Registration Dossier

Diss Factsheets

Administrative data

epidemiological data
Type of information:
experimental study
Adequacy of study:
supporting study
Study period:
Published 1988
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Non-GLP compliant, non-guideline experimental investigation. Study published in scientific, peer reviewed journal.

Data source

Reference Type:
A survey into the respiratory effects of prolonged exposure to pulverised fuel ash
Schilling C J, Tams I P, Schilling R S, et al.
Bibliographic source:
Br J Ind Med. 1988 Dec;45(12):810-7.

Materials and methods

Study type:
cohort study (retrospective)
Endpoint addressed:
not applicable
Principles of method if other than guideline:
Retrospective cross-sectional occupational cohort study
GLP compliance:

Test material

Constituent 1
Reference substance name:
Pulverized fuel ash
Pulverized fuel ash
Details on test material:
Pulverised ash of 6 coal fired power stations in south east England (UK)


Type of population:
Ethical approval:
not specified
Details on study design:
HYPOTHESIS TESTED (if cohort or case control study): Long term exposure to PFA affects the respiratory system

- Type: other: lung function tests, ventilation measurements, chest radiology



- Total population (Total no. of persons in cohort from which the subjects were drawn): 304
- Selection criteria: 1. Males; 2. Boiler cleaners, ash plant attendants, fitters (electrical and mechanical), unit adjusters, and turbine operators who had worked for 10 years or more in coal fired power stations in the south eastern region of the Central Electricity Generating Board (UK) and all retired employees who had worked for 10 years or more as boiler cleaners, ash plant attendants, and turbine operators.
- Total number of subjects participating in study: 243
- Sex/age/race: male/26-74 years/white
- Smoker/nonsmoker: 99/144 (37 nonsmokers + 107 ex-smokers)
- Total number of subjects at end of study: 288

- Type: The cohort was assigned into 4 categories based on exposure: low, medium, high <20 years, high >=20 years

- Diagnostic procedure: respiratory symptoms questionnaire, lung function tests, ventilation measurements, chest radiology, interview
- Other health effects:

Exposure assessment: The cohort was assigned into 4 exposure categories based on occupational history. Personal shift exposures were estimated based on air sampling.
Details on exposure:
TYPE OF EXPOSURE: Occupational


High <20 years
High >=20 years

EXPOSURE PERIOD: from 10 to >=20 years



Less than 10 years experience in the medium group. Ten years or more as turbine operators.
Some turbine operators described regular duties on jobs involving heavy exposure to PFA. Those with 10
or more years experience of these duties were ascribed to the medium group.

Electrical and mechanical fitters and unit adjusters with 10 or more years experience in this group.

High <20 years
Boiler cleaners and ash plant attendants, less than 20 but more than 10 years.
Those with less than 10 years exposure in this group but at least 10 years altogether in power stations were
assigned to the medium category.

High >=20 years
Boiler cleaners and ash plant attendants, 20 or more years.
Statistical methods:
Stepwise multiple linear regression analyses were carried out on each lung function variable with age, height, weight, smoking status, and exposure acting as explanatory variables. An age/smoking interaction term was also incorporated in the model.

To investigate the relation between the median x ray readings, age, and the indices of the exposure (exposure group and duration, previous exposure to dust in another job, and exposure to asbestos), cross tabulations were first made of the numbers in each possible subgroup together with the prevalences of the abnormality to be analysed. The prevalences were then related to age group, exposure group, smoking habit, previous exposures to dust, and exposure to asbestos using logistic regression. This technique is similar to multiple regression, except that the logit transformation is used to transform the prevalence rates as analysis of untransformed prevalences by regression techniques is not valid.

Results and discussion

- Number of measurements: 360
- Average concentrations: see below
- Other:
Table 2 (see attached publication) shows the results for total dust exposure for all groups by power station. Technical problems were encountered at one power station and the measurements were unsatisfactory; for this reason they are not shown on table 2 (see attached publication). Those for A and C were not representative since these stations were closing and the duties of the various categories of staff did not conform to practices obtaining during the major part of the lifetime of the station. The results for the more representative stations (B, E, and F) show a relation between the low exposure, medium exposure, and high exposure groups that approximates to a ratio of 1:2:3.


Table 3 (see attached publication) lists the responses to several sections of the questionnaire by smoking habit and exposure group. Dyspnoea was assessed using the grading system of Fletcher. Current smokers showed a higher prevalence than non-smokers and ex-smokers of chronic cough, chronic phlegm, regular wheeze, and work related wheeze. Those in the high exposure (greater than 20 years) group showed a higher prevalence of chronic cough, chronic phlegm, dyspnoea grade > 2, regular wheeze, dyspnoea and wheeze, and wheeze and chest tightness compared with all the other groups. The work related symptoms showed no definite trends with exposure. Table 4 shows that there were relatively few non-smokers and ex-smokers in the high exposure groups and this should be taken into account when interpreting these data.

To assess the effect of exposure on lung function, a multiple linear regression analysis using exposure as one of the predictor variables was applied to all the lung function data. The mean total dust concentration for the high exposure group is roughly three times that of the low exposure group; the corresponding increase for the medium exposure group is twofold (see table 2 (see attached publication)). The variables affected significantly were PVC, PF, VC, and DCO. The remaining four variables (FEV,, FEVI/FVC%, FEF2575%, MEF5O%,25%) had a negative coefficient, establishing that the exposure effect was in the same direction for all responses (table 5 (see attached publication)).

When exposure was treated as a qualitative variable the only significant regression coefficient was for DCO in the high exposure group (greater than 20 years). Smoking habit was shown to influence significantly all variables except FVC and VC. Ex-smokers were generally more adversely affected than smokers. For DCO only smokers showed a significant effect. Because An occupational history of exposure to asbestos was not associated with any significant effects on the variables of lung function. Table 6 shows the lung function results standardised for a man aged 50 and 175 cm tall. This highlights the significant effects of exposure upon FVC, VC, PF, and DCO shown in table 5 (see attached publication).

x Ray readings were obtained for 208 men. Table 7 (see attached publication) shows how these men are distributed with respect to the four exposure categories. A much higher proportion of the more heavily exposed workers did not have an x ray examination as most had retired and were unavailable.

Small opacities
Table 8 (see attached publication) gives the distribution of small opacities. The overall prevalence of any change is 15% with 9% of category 0/1 or more. Most of these opacities were of irregular shape and occurred most often among the smokers (table 9 (see attached publication))

The logistic analysis showed that the prevalence of small opacities of category 1/0 or more was associated but not significantly with exposure group (chi squared = 5-80 with 3 degrees of freedom p = 0.12) and with smoking (chi-squared = 3.11 with 2 degrees of freedom p = 0-21) and to a lesser degree with age group (chi-squared = 4-59 with 4 degrees of freedom p = 0-33). Based on the logistic regression coefficient for the prevalence of small opacities of category 1/0 the figure shows how the prevalence varied with exposure group and smoking for men aged 55-64 who represented 42% of those who underwent x ray examination (see table 10). Although the coefficients differed significantly from zero at the 5% level, this shows the relative importance of exposure and smoking on the prevalence of small opacities. The highest fitted rates occurred for those men with high exposure for less than 20 years; indeed seven of the 31 men (23%) in this group showed such an abnormality compared with 11% among the remainder.
Of these seven men, five were current smokers and two ex-smokers. One had a history of pulmonary tuberculosis and one of pulmonary embolism. Three men had occupational histories of exposure to dust before joining the CEGB-one as a coal face worker, one as a tiler, and one as a stoker in coal fired ships. Of the two men without medical and occupational histories that might have contributed to their x ray changes, one had worked for 17 years as an ash plant attendant smoking 10 cigarettes a day and the other 18 years as a boiler cleaner smoking two ounces of tobacco a week. Compared with category 1/0 or more, the prevalence of category 0/1 or more was not as strongly associated with exposure as with smoking, nor was there a pronounced association with age. This result is not unexpected since irregular small opacities of category 0/1 are particularly associated with smoking and the analysis confirms that the smokers have a higher prevalence of any small opacities than the ex-smokers or the non-smokers. A history of exposure to dusts before employment in the CEGB and to asbestos was not significantly related to the prevalence of small opacities.

Pleural thickening
The prevalence of pleural thickening was strongly related to age (chi-squared = 13-32 with 4 degrees of freedom; p < 0-01), as shown in table 9 (see attached publication), but to none of the indices of exposure, nor to smoking. Below age 45, only one man had pleural thickening compared with one in five of those above that age. Four of the 19 retired workers (21%) had pleural thickening.

Pleural calcification
Table 11 (see attached publication) shows the pronounced relation between pleural calcification and exposure to asbestos. Altogether only eight men had pleural calcification and five of them had also had exposure to asbestos. No other index of exposure was predictive of pleural calcification.

Applicant's summary and conclusion

Effects on lungs were largely similar to those induced by concentrated ambient particles, i.e. mild neutrophil / mononuclear cell inflammation, focal alveolar septal thickening, increased cellularity, and slight fibrosis. Lung function tests after >20 years of heavy occupational exposure to PFA showed modest effect on forced vital capacity, vital capacity, forced expiratory volume in one second, peak flow, and gas transfer.
Executive summary:

Previous studies of respiratory disorders in workers exposed to pulverised fuel ash (PFA) have been confined to radiological effects, that were found to be minimal. The present survey included 268 men (88% of the defined population) with a history of more than 10 years of exposure to PFA in six power stations in the south east of England. Respiratory questionnaires with full occupational histories were obtained from all of these subjects, of whom 207 were actively employed and 61 had retired; 243 had lung function tests and 208 had chest x ray examinations. The men were grouped, using their occupational histories, into high, medium, and low exposure categories. Dust concentrations were obtained by personal sampling on a representative sample of men from the three exposure categories. Lung function tests showed that a modest effect on forced vital capacity, vital capacity, forced expiratory volume in one second, peak flow, and gas transfer (DCO) was associated with prolonged heavy exposure to PFA. The men with prolonged heavy exposure also showed higher prevalences of respiratory symptoms. No definite relation between exposure and x ray changes was established. The results of this cross sectional survey indicate that exposures to PFA should not exceed the limits recommended by the Health and Safety Executive for low toxicity dusts [long-term exposure limit 8-h TWA for pulverised fuel ash: 10 mg/m3 (inhalable dust), 4 mg/m3 (respirable dust)].