The workers in the development section of the mine were significantly more affected by the acute symptoms of breathlessness and blocked nose compared with the other production workers. The higher exposure to respirable dust and quartz compared with other workers might explain this . Our study also associated the presence of chronic respiratory symptoms and exposure to quartz and respirable coal mine dust. The fact that the specific group of workers from the development section has higher exposure and higher occurrence of symptoms has not been shown before. It might be that, awareness of high exposure is related to greater willingness to respond positively to questions about symptoms. However, in our study symptoms like runny nose and sneezing, not traditionally considered to be related to dust exposure were not different between the development, mine and the others and this strengthens our findings.
Our study showed a lower prevalence of chronic symptoms than previous studies from the United States, the United Kingdom and China. This might be explained by lower dust exposure levels in the present study. Workers in the mine team (coal face) had an average exposure of 0.66 mg/m3 [14, 19], which was lower than in previous studies in the United States (1.1 ± 0.5 mg/m3), Australia (1.51 ± 1.08 mg/m3) and South Africa (0.9 – 1.9 mg/m3).
As a reminder, the frequency of chronic symptoms in the current study were 25.3% for any cough, 5.6% for chronic cough, 13.3% for any cough with sputum, 3.2% for chronic cough with sputum, 34.5% for short of breath when hurrying on level ground, and 8.1% for wheeze. The National Study of Coal Workers' Pneumoconiosis in the United States showed that 35% of the workers employed in coal mines before 1970 had chronic bronchitis (chronic cough and phlegm), 43% had shortness of breath and 42% had wheezing . Seixas et al.  studied 1185 workers who started mining from 1970 and later; the prevalence of respiratory symptoms was lower, by reporting that 28% had cough, 32% phlegm, 21% chronic bronchitis, 22% breathlessness and 27% wheezing. Another study  among coal miners in the United States reported the prevalence of chronic bronchitis to be 33%, and studies of coal miners in the United Kingdom found that the prevalence of chronic bronchitis was 37%  and 39% . A study of coal mine workers in China  showed that 77% had breathlessness walking at a normal pace on level ground, 47% had chronic cough and 37% had chronic phlegm.
The studies in the United Kingdom and the United States showed that chronic respiratory symptoms were associated with both smoking and dust exposure levels [5, 11, 24]. When converting gm hr/m3 to mg- yrs/m3 by using a factor of 1.74, the cumulative dust exposurefor coal miners of 250 gm-hr/m3reported by Rae et al.  is equivalent to 144 mg-yrs/m3 which is close to 136.3 mg- yrs/m3 presently found for development workers. Further, the mean age of the development workers of 36 years falls half way between the age groups of 25–34 and 35–44 described by Rae et al. . For never-smokers in these two age groups, Rae et al.  reported an observed prevalence of cough with sputum for most days for 3 months of 20% and 22.2%. This is about 3 times greater than 7.7% reported by the development workers in the present study.
Kibelstis et al  showed that in each age group cigarette smoking coal face workers had significantly higher prevalence of respiratory symptoms than their non-smoking counterparts. In the study by Seixas and co-workers , never smokers had lower prevalence of respiratory symptoms than ex smokers and current smokers. However, the prevalence of respiratory symptoms in our study is lower than reported by Seixas et al  also among never-smokers.
The current prevalence of chronic cough of 5.6% is comparable to that reported by Naidoo et al. in South Africa (5.3%), who also reported relatively low prevalence of cough (9.0%), chronic phlegm (8.6%) and chronic bronchitis (7.5%) [23, 25].
The prevalence of acute respiratory symptoms has to be interpreted with caution, as they correlate significantly with chronic symptoms. This may imply either that people with chronic symptoms also experience more acute symptoms or that people with chronic symptoms report the problem as an acute symptom. The definition of acute symptoms might confuse workers with chronic symptoms, thus exaggerating the acute respiratory problems among the coal mine workers.
The strengths of the current study include the availability of quantitative exposure data and the large contrast in exposure between the groups. However, we could only investigate relative differences in symptom prevalence between the exposed groups since we did not include an external group not exposed to mixed coal dust. The results indicate an association between dust exposure and respiratory symptoms, since stratification by smoking habits did not alter the significant difference in the prevalence of cough as much as 4–6 times a day for 4 days or more in a week and shortness of breath walking with people of own age between the groups; but a cross-sectional study cannot confirm causal relationships.
Further, information bias might have affected the reporting of symptoms. Our study took place when Tanzania was implementing public sector reform: moving from public ownership of industry into private or mixed public-private ownership. The planning of this process had started in the present mine at the time of our study and some workers were presumably afraid of losing their jobs because they could not be absorbed into the private sector immediately. In this context, some workers in the mine might not have given correct information on respiratory symptoms by thinking that such information could be used as a screening criterion to prevent future employment. This might have contributed to the low symptom prevalence observed in this study, although all workers were assured confidentiality during participation.
The healthy worker effect might also be an issue since only the current workers in the mine were studied. Workers who had developed respiratory symptoms and airflow limitation might have left the mining industry, thus contributing to underestimating the effect of exposure.
The use of respirable coal mine dust samples might be misleading, since the development of some of the respiratory symptoms might be more closely related to larger dust particles. However, Seixas et al.  addressed this issue and concluded that a respirable dust concentration is a sensible proxy for measuring larger particles. The exclusive use of current exposure data in the construction of cumulative exposure is a limitation of the study. However, according to the management the coal production was fairly stable for the past two decades and no major changes in the production processes had been done, indicating that the current data is representative for also the past exposures. The exposure levels were also similar in the two periods of sampling in this study.
This study was conducted in a mine in Tanzania, and the results may be difficult to generalize to other countries, although the information might be valid for the mines elsewhere with similar characteristics. However, the information obtained will be useful in improving the working conditions in the mine.