March 9, 2018
Scientists from Exponent's Center for Health Sciences have published an article entitled "Reanalysis of Diesel Engine Exhaust and Lung Cancer Mortality In The Diesel Exhaust in Miners Study (DEMS) Cohort Using Alternative Exposure Estimates and Radon Adjustment," now available online from the American Journal of Epidemiology. In this article, Dr. Ellen Chang, Dr. Suresh Moolgavkar, and Mr. Edmund Lau and their co-authors present a statistical re-analysis of data from Diesel Exhaust in Miners Study (DEMS), a landmark research study conducted by the National Cancer Institute and the National Institute for Occupational Safety and Health. DEMS was designed to investigate the association between exposure to diesel engine exhaust and lung cancer mortality in workers at eight U.S. nonmetal mines.
In their new study, Chang et al. used well-justified alternative estimates of diesel engine exhaust exposure, generated by co-authors Dr. Kenny Crump and Ms. Cynthia Van Landingham, along with statistical control for exposure to radon, an established lung carcinogen, to investigate in depth the association of diesel engine exhaust with lung cancer mortality. The authors found that the association was generally weakened and statistically insignificant with use of the best available alternative exposure estimate, and that control for radon further weakened the association. No association between diesel engine exhaust exposure and lung cancer mortality was detected among miners who worked exclusively underground — that is, the group anticipated to have the highest levels of continuous exposure to diesel engine exhaust — nor was any association detected in seven out of the eight mines.
The findings of Chang et al. build on those of a previous Exponent reanalysis of the DEMS cohort, in which the authors found that temporal factors including duration of exposure and timing of exposure initiation and cessation, as well as effect modification by age, had an important influence on the association with lung cancer mortality. Taken together, these results identify key areas of uncertainty in the DEMS data and highlight the value of making scientific data available for independent reanalysis by multiple investigators.