Occupational Exposure to Vapor, Gas, Dusts, and Fumes Among Rural Residents

Details

• Personal Author:
  Houseman E ; Virji M
• Description:
  Objective: Direct-reading instruments are valuable tools for measuring exposure. They provide real-time data and valuable information on short-term exposure variability. However, statistical analysis is complicated by autocorrelation among successive measurements, nonstationary time-series, and presence of left-censoring due to limit-of-detection (LOD). A Bayesian framework is proposed for analyzing exposure timeseries that accounts for nonstationary autocorrelation and LOD issues. Methods: A spline based approach was used to model nonstationary autocorrelation with relatively few assumptions about autocorrelation structure. Left censoring was addressed by integrating over the left tail of the distribution. The model was fit using Markov-Chain Monte Carlo within a Bayesian paradigm. The method can flexibly account for hierarchical relationships, random effects and fixed effects of covariates. The method was implemented using the rjags package in R and is illustrated by applying it to real-time exposure data. Estimates for covariates from the Bayesian model were compared to those from the frequentist models including linear regression and mixed effects models with different autocorrelation structures. Simulations studies were conducted to evaluate method performance. Results: Simulation studies with LODs ranging from 0-50% showed lowest root mean squared errors for task means and the least biased standard deviations from the Bayesian model compared to the frequentist models across all levels of LOD. In the application, task means from the Bayesian model were similar to means from the frequentist models, while the standard deviations were different. Parameter estimates for covariates, e.g., source enclosure, were significant in some frequentist models, but in the Bayesian model their credible intervals contained zero; such discrepancies were observed in multiple datasets. Variance components from the Bayesian model reflected substantial autocorrelation, consistent with the frequentist models. Plots of means from the Bayesian model showed good fit to the observed data. Conclusions: The proposed Bayesian model out performs the frequentist models in estimating task means, standard deviations and parameter estimates for covariates, thus providing an approach for modeling nonstationary autocorrelation in a hierarchical modeling framework. [Description provided by NIOSH]
• Subjects:
  Occupational Health Safety
• Keywords:
  Direct-reading Methods Exposure Assessment Mathematical Models Simulation Methods Statistical Analysis
• Publisher:
  American Industrial Hygiene Association
• Document Type:
  Text
• Genre:
  Abstract or Summary
• Place as Subject:
  Maryland ; Oregon ; OSHA Region 10 ; OSHA Region 3 ; West Virginia
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Division:
  RHD - Respiratory Health Division
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Pages in Document:
  31-32
• NIOSHTIC Number:
  nn:20065989
• Citation:
  AIHce 2016: American Industrial Hygiene Conference and Exposition Pathways to Progress, May 21-26, 2016, Baltimore, Maryland. Falls Church, VA: American Industrial Hygiene Association, 2016 May; :31-32
• Contact Point Address:
  E. Houseman, Oregon State University, Corvallis, OR 97331
• Federal Fiscal Year:
  2016
• Peer Reviewed:
  False
• Source Full Name:
  AIHce 2016: American Industrial Hygiene Conference and Exposition Pathways to Progress, May 21-26, 2016, Baltimore, Maryland
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:18cb7d12090488a12f440edb077293bb742dd51b24fa17d3abccc38e460ae82670f155dfb65e9ea5c74bb09e907ca68cd2a854fb0b73865b09e3c9a74e55e439
• Download URL:
  https://stacks.cdc.gov/view/cdc/209081/cdc_209081_DS1.pdf
• File Type:
  [PDF - 1.41 MB ]