Cumulative Retrospective Exposure Assessment (REA) as a predictor of amphibole asbestos lung burden: validation procedures and results for industrial hygiene and pathology estimates

Our plain-English summary of Cumulative Retrospective Exposure Assessment (REA) as a predictor of amphibole asbestos lung burden: validation procedures and results for industrial hygiene and pathology estimates, by J.O. Rasmuson, V.L. Roggli, F.W. Boelter, et al (Inhalation Toxicology, 2014).

The study

• Authors: J.O. Rasmuson, V.L. Roggli, F.W. Boelter, E.J. Rasmuson, C.F. Redinger
• Journal: Inhalation Toxicology
• Published: 2014;26(1):1-13
• Cited by: 28 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Context: A detailed evaluation of the correlation and linearity of industrial hygiene retrospective exposure assessment (REA) for cumulative asbestos exposure with asbestos lung burden analysis (LBA) has not been previously performed, but both methods are utilized for case-control and cohort studies and other applications such as setting occupational exposure limits.

Objective: (a) To correlate REA with asbestos LBA for a large number of cases from varied industries and exposure scenarios; (b) to evaluate the linearity, precision, and applicability of both industrial hygiene exposure reconstruction and LBA; and (c) to demonstrate validation methods for REA.

Methods: A panel of four experienced industrial hygiene raters independently estimated the cumulative asbestos exposure for 363 cases with limited exposure details in which asbestos LBA had been independently determined. LBA for asbestos bodies was performed by a pathologist by both light microscopy and scanning electron microscopy (SEM) and free asbestos fibers by SEM. Precision, reliability, correlation and linearity were evaluated via intraclass correlation, regression analysis and analysis of covariance. Plaintiff's answers to interrogatories, work history sheets, work summaries or plaintiff's discovery depositions that were obtained in court cases involving asbestos were utilized by the pathologist to provide a summarized brief asbestos exposure and work history for each of the 363 cases.

Results: Linear relationships between REA and LBA were found when adjustment was made for asbestos fiber-type exposure differences. Significant correlation between REA and LBA was found with amphibole asbestos lung burden and mixed fiber-types, but not with chrysotile. The intraclass correlation coefficients (ICC) for the precision of the industrial hygiene rater cumulative asbestos exposure estimates and the precision of repeated laboratory analysis were found to be in the excellent range. The ICC estimates were performed independent of specific asbestos fiber-type.

Conclusions: Both REA and pathology assessment are reliable and complementary predictive methods to characterize asbestos exposures. Correlation analysis between the two methods effectively validates both REA methodology and LBA procedures within the determined precision, particularly for cumulative amphibole asbestos exposures since chrysotile fibers, for the most part, are not retained in the lung for an extended period of time.

In plain English

Four experienced industrial hygienists, working independently and blind to laboratory results, were asked to estimate cumulative asbestos exposure for 363 cases from a pathologist's database, using brief job histories from medical records and legal documents. All four gave estimates for 329 cases; 328 were analyzed after excluding one ambiguous estimate. Estimates were compared with lung asbestos body and fiber counts. The four-rater average had an adjusted reliability score of 0.92; repeated laboratory measurements scored 0.88 to 0.91. Rater estimates correlated with combined lung burden (r of 0.67). For 216 cases where commercial amphibole asbestos was the majority lung fiber and no chrysotile was found, lung burden rose roughly in proportion to estimated exposure (log-scale slope not significantly different from one). It was nearly flat where amosite or crocidolite was not the majority fiber or absent. Among 11 automobile and friction-product cases, essentially no correlation appeared. The authors concluded that lung asbestos body and amphibole fiber burden analysis is a valid, linear relative indicator of past cumulative asbestiform amphibole fiber exposure. They noted their research does not settle whether disease dose-response is linear.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Exposure-assessment validation (correlation) study. Blind retrospective exposure estimates by four industrial hygienists were compared with lung burden analysis, using case histories taken from medical records and litigation documents.
• Who was studied: 363 individuals from the consultation database of one author, a Duke University pathologist. These were cases with lung fiber analysis and occupational exposure information. All four raters gave estimates for 329 cases; one was then excluded as ambiguous, leaving 328 for rater correlation and at most 327 for lung burden correlations. Fiber-type data were available for 308 of those 327. Years of collection and the cases' diagnoses are not stated.
• Exposure: Cumulative asbestos exposure in fiber per cubic centimeter years (f/cc-years), estimated from brief job descriptions and exposure durations. Examples include insulator, brake mechanic, shipyard worker, pipefitter and household contact. Lung fibers were typed as amosite, crocidolite, chrysotile, tremolite, actinolite or anthophyllite.
• Main results:
  • Raters' log-transformed estimates had an ICC of 0.76. The four-rater average had an adjusted Spearman-Brown ICC of 0.92.
  • Pairwise comparisons between individual raters gave r-values from 0.76 to 0.90.
  • Repeat laboratory analyses had ICCs above 0.75 (Table 3: 0.91 for light microscopy asbestos bodies, 0.90 for SEM asbestos bodies, 0.88 for fibers by SEM).
  • The combined lung burden measure (LAB) correlated with the average rater estimate at r of 0.67, explaining 45% of the variability.
  • Adding exposure-type categories gave r of 0.72. The common slope was 0.96, not significantly different from one (p equals 0.66).
  • Each of the top four amphibole exposure categories (shipyard insulator, power plant worker, asbestos insulator, shipyard worker) differed significantly from the automobile and friction product category.
  • With fiber-type classes added, r was 0.75. For Class 1 (commercial amphibole majority fiber, no chrysotile, n of 216), the slope on log-transformed values (1.11 in Table 6) was not significantly different from one. The authors define a slope of one on the log scale as a linear relationship.
  • Class 3 (amosite and/or crocidolite detected but not the majority fiber, n of 35) and Class 4 (no crocidolite or amosite detected, n of 40) had nearly flat slopes. These slopes were significantly different from one but not from zero.
  • For the automobile and friction product category (n of 11), regression of lung burden on rater estimate gave r of 0.01 and slope of 0.01.
  • Only six of 49 insulator cases had detectable chrysotile remaining in the lung.
• The authors' conclusion: The authors concluded that lung asbestos body and amphibole fiber burden analysis is a valid and linear relative indicator of past cumulative asbestiform amphibole exposure.
• Limitations the authors state:
  • Exposure information given to raters was limited. The authors expected better agreement with more detailed exposure information.
  • High agreement between raters does not necessarily mean high accuracy.
  • Much variability remained unexplained, even with fiber type included.
  • The refinery worker group was small (n of six), and the exact nature of their work was unknown.
  • The analysis did not include miner and miller exposures.
  • Log transformation can mask large differences between high values.
  • The study does not settle whether asbestos disease dose-response is linear or non-linear.

Topics

Medical terms in this study, explained in our Lexicon: Amphibole Asbestos, Industrial Hygiene, Pathology, Asbestos Exposure, Occupational Exposure, Exposure Reconstruction, Asbestos Bodies, Scanning Electron Microscopy (SEM).

Related studies

• Retrospective Exposure Assessment Methods Used in Occupational Human Health Risk Assessment: A Systematic Review — F. Borghi, L.A. Mazzucchelli, et al. (2020)
• An inter-laboratory study to determine the effectiveness of procedures for discriminating amphibole asbestos fibers from amphibole cleavage fragments in fiber counting by phase-contrast microscopy — M. Harper, E.G. Lee, et al. (2012)
• A novel strategy for retrospective exposure assessment in the Norwegian silicon carbide industry — S. Føreland, M.D. Bugge, et al. (2012)
• Lung fiber burden in the Nottingham gas mask cohort — G. Berry, F. Pooley, et al. (2009)
• Heavy equipment maintenance exposure assessment: using a time-activity model to estimate surrogate values for replacement of missing data — F.W. Boelter, J.W. Spencer, et al. (2007)
• The diesel exhaust in miners study: I. Overview of the exposure assessment process — P.A. Stewart, J.B. Coble, et al. (2010)

Browse more research in Inhalation Toxicology, or return to medical research.