Article Text
Statistics from Altmetric.com
151 WOOD DUST AND COPD: IS THE TLV PROTECTIVE?
Objectives:
Quantitative exposure estimates are needed to determine exposure–response relationships, which can then inform occupational exposure limits. We evaluated the exposure–response relationship between quantitative wood dust levels and hospitalisation for chronic obstructive pulmonary disease (COPD) in a cohort of Canadian sawmill workers.
Methods:
The study population consisted of 11 273 male sawmill workers that met the following criteria: (1) employed ⩾1 year between 1950 and 1995, and (2) employed >1 day between 1985 and 1995. This cohort was probabilistically linked to hospital discharge records (1985–1998) and 132 COPD cases were identified. Quantitative wood dust estimates based on statistical models were available for all workers. We evaluated the shape (log-linear vs log-log models), goodness of fit, and precision of the exposure–response relationship using Poisson regression.
Results:
The median, 80th percentile, and maximum cumulative wood dust exposure levels were 6.8, 27 and 89 mg/m3-year, respectively. The reference group were sawmill workers exposed to less than 5 mg/m3-year. Wood dust was strongly associated with COPD hospitalisations using a log-log model and only weakly associated with a log-linear model. The log-log model was ln(RR) = 0.203*ln(CE−Xref+1), where RR is the relative risk, CE is the cumulative exposure, and Xref is reference group’s mean exposure (2.3 mg/m3-year). Thus, there would be an 80–110% increase in COPD hospitalisations for workers exposed from 20 to 40 years at the current ACGIH TLV of 1 mg/m3. The cumulative exposure associated with a RR of 2 is 31 mg/m3-year. COPD was not associated with cumulative exposure to non-specific particulate.
Conclusion:
This study suggests that a lifetime of exposure to wood dust at the current TLV is associated with a doubling in the risk of hospitalisations for COPD, and thus the TLV may not be sufficiently protective of respiratory health. The actual health burden may be greater …