A meta-analysis of colorectal cancer and asbestos exposure

Our plain-English summary of A meta-analysis of colorectal cancer and asbestos exposure, by D.M. Homa, D.H. Garabrant, B.W. Gillespie (American Journal of Epidemiology, 1994).

The study

• Authors: D.M. Homa, D.H. Garabrant, B.W. Gillespie
• Journal: American Journal of Epidemiology
• Published: 1994;139(12):1210-22
• Type of article: Meta-analysis
• Cited by: 33 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

A meta-analysis of the relation between asbestos exposure and colorectal cancer mortality was conducted, using published reports of 20 asbestos-exposed cohorts. Summary standardized mortality ratios (SMRs) for colorectal cancer were examined in relation to asbestos type and estimates of dust exposure (as direct estimators of asbestos exposure) and in relation to lung cancer SMR and the proportion of all deaths due to mesothelioma (as proxy estimators of asbestos exposure). An elevated summary SMR was observed in cohorts exposed to amphibole asbestos (summary SMR = 1.47; 95% confidence interval (CI) 1.09-2.00), but not in cohorts exposed to serpentine asbestos (summary SMR = 1.04; 95% CI 0.81-1.33) or in cohorts exposed to both serpentine and amphibole asbestos (summary SMR = 1.03; 95% CI 0.74-1.42). Cohorts having a lung cancer SMR greater than 2.00 had a summary SMR of 1.51 (95% CI 1.29-1.76), and cohorts in which more than 1% of all deaths were attributed to mesothelioma had a summary SMR of 1.24 (95% CI 0.94-1.64). After stratifying the cohorts based on mortality due to all cancers excluding those known or suspected to be associated with asbestos exposure, lung cancer mortality was not clearly associated with colorectal cancer mortality, suggesting that the crude association between these factors may be due to misdiagnosis of lung cancer as other types of cancer in the reported causes of death. These results suggest that exposure to amphibole asbestos may be associated with colorectal cancer, but these findings may reflect an artifact of miscertification of cause of death. The results also suggest that serpentine asbestos is not associated with colorectal cancer.

In plain English

This meta-analysis pooled colorectal cancer deaths in 20 asbestos-exposed worker cohorts from 16 published studies. The authors grouped cohorts by asbestos type (serpentine, amphibole or mixed), dust level, lung cancer standardized mortality ratio (SMR), and percentage of all deaths attributed to mesothelioma. Overall, the summary colorectal cancer SMR was 1.10 (95% confidence interval, CI, 0.92 to 1.32). The three amphibole-exposed cohorts had a summary SMR of 1.47 (95% CI 1.09 to 2.00); serpentine and mixed cohorts showed no appreciable elevation. Cohorts with a lung cancer SMR above 2.00 had a summary SMR of 1.51 (95% CI 1.29 to 1.76). The authors also split cohorts at 1.11, the median SMR for cancers other than lung, mesothelioma, esophageal, stomach and colorectal. Summary colorectal cancer SMRs were 0.85 (95% CI 0.68 to 1.06) in the lower 10 and 1.43 (95% CI 1.25 to 1.65) in the upper 10. The authors suggested this may reflect misrecorded lung cancer and mesothelioma deaths. They concluded that amphibole asbestos may be associated with colorectal cancer, possibly as an artifact of death-certificate miscertification, and that the results suggest serpentine asbestos is not.

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: Meta-analysis of published asbestos-exposed cohort mortality studies, using Poisson regression and a random effects method to compute summary SMRs
• Who was studied: 20 asbestos-exposed cohorts from 16 published studies, including shipyard, insulation, mining, asbestos cement, textile and factory workers. Results are presented for all cohorts combined; most members were male.
• Exposure: Asbestos type, classed as serpentine (chrysotile), amphibole (amosite, crocidolite) or mixed. Dust level, classed as low, moderate, high or undetermined. The lung cancer SMR and the percentage of all deaths attributed to mesothelioma were used as proxy measures of exposure.
• Main results:
  • Across all 20 cohorts, the summary colorectal cancer SMR was 1.10 by the random effects method (95% CI 0.92 to 1.32, per Table 3) and 1.02 by Poisson regression (95% CI 0.92 to 1.13, per Table 3). Considerable heterogeneity remained between studies.
  • In the three amphibole-exposed cohorts, the summary colorectal cancer SMR was 1.47 (95% CI 1.09 to 2.00).
  • Raised colorectal cancer mortality was not related to dust level or to serpentine or mixed asbestos exposure. By the random effects method (Table 3), the summary SMR was 1.04 (95% CI 0.81 to 1.33) for the 9 serpentine cohorts and 1.03 (95% CI 0.74 to 1.42) for the 8 mixed cohorts.
  • In the 8 cohorts with a lung cancer SMR greater than 2.00, the summary colorectal cancer SMR was 1.51 by the random effects method (95% CI 1.29 to 1.76, per Table 3) and 1.48 by Poisson regression.
  • In cohorts where mesothelioma made up more than 1% of all deaths, the summary colorectal cancer SMR was 1.24 by the random effects method and 1.23 by Poisson regression. The random effects confidence interval (0.94 to 1.64, per Table 3) included 1.00.
  • Cohorts were split at 1.11, the median SMR for all other cancers (excluding lung, mesothelioma, esophageal, stomach and colorectal cancer). In the 10 lower cohorts, the summary colorectal cancer SMR was 0.85 (95% CI 0.68 to 1.06, Table 5). In the 10 higher cohorts, it was 1.43 (95% CI 1.25 to 1.65, Table 5). In the lower group, one very small cohort had an elevated colorectal cancer SMR based on four deaths.
  • After taking the SMR for all other cancers into account, the association between lung cancer SMR and colorectal cancer SMR weakened appreciably.
  • Among serpentine-exposed cohorts, colorectal cancer risk showed a pattern of decreasing as dust exposure increased.
• The authors' conclusion: The authors concluded that amphibole asbestos exposure may be associated with colorectal cancer, but that this may reflect miscertification of cause of death. They also stated that the results suggest serpentine asbestos is not associated with colorectal cancer.
• Limitations the authors state:
  • Dust level and asbestos type were coded from each paper's reported values. These were not always specific or complete, so misclassification was possible.
  • All amphibole-exposed cohorts were in the high dust category, so the effects of asbestos type and dust level could not be separated.
  • Only one amphibole-exposed cohort took a latency period into account, and that period was 5 years.
  • Publication bias was possible, since negative findings on colorectal cancer may have been less likely to be published.
  • Other cancers linked to asbestos (oral cavity, larynx, kidney) were not excluded from the 'other cancer' SMR, so that SMR could be biased upward.
  • SMRs from populations with different age structures are generally not directly comparable.
  • Summary measures need cautious interpretation because the pooled SMRs were heterogeneous.

Topics

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

Medical terms in this study, explained in our Lexicon: Meta-Analysis, Colorectal Cancer, Asbestos Exposure, standardized mortality ratio (SMR), Lung Cancer, Mesothelioma, Amphibole Asbestos, confidence interval.

Related studies

• Author's response to 'Re: Exposure to asbestos and the risk of colorectal cancer mortality: a systematic review and meta-analysis by Kwak et al' — K. Kwak, D. Paek, et al. (2020)
• Asbestos exposure and neoplasia — I.J. Selikoff, J. Churg, et al. (1964)
• Occupational Asbestos Exposure and Incidence of Colon and Rectal Cancers in French Men: The Asbestos-Related Diseases Cohort (ARDCo-Nut) — C. Paris, I. Thaon, et al. (2017)
• American Cancer Society plans for a long-range study of colonic and rectal cancer — E.C. Hammond (1970)
• Exposure to asbestos and the risk of colorectal cancer mortality: a systematic review and meta-analysis — K. Kwak, D. Paek, et al. (2019)
• Antioxidants in the chemoprevention of colorectal cancer and colorectal adenomas in the general population: a systematic review and meta-analysis — D. Papaioannou, K.L. Cooper, et al. (2011)

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