Article Text

Download PDFPDF
Original research
Estimating impacts of reducing acrylonitrile exposure on lung cancer mortality in an occupational cohort with the parametric g-formula


Objectives To inform the potential human carcinogenicity of acrylonitrile, we estimate associations between acrylonitrile exposures and lung cancer mortality in US workers with the objectives of (1) assessing potential for healthy worker survivor bias and (2) adjusting for this bias while assessing the expected lung cancer mortality under different hypothetical occupational exposure limits on acrylonitrile exposure using the parametric g-formula.

Methods We used data from a cohort of 25 460 workers at facilities making or using acrylonitrile in the USA. We estimated HRs to quantify associations between employment and lung cancer mortality, and exposure and leaving employment. Using the parametric g-formula, we estimated cumulative lung cancer mortality at hypothetical limits on acrylonitrile exposure.

Results Recent and current employment was associated with lung cancer, and exposure was associated with leaving employment, indicating potential for healthy worker survivor bias. Relative to no intervention, reducing the historical exposure under limits of 2.0, 1.0 and 0.45 parts per million would have been expected to reduce lung cancer mortality by age 90 by 4.46 (95% CI 0.78 to 8.15), 5.03 (95% CI 0.96 to 9.11) and 6.45 (95% CI 2.35 to 10.58) deaths per 1000 workers, respectively. A larger lung cancer mortality reduction would be expected under elimination of exposure: 7.21 (95% CI 2.72 to 11.70) deaths per 1000 workers.

Conclusions Healthy worker survivor bias likely led to underestimation of excess risk. Our results corroborate previous study findings of an excess hazard of lung cancer among the highest exposed workers.

  • Epidemiology
  • Occupational Health
  • Statistics
  • Volatile Organic Compounds
  • Longitudinal studies

Data availability statement

Data are available on reasonable request. Data can be made available after approval of NCI data transfer agreement. Mortality data cannot be transferred and can only be requested directly from NCHS via the National Death Index approval process. After the time of publication, data can be made available subject to these approvals. Contact SK, for more information.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.