Multistage modelling of lung cancer mortality in asbestos textile workers

Our plain-English summary of Multistage modelling of lung cancer mortality in asbestos textile workers, by N. Pearce (International Journal of Epidemiology, 1988).

The study

• Author: N. Pearce
• Journal: International Journal of Epidemiology
• Published: 1988;17(4):747-52
• Cited by: 19 articles, as counted by NIH iCite
• Read it: PubMed · Publisher
• In J-M's library: Johns-Manville's company medical library held a copy (J-M no. 000927).

Abstract

The Armitage-Doll multistage model of cancer is applied to data from a cohort study of lung cancer in 1261 white male workers from one asbestos textile manufacturing plant. Three approaches are used: induction time analysis; analysis of the relationship of the excess incidence rate to age at first exposure and time since first exposure; and direct fitting of the Armitage-Doll model. Poisson regression was used for all analyses. The induction time analysis was conducted using Rothman's 'window of exposure' method. This suggested that the increase in rate ratio was primarily due to exposures occurring 15-24 years previously, whereas there was little effect from exposures occurring 0-14 or 25+ years previously. The excess incidence rate increased both with age at first exposure and time since first exposure, suggesting that asbestos acted at a stage intermediate between the first and penultimate stages. Direct fitting of the Armitage-Doll model suggested that the best fit was obtained by assuming that asbestos acts at stage 3, 4 or 5 of a six-stage process. Most analyses of the type presented here are unable to determine at which stage a carcinogen acts, due to the small numbers of cancer deaths occurring in typical occupational cohorts. Furthermore, there is reason to doubt the validity of the Armitage-Doll model. However, such analyses can at least suggest whether a carcinogen appears to act at an early, intermediate or late stage, and the general statistical methods applied here will retain their usefulness as further models are developed and larger data sets become available.

In plain English

This study reanalyzed data from an earlier cohort study by Dement and colleagues. That cohort was 1261 white male workers employed for at least one month between 1940 and 1965 at one asbestos textile manufacturing plant, followed through 1975. The author used each worker's estimated cumulative asbestos exposure, in fibres per cc days, and a statistical method called Poisson regression to test the Armitage-Doll multistage model, a theory that cancer develops through a sequence of cell changes. When each period of past exposure was adjusted for the others, the increase in lung cancer rate ratio appeared mainly due to exposure 15 to 24 years earlier, though these estimates were imprecise. Lung cancer was strongly associated with cumulative exposure. The excess rate rose with both age at first exposure and time since first exposure. Fitting the model directly suggested asbestos acted at stage 3, 4 or 5 of a six-stage process. The author cautioned that there were too few lung cancer deaths to tell the models apart, that individual smoking data were lacking, and that the model's validity is in doubt. The author concluded that such analyses can at least suggest whether a carcinogen acts at an early, intermediate or late stage.

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: Statistical reanalysis of an existing retrospective cohort, applying the Armitage-Doll multistage model of cancer with Poisson regression: an induction time (exposure window) analysis, analysis by age at first exposure and time since first exposure, and direct model fitting
• Who was studied: 1261 white male workers employed for at least one month between 1 January 1940 and 31 December 1965 at one asbestos textile manufacturing plant (cohort originally studied by Dement et al), followed from 1940 to 1975. Expected deaths were based on US white male rates.
• Exposure: Asbestos in a textile plant (the text does not name the fiber type). Cumulative exposure was estimated in fibres/cc days by linking job histories to air measurements. Exposure windows compared 5000 or more with less than 5000 fibre/cc days.
• Main results:
  • In the induction time analysis, adjusted for exposures in other windows, the increase in lung cancer rate ratio appeared mainly due to exposures 15-24 years earlier, with little effect from exposures 0-14 or 25 or more years earlier.
  • For the 20-24 year window, workers with 5000 or more fibre/cc days in that window had an adjusted rate ratio of 4.6 (95% limits 1.3-16.3) compared with less than 5000 fibre/cc days, adjusted for age, calendar year and exposure in other windows.
  • Lung cancer incidence was strongly associated with cumulative asbestos exposure. In the highest cumulative exposure group (100 or more thousand fibre/cc days), 2 lung cancer deaths were observed against 0.13 expected, an SMR of 15.39.
  • The excess rate appeared to increase with age at first exposure, which the author took to suggest asbestos does not act solely at the first stage.
  • The excess rate appeared to increase with time since first employment, which the author took to suggest asbestos does not act solely at the penultimate stage.
  • In direct fitting of the model, the categorical analyses gave the best fit when asbestos was assumed to act at stage 3, 4 or 5 of a six-stage process.
  • The continuous analyses suggested that asbestos acts at stage 4 or 5.
• The authors' conclusion: The author concluded that, despite doubts about the validity of the Armitage-Doll model, such analyses can at least suggest whether a carcinogen acts at an early, intermediate or late stage.
• Limitations the authors state:
  • Most occupational data sets, including this one, do not have enough cancer cases to distinguish between the forms of the model.
  • Results of the direct model fitting should be viewed with considerable reservation because of the small number of lung cancer deaths.
  • Small numbers forced a two-category cumulative exposure classification in the induction time analysis.
  • The induction time estimates were imprecise. The adjusted rate ratios must be interpreted with caution because they assume a multiplicative effect of exposures at different times.
  • Individual smoking data were lacking. The author judged smoking unlikely to be an important confounder, because survey data showed smoking prevalence similar to US white males, but noted it could have been an effect modifier.
  • Interpreting the Armitage-Doll model is problematic because it was derived empirically and lacks a biological foundation.
  • Exposure data, not organ doses, were used, and the relationship between exposure and dose may not be linear.
  • These analyses cannot tell apart the various intermediate stages at which a carcinogen might act.

Topics

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

Medical terms in this study, explained in our Lexicon: Lung Cancer, Cohort Study, Incidence, rate ratio.

Related studies

• Lung cancer mortality in North Carolina and South Carolina chrysotile asbestos textile workers — L. Elliott, D. Loomis, et al. (2012)
• Cause-specific mortality in a Chinese chrysotile textile worker cohort — X. Wang, S. Lin, et al. (2013)
• Increased lung cancer mortality among chrysotile asbestos textile workers is more strongly associated with exposure to long thin fibres — D. Loomis, J.M. Dement, et al. (2012)
• Lung cancer mortality among asbestos textile workers: a review and update — J.M. Dement, D.P. Brown (1994)
• Follow-up study of chrysotile asbestos textile workers: cohort mortality and case-control analyses — J.M. Dement, D.P. Brown, et al. (1994)
• Cancer mortality among Chinese chrysotile asbestos textile workers — X.R. Wang, I.T. Yu, et al. (2012)

Browse more research in International Journal of Epidemiology, or return to medical research.