G-Estimation Methods and Applications to Quantitative Exposure

Details

• Personal Author:
  Picciotto S
• Description:
  In occupational studies, workers more susceptible to the outcome are likely to leave work (or even experience the outcome) earlier, thereby accumulating less exposure-a phenomenon that gives rise to healthy worker survivor bias. This R03 aimed to examine the relationships of exposure to straight (oil-based) metalworking fluids with all-cause and cardiovascular mortality in a cohort of autoworkers, using statistical methods that account for this bias while also incorporating quantitative exposure data. Traditional Cox regression analysis failed to detect any relationship between straight fluids and cardiovascular mortality, while the exposure appeared slightly protective for all-cause mortality. This is because conventional regression analysis cannot adjust correctly for employment status, a time-varying confounder that affected by prior exposure-even though failure to adjust also leads to bias. In contrast to traditional regression analysis, g-estimation of a structural accelerated failure time model treats each year of exposure separately and adds together the estimated effects, rather than adding together the exposures over time and attempting to estimate the effect of the cumulative exposure. This allows it to disentangle exposure and confounders over time in a way that traditional regression cannot do. Previous work with g-estimation had suggested that longer durations of straight metalworking fluid exposure shortened the lives of those who died from ischemic heart disease. The present work considered the intensity of exposure, not just its duration, and considered a few related outcomes. One set of analyses focused on a public health framework by considering a series of hypothetical occupational exposure limits (OELs) and reporting the total number of years of life that would have been saved among those who died of the cause of interest if those limits had been enforced. We estimated that banning straight metalworking fluids could have saved 6047 years of life among those who died during the study period (1941-1994). When we considered more specific outcomes, where the etiologic relationship appeared to be stronger, the estimates were 4003 life-years for IHD, 2932 for AMI, and 916 for cerebrovascular disease, which was the rarest of the outcomes considered. Enforcing exposure limits higher than zero, even if very low, resulted in considerably lower impact (than a ban) on all outcomes except for cerebrovascular disease. A separate set of analyses made a different assumption about the shape of the exposure-response curve by entering exposure into the outcome model as either a categorical variable treated as continuous, or (when possible) as a continuous variable, rather than a binary variable corresponding to an OEL. These results were consistent with those found above, though some outcomes showed attenuation in the slopes due to the etiologic importance of even low levels of exposure. Cerebrovascular disease did not show this attenuation, which corresponds to the fact that exposure limits above zero still had some impact on this outcome. Overall, our results support lower occupational exposure limits for straight metalworking fluids. [Description provided by NIOSH]
• Subjects:
  Automobiles Cardiovascular Diseases Cardiovascular System Environmental Exposure Occupational Health Risk Assessment Risk Factors Safety
• Keywords:
  Automotive Industry Cardiovascular System Cardiovascular System Disease Cardiovascular System Disorders Exposure Levels Metalworking Fluids Models MWFs Oils Regression Analysis Risk Factors

  More +
• Series:
  Grant Final Reports
• Publisher:
  National Institute for Occupational Safety and Health
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Final Grant Report
• Place as Subject:
  California ; OSHA Region 9
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Division:
  OEP - Office of Extramural Programs
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Pages in Document:
  1-16
• NIOSHTIC Number:
  nn:20048386
• NTIS Accession Number:
  PB2016-104196
• Citation:
  Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, R03-OH-010202, 2015 Mar; :1-16
• Contact Point Address:
  Sally Picciotto, UC Berkeley School of Public Health, Environmental Health Sciences, 50 University Hall, Berkeley, CA 94720-7360
• Email:
  sallypicciotto@berkeley.edu
• Federal Fiscal Year:
  2015
• Performing Organization:
  University of California - Berkeley
• Peer Reviewed:
  False
• Start Date:
  20120701
• Source Full Name:
  National Institute for Occupational Safety and Health
• End Date:
  20141231
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:57edee8e601fda30b2b66dd1b87637cc6b452dcafc29defbde2a08b3c3283cfb733dbe4a4ca6541f914c55ed9657831f3c808e8154bfe1e4b457f0473d732bb3
• Download URL:
  https://stacks.cdc.gov/view/cdc/218645/cdc_218645_DS1.pdf
• File Type:
  [PDF - 628.25 KB ]