A quantitative weight of evidence assessment of Hill's guidelines for causal inference for cosmetic talc as a cause of mesothelioma

Our plain-English summary of A quantitative weight of evidence assessment of Hill's guidelines for causal inference for cosmetic talc as a cause of mesothelioma, by A.M. Ierardi, A. Urban, G.M. Marsh (Toxicology and Applied Pharmacology, 2021).

The study

• Authors: A.M. Ierardi, A. Urban, G.M. Marsh
• Journal: Toxicology and Applied Pharmacology
• Published: 2021;417:115461
• Type of article: Review
• Cited by: 6 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Cosmetic talc has been suggested to cause mesothelioma. To assess a potential causal relationship between cosmetic talc and mesothelioma, a quantitative weight of evidence analysis was performed in accordance with Hill's nine original guidelines for causal inference using a published empirical model to weight each respective guideline. Various epidemiological, toxicological, and exposure studies related to cosmetic talc and risk of mesothelioma were included in an evaluation of each of Hill's guidelines. Probabilities that the guidelines were true were assigned based on expert judgment. We applied a sensitivity analysis to evaluate the variability of our probability estimates. The overall probability of causality for cosmetic talc and mesothelioma was approximately 1.29% (range: 0.73%-3.96%). This low probability of causality supports the conclusion that cosmetic talc is not related to the development of mesothelioma.

Copyright © 2021 Elsevier Inc. All rights reserved.

In plain English

This article was a quantitative weight of evidence assessment of earlier research, not a new study of people. The authors gathered published and unpublished epidemiological, toxicological and exposure studies on cosmetic talc and mesothelioma and sorted them under Hill's nine guidelines for judging causation. They used expert judgment to give each guideline a probability of being true, then entered these into a published model to estimate an overall probability of causality. Much of the evidence came from cosmetic talc miner and miller cohorts: a 2020 pooled analysis had found one confirmed mesothelioma case among 4151 workers. The calculated probability that cosmetic talc causes mesothelioma was 1.29%. In a sensitivity analysis of 243 combinations, results ranged from 0.73% to 2.27% when inputs varied by 50%, and from 0.41% to 3.96% when they varied by 100%. The authors concluded that the evidence supports the conclusion that cosmetic talc is not related to the development of mesothelioma. They noted that the probability assignments are subjective, and that the model was built on data unlike the cosmetic talc data.

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: Quantitative weight of evidence assessment (a structured review of existing literature) applying Hill's nine guidelines for causal inference. The authors weighted the guidelines with the empirical model of Swaen and van Amelsvoort (2009) and added a sensitivity analysis.
• Who was studied: No new participants. The evidence was existing literature: epidemiological, toxicological and exposure studies, including unpublished reports, book chapters and government documents. Much of it concerned five cohorts of cosmetic talc miners and millers in Italy, Norway, France, Austria and Vermont, which a previous pooled analysis (Ierardi and Marsh, 2020) put at 4151 workers and 130,514 person-years. No search dates were given.
• Exposure: Cosmetic talc, both from consumer use of talcum powder products and from occupational exposure of talc miners and millers. The article discusses allegations that talc may contain asbestiform tremolite, anthophyllite and possibly chrysotile.
• Main results:
  • The authors assigned these probabilities by expert judgment:
• Strength: 20%
• Consistency: 0%
• Specificity: 0%
• Temporality: 100%
• Biological Gradient: 30%
• Plausibility: 0%
• Coherence: 30%
• Experiment: 5%
• Analogy: 30%

Consistency, Specificity and Plausibility were fixed at 0%, and Temporality at 100%.

• For the Strength and Consistency guidelines, the authors relied on a previous pooled analysis (Ierardi and Marsh, 2020) of the cosmetic talc miner and miller cohorts. That analysis reported one confirmed mesothelioma case, reported a pooled SMR of 0.299 (95% CI: 0.015, 1.42), and described this as a non-significant deficit.
• Using the central estimates, the overall probability that cosmetic talc causes mesothelioma was 1.29%.
• Sensitivity analysis over 243 combinations of probabilities, with a median of 1.29% under both approaches:
• Approach 1 (inputs varied by plus or minus 50%): minimum 0.73%, maximum 2.27%.
• Approach 2 (inputs varied by plus or minus 100%): minimum 0.41%, maximum 3.96%.
  • The most conservative assignment of probabilities gave a 3.96% probability of a causal association.
  • When the Temporality and Analogy probabilities were both set to 0%, the overall probability was 2.58%.
• The authors' conclusion: The authors concluded that the available evidence supports the conclusion that cosmetic talc is not related to the development of mesothelioma.
• Limitations the authors state:
  • Assigning probabilities to each guideline is subjective and based on expert judgment, as the model's developers recognized. The authors addressed this with a sensitivity analysis.
  • Some researchers argue that a Hill analysis should not be done without an initial positive association between exposure and disease, and no such association exists here.
  • The weighting model was derived only from IARC Group 1 and 2A carcinogens. Its developers recommended using it only for similar datasets, and the authors stated that the cosmetic talc and mesothelioma data do not resemble those.
  • Two guidelines (Temporality and Analogy) behave counterintuitively in the model: raising their probabilities slightly lowers the overall probability of causality.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Mesothelioma, Talc and vermiculite.

Medical terms in this study, explained in our Lexicon: Cosmetic Talc, Mesothelioma.

Related studies

• Cosmetic talc as a risk factor for pleural mesothelioma: a weight of evidence evaluation of the epidemiology — B.L. Finley, S.M. Benson, et al. (2017)
• Re: Brent L. Finley, Stacey M. Benson & Gary M. Marsh (2017): Cosmetic talc as a risk factor for pleural mesothelioma: a weight of evidence evaluation of the epidemiology, Inhalation Toxicology, DOI: 10.1080/08958378.2017.1336187 — M.M. Finkelstein (2017)
• Letter on: "Cosmetic talc as a risk factor for pleural mesothelioma: a weight of evidence evaluation of the epidemiology" — D. Mirabelli (2017)
• Response to letters regarding "Cosmetic talc as a risk factor for pleural mesothelioma: a weight of evidence evaluation of the epidemiology" — B.L. Finley, S.M. Benson, et al. (2018)
• Toxicology and epidemiology: improving the science with a framework for combining toxicological and epidemiological evidence to establish causal inference — H.O. Adami, S.C. Berry, et al. (2011)
• Evidence Does Not Support Exposure to Cosmetic Talc as Cause of Malignant Mesothelioma — S.J. Geyer (2020)

Browse more research in Toxicology and Applied Pharmacology, or return to medical research.