'Environment' in cancer causation and etiological fraction: limitations and ambiguities

Our plain-English summary of 'Environment' in cancer causation and etiological fraction: limitations and ambiguities, by P. Boffetta, J.K. McLaughlin, C. la Vecchia, et al (Carcinogenesis, 2007).

The study

• Authors: P. Boffetta, J.K. McLaughlin, C. la Vecchia, P. Autier, P. Boyle
• Journal: Carcinogenesis
• Published: 2007;28(5):913-5
• Type of article: Review
• Cited by: 31 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Summary of the article

In this commentary, the authors examined how the term 'environment' is used in cancer research and the limitations of estimating the fraction of cancers attributable to environmental factors. They noted that 'environment' is used both broadly, for all non-genetic factors, and narrowly, for pollutants of air, water, soil and food. They proposed replacing it with 'non-genetic or modifiable determinants' and 'pollutants'. They contrasted a World Health Organization estimate that 19% of cancers worldwide were attributable to environmental factors, including occupational exposures, with other reviews, excluding occupation, that gave 1–3%. They described several sources of error: combining risk estimates from heavily exposed groups with prevalence data based on looser exposure definitions, residual confounding (for example by smoking), applying risks from past exposures to present conditions, underestimating past exposure levels in dose–response models, and ignoring effect modifiers. Non-occupational asbestos exposure was a main example. A meta-analysis of 10 residential studies gave a relative risk of mesothelioma of 3.5 (95% CI 1.8–7.0). Combined with a 5% exposure estimate for Europe, this gave an attributable fraction of 11%, or 174 cases per year in men and 30 in women in a country like Great Britain, which the authors considered highly probably an overestimate. An earlier nickel example was also discussed. The authors concluded that errors combined in one direction could inflate estimates by an order of magnitude, and they urged caution.

PubMed has no abstract for this study, and the article doesn't print one. We wrote this summary from the full article with the help of AI. It is not the authors' wording.

In plain English

This was a commentary, not a new study. The authors argued that the word "environment" is used ambiguously in cancer research. They described errors that can inflate estimates of the share of cancers caused by pollutants. As an asbestos example, they cited an earlier meta-analysis of 10 studies of non-occupational asbestos exposure. Most of those studies looked at people living in mining areas or near asbestos plants or other sources. That meta-analysis found a relative risk of mesothelioma of 3.5 (95% confidence interval 1.8–7.0). The authors combined this with a World Health Organization figure that 5% of people in Europe have non-occupational exposure. The result was that 11% of mesothelioma would be attributed to it, or 174 cases a year in men and 30 in women in a country like Great Britain. The authors said this was highly probably an overestimate. The 5% figure used a broader definition of exposure, and the relative risk ignored possible confounding by work in the asbestos industry. They also explained how underestimating past high exposures can overstate risk at low doses. The authors concluded that such estimates should be used only as benchmarks for comparing risk factors.

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 / methodological discussion (no new data collected)
• Who was studied: No study population of its own. As an example, the article lists 10 published epidemiological studies of mesothelioma and non-occupational asbestos exposure, from South Africa, Canada, Italy, Spain, Switzerland, the UK, New Caledonia and Australia.
• Exposure: Non-occupational (residential or environmental) asbestos exposure, defined in the listed studies as living in a mining area, living near an asbestos cement plant, asbestos factory or other potential source, living in a polluted city, or using contaminated building materials. The listed studies were chosen to reflect high exposure.
• Main results:
  • An earlier meta-analysis (the article's reference 5) of the 10 listed studies found an overall relative risk of mesothelioma of 3.5 (95% confidence interval 1.8–7.0) for non-occupational asbestos exposure.
  • The authors combined the relative risk of 3.5 with a 5% exposure frequency. This gave an 11% fraction of mesothelioma attributable to non-occupational exposure, corresponding to 174 cases a year among men and 30 among women in a country like Great Britain.
  • The authors judged that the 11% figure is highly probably an overestimate of the true effect of non-occupational asbestos exposure.
  • The 5% European exposure figure, from a World Health Organization air quality report, used a broader definition of asbestos exposure than the studies behind the relative risk.
  • The relative risk of 3.5 ignores possible confounding by employment in the asbestos industry, which not all of the listed studies excluded.
  • Underestimating past high exposures overstates the slope of the dose–response relationship and the estimated risk at low doses. The article notes this problem has been discussed by other researchers for retrospective assessment of occupational asbestos exposure.
• The authors' conclusion: The authors concluded that ambiguous terminology should be avoided and that estimates of attributable cancers should be treated with caution and used only as benchmarks for the relative importance of risk factors.

Topics

Medical terms in this study, explained in our Lexicon: Causation, Occupational Exposure, Confounding, Asbestos Exposure, Meta-Analysis, Relative Risk, Mesothelioma.

Related studies

• 'Environment' in cancer causation and etiological fraction: limitations and ambiguities — D. Mirabelli, E. Merler (2008)
• Elongate mineral particles in the natural environment — M.E. Gunter (2018)
• Children's environment in Central Europe: threats and chances — S. Böse-O'Reilly, U. Heudorf, et al. (2007)
• Evaluation of asbestos exposure in the working environment. Recommendation of subcommittee on asbestosis of the permanent commission and international association on occupational health — (1972)
• Unbiased vs. conservative estimators of etiological fractions: examples of misclassification from studies of occupational lung cancer — O. Olsen (1995)
• The future excess fraction of occupational cancer among those exposed to carcinogens at work in Australia in 2012 — R.N. Carey, S.J. Hutchings, et al. (2017)

Browse more research in Carcinogenesis, or return to medical research.