men without PMF in the same sub-group.Age and CVF category at third survey were not included as matching factors. As anticipated, therefore, the cases were on average older, with more severe simple pneumoconiosis at the start of the observation periods, than their matched controls. The estimated effect of age was similar to that described previously, while on average the incidence risks for a man with Category 3CWP were estimated as more than 12 times that of a man with Category 0 of similar age and previous exposure, and work inF at the same colliery.The anthropometric data revealed clear and systematic differences between cases and matched controls, with tall, light men more likely to experience an attack than were short, heavy ones. This result was found among men throughout the various coalfields in Britain, and was effectively independent of differences in age and in CWP category.Intensive analyses of the case-control data, overall and in many sub-groups, revealed no clearcut associations between the composition of the dust to which men were exposed before third survey, and subsequent incidence of PKF. There was sone evidence to suggest that particular combination’s of quartz and kaolin plus mica exposures may be related to subsequent attacks of PMF, but this tendency was no longer evident when age and CWP category were taken into account. A mild association was found, most clearly in the South Wales collieries, between incidence of PMF and estimates of the projected diameter of the dust to which men had been exposed. (No information was available on other relevant particle size parameters including aerodynamic diameter and particle shape.) Finally, an index of the residence time of dust in the lung was defined somewhat pragmatically. Higher values of the index were associated with higher risks of incidence. Indeed, in the South Wales collieries, the age-related variations were entirely explicable in terms of the relationship of incidence with residence time. Overall, it appeared that the combination of longer residence time together with shorter time exposed to the dust (or, equivalently, exposure to higher concentrations on average) led to increased risks of attack.Ken who at third medical survey reported respiratory symptoms, especially breathlessness or recent chest illness, were also at higher risk of subsequently suffering an attack of PKF, but not so in South Wales. Elsewhere, our estimates indicated that the probabilities of incidence were increased by a factor of about 2.5 among men who had reported both symptoms, relative to those with neither. Case-control differences in lung function and smoking habits were generally negligible, however. The only suggestive finding was that cases in South Wales contained relatively fewer current smokers than their matched controls ar.d had higher values of the ratio (FEV1/FVC), evaluated after adjustment for age, height and weight.Identification of men at special risk: The extent to which men at special risk of developing PKF could be identified early using certain routinely available characteristics was also examined in the k 7^9 men by constructing an allocation rule and estimating the proportions of PMF attacks and non-attacks correctly classified. The characteristics of the men used to construct the rule were age, height and weight combined as ^uetelet’s index, and profusion of small rounded opacities (CWP), all obtained at third survey.A range of related rules was examined and estimates of the associated classification errors obtained. Assuming that it was important to identify a substantial majority of those most likely to experience an attack of PKF, attention was directed to a procedure which successfully identified 835c of the attacks, and 61# of the non-attacks. This allocation rule required that practically all men of Category 1 CWP or more be regarded as being at special risk of developing PKF. Within Category 0 CWP, some useful discrimination was possible based on age and Quetelet’s index.Practical conclusions: The results are interpreted as reinforcing in two ways the existing policy of dust control underground to- prevent PKF. Firstly, our findings confirm results from earlier studies which showed that risks of developing PKF increase sharply with increasing category of simple pneumoconiosis. Thus dust control measures that reduce the incidence of simple pneumoconiosis will reduce the incidence of PMF ae well. In addition, our finding that the incidence of PMF in men with no simple pneumoconiosis (Category 0) is related directly to their previous dust exposures underlines the importance of dust suppression for all men at risk. We found no direct evidence of a need for special measures to control the quartz content of the dust, though results from earlier studies linking quartz levels with incidence of simple pneumoconiosis in some men remain relevant to approaches for preventing PKF.The second main preventive strategy is to identify, and to take special measures for the protection of, the men most likely to experience an attack of PKF. Our direct investigation of this issue was limited, but results were sufficient to show that the identification of men at special risk is feasible on the basis of age, simple pneumoconiosis category and Quetelet’s index. More refined predictions can almost certainly be made by considering additional characteristics such as respiratory symptoms and geographical location. “”