Attributable risk in men in two French case-control studies on mesothelioma and asbestos

Our plain-English summary of Attributable risk in men in two French case-control studies on mesothelioma and asbestos, by A. Lacourt, P. Rolland, C. Gramond, et al (European Journal of Epidemiology, 2010).

The study

• Authors: A. Lacourt, P. Rolland, C. Gramond, P. Astoul, S. Chamming's, S. Ducamp, C. Frenay, F. Galateau-Sallé, A. Gilg Soit Ilg, E. Imbernon, N. Le Stang, J.C. Pairon, M. Goldberg, Y. Iwatsubo, L.R. Salmi, P. Brochard
• Journal: European Journal of Epidemiology
• Published: 2010;25(11):799-806
• Cited by: 19 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Pleural mesothelioma is a primary tumor of the pleura that is mainly due to asbestos exposure. To study the relationship between mesothelioma and occupational asbestos exposure in France, two case-control studies (A and B) were conducted. A substantial difference in the attributable risk in the population (AR(p)) was observed among men: 44.5% (95% CI: [32.6-56.4]) in study A and 83.2% (95% CI: [76.8-89.6]) in study B. As different exposure assessment expert methods were used, the main objective of this work was to re-estimate the AR(p) men in two case-control studies according to a common standardized exposure assessment by using a Job Exposure Matrix (JEM) and to assess the role of subjects' selection. The initial observed AR(p) difference was maintained: 36.3% (95% CI: [24.3-50.3]) in study A and 69.7% (95% CI: [51.7-83.2]) in study B. Further investigations highlighted the potential selection bias introduced in both studies, especially among controls. The AR(p) could be underestimated in study A and overestimated in study B. After weighting subjects according to distribution of socio-economic status in the general population for controls and according to distribution of socio-economic status of cases registered by the French National Mesothelioma Surveillance Program, re-estimated AR(p) values were 52.4% in study A and 70.2% in study B. These results provide additional information to describe the relationship between pleural mesothelioma and occupational asbestos exposure, but also confirm the importance of subjects' recruitment in case control studies, particularly control selection.

In plain English

This article re-analysed men from two French case-control studies of pleural mesothelioma. Study A was hospital-based: 460 male cases, 421 controls. Study B was population-based, drawn from a national surveillance program, with 371 male cases and 732 controls. Earlier analyses disagreed on the attributable risk in the population: the share of cases attributable to occupational asbestos exposure. With one job exposure matrix applied to both, the gap remained: 36.8% (95% confidence interval 18.8 to 50.8) in study A and 72.3% (63.8 to 78.8) in study B. Odds of mesothelioma rose with cumulative exposure in both, more steeply in study B. Controls in both studies differed from the general population in socioeconomic status; cases did not differ significantly on that measure from surveillance-program cases. With cases and controls reweighted, estimates were 52.4% (37.9 to 63.5) in study A and 70.2% (60.7 to 77.4) in study B. The authors concluded the difference was probably due to how subjects, especially controls, were recruited, and that study B seemed more reliable. Selection bias, they added, could not explain all of it.

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: Re-analysis of two multicenter case-control studies, one hospital-based (study A) and one population-based (study B)
• Who was studied: Men in France with pleural mesothelioma and matched controls: study A had 460 male cases and 421 controls; study B had 371 male cases and 732 controls
• Exposure: Occupational asbestos exposure, assessed in both studies with the same job exposure matrix (probability, frequency and intensity of exposure, combined into a cumulative exposure index in fibers/ml-years). No specific fiber type was analysed.
• Main results:
  • The difference between the two studies seen with the original expert exposure assessments persisted after the standardized assessment.
  • Using the job exposure matrix, the attributable risk in the population for occupational asbestos exposure in men was 36.8% (95% CI 18.8 to 50.8) in study A and 72.3% (95% CI 63.8 to 78.8) in study B.
  • For definite exposure, the text reports odds ratios of 2.9 (95% CI 2.0 to 4.2) in study A and 10.6 (95% CI 7.2 to 15.6) in study B.
  • In study B, the text reports odds ratios rising from 1.8 (95% CI 1.2 to 2.9) for men with less than 0.1 f/ml-years to 19.6 (95% CI 11.3 to 34.0) for those with more than 10 f/ml-years.
  • In study A, the text reports odds ratios rising from 1.3 (95% CI 0.9 to 2.0) in the lowest cumulative exposure category to 5.3 (95% CI 2.8 to 10.1) in the highest.
  • Controls in both studies differed in socioeconomic status from the general population. Study B controls included 32.1% blue-collar workers, compared with 41.1% in the general population.
  • Cases from both studies did not differ significantly in socioeconomic status from cases registered by the national surveillance program (p greater than 0.05). This comparison covered socioeconomic status only.
  • When only controls were reweighted to match the general population (simulation 1), the attributable risk was 50.1% (95% CI 36.3 to 60.9) in study A and 70.1% (95% CI 60.6 to 77.3) in study B.
  • When cases and controls were both reweighted (simulation 3), the attributable risk was 52.4% (95% CI 37.9 to 63.5) in study A and 70.2% (95% CI 60.7 to 77.4) in study B.
• The authors' conclusion: The authors concluded that the difference in attributable risk between the two studies is probably due to how subjects, particularly controls, were recruited. They also judged that study B's results seem more reliable than study A's.
• Limitations the authors state:
  • Selection bias cannot explain the entire difference between the studies. Confounding by another type of asbestos exposure or another cause should be considered.
  • Only occupational asbestos exposure was considered. Domestic or environmental asbestos exposure could have been a confounding factor.
  • The job exposure matrix produces misclassification, so odds ratios and attributable risk were under-estimated, similarly in both studies.
  • Differences in the validity and reliability of interview-based job histories between the studies cannot be ruled out.
  • The cumulative exposure index is limited by retrospective exposure assessment.
  • The true attributable risk cannot be precisely determined from these data.

Topics

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

Medical terms in this study, explained in our Lexicon: Attributable Risk, Mesothelioma, Pleural Mesothelioma, Pleura, Asbestos Exposure, Selection Bias.

Related studies

• Occupations and industries in France at high risk for pleural mesothelioma: A population-based case-control study (1998-2002) — P. Rolland, C. Gramond, et al. (2010)
• Occupational and non-occupational attributable risk of asbestos exposure for malignant pleural mesothelioma — A. Lacourt, C. Gramond, et al. (2014)
• Compensation of pleural mesothelioma in France: data from the French National Mesothelioma Surveillance Programme — S. Chamming's, B. Clin, et al. (2013)
• Occupational exposure to unintentionally emitted nanoscale particles and risk of cancer: From lung to central nervous system - Results from three French case-control studies — G. Manangama, C. Gramond, et al. (2020)
• Temporal patterns of occupational asbestos exposure and risk of pleural mesothelioma — A. Lacourt, K. Leffondré, et al. (2012)
• Characterization of a French series of female cases of mesothelioma — E. Camiade, C. Gramond, et al. (2013)

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