Examining the association of lung cancer and highly correlated fibre size-specific asbestos exposures with a hierarchical Bayesian model

Our plain-English summary of Examining the association of lung cancer and highly correlated fibre size-specific asbestos exposures with a hierarchical Bayesian model, by G.B. Hamra, D. Loomis, J. Dement (Occupational and Environmental Medicine, 2014).

The study

• Authors: G.B. Hamra, D. Loomis, J. Dement
• Journal: Occupational and Environmental Medicine
• Published: 2014;71(5):353-7
• Cited by: 13 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Background: Asbestos is a known carcinogen. However, little is known about the differential effects of size-specific asbestos fibres. Previous research has examined the relationship with lung cancer of each fibre group in the absence of others. Attempts to model all fibre groups within a single regression model have failed due to high correlations across fibre size groups.

Methods: We compare results from frequentist models for individual fibre size groups, and a hierarchical Bayesian model that included all fibre groups to estimate the relationship of size-specific asbestos fibre groups to lung cancer mortality. The hierarchical model assumes partial exchangeability of the effects of size-specific asbestos fibre groups to lung cancer, and is capable of handling the strong correlation of the exposure data.

Results: When fibre groups are modelled independently with a frequentist model, there appears to be an increase in the dose-response with increasing fibre size. However, when subject to a hierarchical structure, this trend vanishes, and the effects of distinct fibre groups appear largely similar.

Conclusions: This is the first occasion where distinct asbestos fibre groups have been assessed in a single regression model; however, even the use of a hierarchical modelling structure does not appear to overcome all the statistical fluctuations arising from the high correlations across fibre groups. We believe these results should be compared with other occupational cohorts with similar fibre group information. Finally, results for the smallest fibre group may be suggestive of a carcinogenic potential for nanofibres.

In plain English

The authors re-analysed a North Carolina asbestos textile cohort (2419 men and 1384 women in production jobs, 1950 to 1973) to examine chrysotile fibre size and lung cancer deaths. Workers whose exposure could not be estimated (1969 people) were excluded. Using 181 lung cancer cases and matched controls, they compared separate models for each of 21 fibre size groups with one hierarchical Bayesian model including all groups. Exposures to the groups were moderately to highly correlated (0.40 to 0.99). Separate models suggested effects rose with fibre length and diameter; the hierarchical model did not show this trend. Most groups were close to a shared median of 0.0603 (change in log odds of lung cancer per 100 fibre-years/mL). The smallest fibres (under 0.25 μm wide, under 1.5 μm long) had the most precise estimate, 0.0619. The authors considered one group's negative estimate an artefact of the correlations. They concluded most size groups share a similar relationship with lung cancer risk. They called the analyses preliminary, noted missing smoking information, and said the results do not support using the coefficients for risk estimation.

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: Nested case-control analysis within an occupational cohort, with 10 age-matched controls per lung cancer case. Separate frequentist logistic models for each fibre size group were compared with a single hierarchical Bayesian model including all fibre size groups.
• Who was studied: A cohort described as 2419 men and 1384 women employed in production jobs at North Carolina, USA asbestos textile facilities for at least 1 day between 1 January 1950 and 31 December 1973. Workers whose occupational exposure history could not be estimated (n=1969) were excluded from the analysis. The article does not say whether this exclusion applies to the 3803 listed workers or to a larger original group. Follow-up ran through 31 December 2003. The analysis included 181 lung cancer cases.
• Exposure: Cumulative exposure to chrysotile asbestos, with limited amosite use at one facility between 1963 and 1976. Fibres were divided by transmission electron microscopy into 4 diameter and 6 length groups using 77 dust samples from 1964 to 1971. 21 of the 24 groups had observations. Exposure was expressed in fibre-years/mL with a 10-year lag.
• Main results:
  • Correlations between exposures to the different fibre size groups ranged from 0.40 to 0.99, with a median of 0.79.
  • The smallest chrysotile fibre group made up 53% of all fibres counted.
  • In separate frequentist models for fibres 10–20 μm long, coefficients (SE) rose with diameter: 0.1736 (0.0965), 0.3722 (0.1907), 1.0288 (0.6278) and 3.2620 (1.8527) for diameters <0.25, 0.25–1.0, 1.0–3.0 and >3.0 μm.
  • In the hierarchical model, no increasing dose-response trend with fibre size was seen.
  • The shared (second-stage) median coefficient was 0.0603 (SD 0.1378) per 100 f-y/mL, and most fibre groups had medians near it.
  • Across fibre size groups, the results showed a great deal of similarity: in the hierarchical model, most groups' medians were near the shared median.
  • The smallest fibre group (diameter <0.25 μm, length <1.5 μm) had the most precise estimate in the hierarchical model: 0.0619 (SD 0.0362).
  • The group with diameter <0.25 μm and length 1.5–5.0 μm had a negative estimate, -0.2353 (SD 0.1411), in the hierarchical model.
• The authors' conclusion: The authors concluded that, under the hierarchical model, most chrysotile fibre size groups share a similar relationship with lung cancer risk.
• Limitations the authors state:
  • There was no information on tobacco smoking, and a medical surveillance programme may have had an effect.
  • Fibre size estimates were based on dust samples from 1964–1971 only.
  • Only a limited number of samples could be analysed, creating unknown uncertainty in fibre number and concentration estimates.
  • Textile workers may differ from other industries in exposure response and fibre size distribution.
  • The authors believe the apparent protective (negative) estimate is an artefact of the strong correlations between fibre groups.
  • The authors described the analyses as preliminary and needing replication in other cohorts.
  • The results could not be compared directly with a non-hierarchical model including all fibre groups.
  • The authors stated the results do not support using the size-specific coefficients for risk estimation.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Lung cancer.

Medical terms in this study, explained in our Lexicon: Lung Cancer, Asbestos Exposure, Dose-response relationship, Carcinogenic.

Related studies

• Development of a fibre size-specific job-exposure matrix for airborne asbestos fibres — J.M. Dement, E.D. Kuempel, et al. (2008)
• Pulmonary effects in rats inhaling size-separated chrysotile asbestos fibres or p-aramid fibrils: differences in cellular proliferative responses — D.B. Warheit, M.A. Hartsky, et al. (1996)
• An epidemiological study of the role of chrysotile asbestos fibre dimensions in determining respiratory disease risk in exposed workers — L. Stayner, E. Kuempel, et al. (2008)
• Hierarchical latency models for dose-time-response associations — D.B. Richardson, R.F. MacLehose, et al. (2011)
• Airborne fiber size characterization in exposure estimation: Evaluation of a modified transmission electron microcopy protocol for asbestos and potential use for carbon nanotubes and nanofibers — J.M. Dement, E.D. Kuempel, et al. (2015)
• Size- and type-specific exposure assessment of an asbestos products factory in China — M.N. Courtice, D.W. Berman, et al. (2016)

Browse more research in Occupational and Environmental Medicine, or return to medical research.