Accelerated lung function decline in an aluminium manufacturing industry cohort exposed to PM2.5: an application of the parametric g-formula

Objective Occupational dust exposure has been associated with accelerated lung function decline, which in turn is associated with overall morbidity and mortality. In the current study, we assess potential benefits on lung function of hypothetical interventions that would reduce occupational exposure to fine particulate matter (PM 2.5 ) while adjusting for the healthy worker survivor effect. Methods Analyses were performed in a cohort of 6485 hourly male workers in an aluminium manufacturing company in the USA, followed between 1996 and 2013. We used the parametric g-formula to assess lung function decline over time under hypothetical interventions while also addressing time-varying confounding by underlying health status, using a composite risk score based on health insurance claims. Results A counterfactual scenario envisioning a limit on exposure equivalent to the 10th percentile of the observed exposure distribution of 0.05 mg/m 3 was associated with an improvement in forced expiratory volume in one second (FEV 1 ) equivalent to 37.6 mL (95% CI 13.6 to 61.6) after 10 years of follow-up when compared with the observed. Assuming a linear decrease and (from NHANES reference values), a 20 mL decrease per year for a 1.8 m-tall man as they age, this 37.6 mL FEV 1 loss over 10 years associated with observed exposure would translate to approximately a 19% increase to the already expected loss per year from age alone. Conclusions Our results indicate that occupational PM 2.5 exposure in the aluminium industry accelerates lung function decline over age. Reduction in exposure may mitigate accelerated loss of lung function over time in the industry.