Additional thoughts on the review of epidemiologic studies related to ingested asbestos

Our plain-English summary of Additional thoughts on the review of epidemiologic studies related to ingested asbestos, by G.M. Marsh (Environmental Health Perspectives, 1983).

The study

• Author: G.M. Marsh
• Journal: Environmental Health Perspectives
• Published: 1983;53:185-7
• Cited by: 1 article, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

In plain English

This was a commentary, not a new study. The author answered questions raised after presenting an earlier critical review of 13 epidemiologic studies related to ingested asbestos. For a Connecticut study with very low fiber concentrations, the author wrote that, with no established dose-response relationship, no exposure level could be called insufficient to raise cancer rates. The author judged that study's regression assumptions were probably not completely satisfied but suspected the departures did not drastically alter its conclusions. For the San Francisco Bay Area studies, the author described another researcher's (Tarter's) critique of their log transformation. Tarter reported that his form of it, with a larger constant (K = 32 instead of 0.01), greatly enhanced the significance of the drinking water cancer association. The author explained that the 'points' tallying study weaknesses were purely descriptive. Probability calculations testing whether positive findings could be due to chance were meant only as a rough guide for future research. The author concluded that uncertainty would remain until hypotheses from ecologic studies were tested at the individual level.

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: commentary (the author's answers to questions about the author's own earlier review of ingested-asbestos studies); no new data
• Who was studied: No study population; the commentary discusses the 13 epidemiologic studies covered in the author's earlier review, including the Connecticut and San Francisco Bay Area drinking-water studies
• Exposure: ingested asbestos; the studies discussed concern waterborne asbestos in drinking water
• Main results:
  • On the low fiber concentrations in Connecticut, the author wrote that, since no dose-response relationship between ingested asbestos and cancer had been established, no exposure level could be characterized as insufficient to raise cancer above background
  • The author suspected that departures from multiple regression assumptions in the Connecticut study were not severe enough to drastically alter its reported conclusions
  • For the San Francisco Bay Area studies, the author reported Tarter's finding (not the author's own) that using Tarter's form of the log transformation with a large constant, K = 32 rather than the K = 0.01 used by Kanarek et al., greatly enhanced the overall significance level of the asbestos in drinking water cancer association
  • The 'points' totals in the author's table of study weaknesses were purely descriptive and did not influence how the findings were integrated
  • The author stated that the results of the probability (PD) calculations, which assessed whether positive male and female findings could be due to chance, were meant only as a rough guide for future research, and that the ranking based on them did not depend on the choice of p (0.05 or 0.10)
  • Because of very small sample sizes, coefficients measuring agreement between male and female findings could not be computed for most cancer sites
• The authors' conclusion: The author concluded that more sophisticated analysis can strengthen causal inference from ecologic data, but some uncertainty will remain until hypotheses from ecologic studies are tested more definitively at the individual level
• Limitations the authors state:
  • The PD approach in the author's review was limited by small sample size and by the need to assume the studies provided equivalent information
  • Regression assumptions in the Connecticut study, such as negligible measurement error and linearity, were likely not completely satisfied

Related studies

• Critical review of epidemiologic studies related to ingested asbestos — G.M. Marsh (1983)
• Comparing epidemiologic studies of ingested asbestos for use in risk assessment — L.S. Erdreich (1983)
• The epidemiologic evidence for elongate mineral particle (EMP)-related human cancer risk — P. Boffetta, K.A. Mundt, et al. (2018)
• Additional notes on the case-control study in western Washington on the cancer risk from asbestos in drinking water — L. Polissar, R.K. Severson, et al. (1983)
• Additional comments on the study of the effect of population density on the results of the California Bay area study — P.M. Conforti (1983)
• Human cancer risk from ingested asbestos: a problem of uncertainty — W.J. Nicholson (1983)

Browse more research in Environmental Health Perspectives, or return to medical research.