COAL MINERS EXPOSED TO DIESEL EXHAUST EMISSIONS

ABSTRACT

The possibility of adverse respiratory health effects amongst coal miners in and at diesel-use mines has been investigated. Coal miners working in and at non-diesel-use mines served as matched controls. Differences in symptom prevalence and pulmonary function performance are documented. Several health indices have been related to a measure of diesel exposure. A preliminary characterization of the underground environment is shown. Underground miners in diesel-use mines reported more symptoms of cough and phlegm and had generally lower pulmonary function performance than matched controls. Similar trends were noted for surface workers at diesel-use mines compared with matched controls. Although a pattern consistent with small airways disease is shown, factors other than diesel exposure may be responsible. Exposure time to diesel emissions for miners in this study is relatively short and measured concentrations are low. Based on present information, sufficient and consistent evidence does not exist allowing for the rejection of hypotheses of health equality between matched groups. A prudent public health stance dictates reservation of judgment pending prospective examinations and detailed environmental surveys.

METHODOLOGY

Six coal mines utilizing diesel units underground were included in this study. Three were located in Utah and one each in Wyoming, Colorado and Kentucky. Four mines used diesel-powered units for face haulage of coal in place of conventional electric shuttle cars. Diesel front end scoops (for mucking operations) were also present at the four operations. Another mine used diesel-powered dump trucks for haulage of coal between the face and the tipple. One mine routinely used diesel units for mantrips and supply haulage.

All miners at the six operations were invited for examination. Nearly 1000 were examined and the participation rate was 93%. Each examination included pos­terior-anterior and left lateral chest films, a minimum of five spirometric manoeuvres, the administration of a modified version of the MRC respiratory symptom questionnaire , an employment and smoking history, and the determination of selected demographic characteristics.

A dry rolling sealed spirometer was used for lung function testing. Flow volume curves of forced vital capacity manoeuvres were recorded on an FM analog tape and later processed on a digital computer. During testing, each curve was displayed on an oscilloscope for evaluation of reproducibility . Pulmonary function indices derived were the forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1) and flow rates at 5 % intervals of the FVC exhaled. The primary measurements chosen for analysis were the FVC, FEV1 and flow rate at 50% of the FVC (FEF50). Peak expiratory flow and flow rates at 25, 75, and 90% were also considered. In the absence of total lung capacity (TLC), flow rates were not directly corrected for lung volumes; this was partially achieved through case matching procedures. The curve with the largest FVC and a peak flow within 15% of the maximum observed peak flow was chosen as the best curve. Flow rates were obtained from this 'best curve' determination. Volumetric measurements, viz. FVC and FEV1 were taken as the maximum value, regardless of the curve on which they occurred.

Persistent cough was defined by a positive answer to: 'Do you usually cough first thing in the morning in the winter' or 'Do you usually cough during the day in the winter', plus a positive answer to: 'Do you cough like this on most days for as much as 3 months each year?' Persistent phlegm production was defined similarly. Moderate to severe shortness of breath was indicated by a positive answer to: 'Do you get short of breath walking with other people of your own age on level ground?'

The radiological classification of coalworkers' pneumoconiosis was based on the median of independent assessments from three 'B' readers using the 1971 ILO U/C standards .

Coal miners, at nearby mines, not exposed to diesel emissions were selected as the comparison group. The control subjects were examined in the National Study of Coalworkers' Pneumoconiosis, details of which are published elsewhere . Several criteria, on a man-by-man basis, were employed in matching control to diesel exposed miners. Geographic area, smoking status, and race were important considerations and matches achieved on these factors were undeviating. Three other variables were considered in the matching procedure: age, height and years spent in underground mining. Although exact matching was attempted on these factors, it could not be achieved and variance was allowed. Table 1 shows distributions relating to matching for age, height, and years spent in underground mining. Cases which could not be matched within reasonable limits were excluded. Women miners, office workers and others not specifically performing mining jobs were also excluded. Hence, data from 823 pairs of miners were available for analysis. Table 2 portrays the general precision achieved in the matching process. To a large degree, effects from extraneous variables which could bias the results were neutralized. Means and percentages for (matching) variables are roughly equivalent.

The analysis was dichotomized to insure a distinct separation between matched pairs of underground and surface workers. As such, two types of controls could be exercised. The first involved the matching of underground workers in such a manner that effects might be attributed to differences in exposure. The second and equally important control involves disparities seen between matched surface workers, where no effect due to diesel exposure is expected. Therefore, checks in effects from one matched group to another could be made to partially quantify non-sampling error and other non-occupational factors.
A sample of miners at the diesel-use mines was asked to wear a personal dust sampler. The unit was operated at a flow rate of 2 1. min-1, such that it would perform similarly to the human respiratory system with respect to deposition and retention of different sized particles. In addition, each subject was asked to wear tandem passive dosimeters to measure concentrations of nitrogen dioxide. Area environmental measurements were also obtained to ascertain levels of CO, C02, NOx, NO2, SO2, CH2O and respirable and total dust.
Statistical procedures utilized and miscellaneous notes are contained in the Appendix.

RESULTS

Symptoms

Significantly more underground miners exposed to diesel emissions reported persistent cough than was the case with their matched controls: 23.6 vs 16.5%. Likewise, surface miners working at diesel-use mines reported more persistent cough than their matched counterparts: 20.1 vs 17.6%. These data are shown in Fig. 1(a).
Figure 1(b) shows that more of the underground diesel-exposed miners reported persistent phlegm production than their matched controls: 26.5 vs 22.8%. Amongst surface workers at diesel-use mines and their matches, the direction of the disparity was the same: 23.6 vs 21.8%.

The prevalence of moderate and severe dyspnea for workers in and at diesel-use mines and their matches is shown in Fig. 1 (c). Directional differences are opposite those shown for cough and phlegm. This is especially so for underground matched pairs where a significantly higher proportion of the controls reported the symptom: 23.8 vs 9.3%. Amongst surface workers, equivalence in reported prevalence existed between matched pairs: 6.3 and 6.6%.

Figure 2(a) shows the reported prevalence of other selected symptoms for underground workers in diesel-use mines and their matched pairs. A significantly greater proportion of workers in diesel-use mines reported exacerbations of cough and phlegm than did their matched counterparts: 21.7 vs 16.2%. Between-group differences for wheezing, hemoptysis, nasal drainage and chest illnesses causing activity limitations were slight and non-significant.

Similar data for surface workers at diesel-use mines and their matches are shown in Fig. 2(b). No directional trends are evident and all between-group disparities are non-significant.

Pulmonary function

Figure 3(a) shows mean differences in several lung function parameters between underground workers exposed to diesel emissions and their matched controls. Significant decrements were evident for FVC, FEV1, FEF75 and FEF90, showing the diesel exposed individuals as having somewhat lower pulmonary performance than their matched controls. A marked reversal exists for peak flow, wherein the diesel-exposed workers showed a significant increase relative to their matched pairs.
The same directional phenomena exist for the surface workers at diesel-use mines and their matched pairs. These data are shown in Fig. 3(b). For these contrasts, significant decrements were evident for FVC, FEF75 and FEF90, showing the surface workers at diesel-use mines as having lower pulmonary function status than their matched counterparts. The surface workers at diesel-use mines also showed significantly elevated peak flows relative to controls.

The proportion of workers in or at diesel-use mines showing evidence of obstructive airways disease is equivalent to their matched controls. These data are shown in Table 3, ca. 12% of each group showing obstruction, the majority being of a mild form.

Although restrictive lung problems are not common amongst working miners (in the absence of PMF), some mild restrictive conditions were observed. These data are shown in Table 4. Roughly 4% of the underground workers in diesel-use mines showed mild restrictive lung problems, the corresponding prevalence for the matched controls being around 2%. An equivalent disparity exists between surface workers at diesel-use mines and their matched counterparts: 6 vs 3%.


Symptoms and pulmonary function related to diesel exposure index
Figure 4(a) shows differences in the prevalence of persistent cough by select subgroups of matched pairs. The pairs have been allocated to a particular exposure group depending on the tenure of time the diesel-exposed member spent underground. Surface workers and their matches are shown separately. A logistic fit to these data shows that an increase in the difference between the prevalence of persistent cough for subgroups does not coincide with an increase in the indirect exposure index. The relationship is nonsignificant and does not show an ever-widening departure from the zero line with increased exposure time.

Figure 4(b) shows similar data as it relates to persistent phlegm production. No consistent trend exists showing the difference in the prevalence of persistent phlegm for subgroups to be related to the indirect exposure index.
Differences in moderate to severe dyspnea by years of diesel exposure are shown in Fig. 4(c). A logistic fit to the data shows that the trend is not significantly different from the zero line.

Figure 5(a) shows mean difference in FEV1 for matched subgroups by years of diesel exposure. As before, the years of diesel exposure relates only to that member of the pair who worked in an underground diesel-use mine. Each subgroup follows the overall trend in that the diesel-exposed individuals had (on average) a lower FEV1 than their matched counterparts. However, a general linear model shows these data not departing significantly from zero.

Figure 5(b) shows the same type of information as it relates to the FVC by years of diesel exposure. The same pattern exists for FVC as with FEV1, but even more so. The trend is more obviously parallel with the zero line. As the exposure index increases, there is no noticeable nor apparent widening of the discrepancy between the matched pairs of subgroups.

Figure 5(c) shows the mean difference in FEF50 by years of diesel exposure. An obvious lack of fit exists with these data and there is no noticeable trend of increasing or decreasing disparities between subgroups as the exposure index increases. The general linear model showed the relationship to be non-significant.

An analysis relating to degree of coalworkers' pneumoconiosis was considered but found to be unwarranted owing to the small number of cases involved. Only 4 cases of simple coalworkers' pneumoconiosis were found in each of the matched groups. No cases of progressive massive fibrosis were found. These data are shown in Table 5 for information only.

Potential interactions

Although not shown, the data were also analysed by age group, smoking status and mine, to ascertain if selected subgroups were contributing disproportionately to the overall differences. A fairly thorough mixing occurred between subgroups for each major factor considered. In short, there was no common thread indicating consistent interacting effects by particular subgroups.

Environmental measurements

Short-term detector tube samples were collected in work areas of the six mines. These data are shown in Table 6. Carbon monoxide (CO) levels ranged from slightly over 23 ppm in mine No. 5 to around 3 ppm in mine No. 1. Carbon dioxide (C02) levels ranged from trace detection upwards to 0.09% in mine No. 6. Oxides of nitrogen (NOx) levels ranged from zero upward to 0.6 ppm in mine No. 1. Sulphur dioxide (So2) and formaldehyde (CH2O) were either undetectable or found only in trace amounts.

In addition to detector tube samples, full shift personal samples (NO2 and respirable dust) were collected from miners at each mine site. These data are shown in Table 7. The miners' personal exposure to NO2 was relatively uniform between mines compared with the variability observed in short-term area samples. Full shift personal samples for NO2 ranged from a high of 0.28 ppm in mine No. 2 to a low of 0.13 ppm in mine No. 6. Respirable dust levels as measured by the cyclone and filter method ranged from a high of 2.73 mg m-3 in mine No. 2 to a low of 0.93 mg m-3 in mine No. 1.

Passive dosimeters to measure NO2 and pumps to measure respirable and total dust were situated at various stations within the mines. The range of these full shift area samples over all mines are reported in Table 8. As can easily be seen, maximum concentrations of NO2 and respirable and total dust existed at the face and in the return airway.
