Airflow limitation among workers in a labour-intensive coal mine in Tanzania

To describe the relationship between cumulative respirable dust and quartz exposure and lung functioning among workers in a labour-intensive coal mine.

The study population comprised 299 men working at a coal mine in Tanzania. Lung function was assessed using a Vitalograph alpha III spirometer in accordance with American Thoracic Society recommendations. Multiple linear regression models were developed to study the relationship between forced expiratory volume in 1 s (FEV(1)), forced vital capacity (FVC) and FEV(1)/FVC and the cumulative dust or quartz exposure while adjusting for age, height and ever smoking. To evaluate trends for dose response, cumulative exposure concentrations for respirable dust and quartz were ranked and categorized in quartiles and the highest decile, with the first quartile as the reference group. Logistic regression models were used to determine odds ratios for FEV(1)/FVC < 0.7 and FEV(1)% < 80 for categories of cumulative dust or quartz exposure.

The prevalence of FEV(1)/FVC < 0.7 among the workers was 17.3%. Workers in the development team (20.5%) had the highest prevalence of FEV(1)% < 80%. The estimates of the effects of cumulative exposure on FEV(1)/FVC were -0.015% per mg years m(-3) for respirable dust and -0.3% per mg years m(-3) for respirable quartz. In logistic regression models, the odds ratios for airway limitation (FEV(1)/FVC < 0.7) for the workers in the highest decile of cumulative dust and quartz exposure versus the referents were 4.36 (95% confidence interval: 1.06, 17.96) and 3.49 (0.92, 13.21), respectively. The upper 10% of workers by cumulative dust and quartz exposure also had higher odds ratios for predicted FEV(1)% < 80% than the reference group odds ratio: 10.38 (1.38, 78.13) and 14.18 (1.72, 116.59), respectively. The results must be interpreted with caution due to a possible healthy worker effect and selection bias.

Exposure to respirable coal mine dust was associated with airway limitation as measured by FEV(1)/FVC and predicted FEV(1)%.