Asbestos-related lung cancer mortality in Piedmont, Italy

Our plain-English summary of Asbestos-related lung cancer mortality in Piedmont, Italy, by M. Martuzzi, P. Comba, M. De Santis, et al (American Journal of Industrial Medicine, 1998).

The study

• Authors: M. Martuzzi, P. Comba, M. De Santis, I. Iavarone, M. Di Paola, M. Mastrantonio, R. Pirastu
• Journal: American Journal of Industrial Medicine
• Published: 1998;33(6):565-70
• Cited by: 16 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

This study investigates the association between pleural neoplasm mortality, a possible proxy for asbestos exposure, and lung cancer mortality among males resident in Piedmont (northwestern Italy). Pleural neoplasm mortality was estimated in the 1,209 municipalities of the region for the period 1980-1992, applying Bayesian methods. The association with lung cancer mortality for municipalities was studied using Poisson regression. Urban/rural indicators and altitude were also included in the analysis. A positive, statistically significant association was found between pleural neoplasm and lung cancer mortality (beta = 0.025, P < 0.001); lung cancer risk was associated also with urban status (vs. rural, beta = 0.223, P < 0.001) and altitude (P = 0.01). The proportion of lung cancer deaths attributable to living in municipalities with increased pleural neoplasm mortality was 3.9% (95% confidence interval, 2.1-5.7%). The observed association and the presence of a dose-response relationship at the municipality level underscore the dangers of asbestos for human health.

In plain English

This geographical study compared death rates among males in the 1,209 municipalities of Piedmont, northwestern Italy, in 1980–1992 (2,239,516 males resident at the 1981 census). Lacking individual exposure records, the authors used each municipality's estimated relative risk of death from malignant pleural neoplasm, calculated with empirical Bayes methods, as an approximate measure of asbestos exposure. They tested its association with lung cancer deaths, allowing for urban or rural status and altitude. Each unit increase in pleural neoplasm relative risk was associated with an estimated 2.5% excess in lung cancer mortality (95% confidence interval 1.4 to 3.7%). Assuming no background pleural neoplasm deaths, the authors estimated that 3.9% of lung cancer deaths (95% confidence interval 2.1 to 5.7%), or 1,016.7 deaths, were attributable to living in municipalities with high pleural neoplasm mortality. With a background rate of 2 per million per year, it was 3.7%. The association held without Casale Monferrato or without Turin. The authors suggested that asbestos effects on lung cancer known from occupational studies are also observable in the general population. Smoking and individual case data were unavailable.

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: Geographical (ecological) study of mortality data at the municipality level, using Bayesian estimates of pleural neoplasm mortality and Poisson regression
• Who was studied: Males living in the Piedmont region of northwestern Italy, 1980–1992: 2,239,516 males resident at the 1981 census (558,577 in Turin), across 1,209 municipalities
• Exposure: Asbestos exposure was not measured directly. The authors used each municipality's relative risk of pleural neoplasm mortality, estimated with empirical Bayes methods, as an approximate measure of asbestos exposure. No fiber types are specified. Urban/rural status and altitude were included as possible confounders.
• Main results:
  • Piedmont males had 794 observed deaths from malignant pleural neoplasm against 516.4 expected (overall SMR 1.5), and 26,415 lung cancer deaths against 26,512.5 expected (overall SMR 1.0). There were no pleural neoplasm deaths in 945 (78.1%) of the 1,209 municipalities.
  • In the multivariate analysis, pleural neoplasm mortality was positively and statistically significantly associated with lung cancer mortality (coefficient 0.025, p less than 0.001). Each unit increase in pleural neoplasm mortality relative risk corresponded to an estimated 2.5% excess in lung cancer mortality (95% confidence interval 1.4 to 3.7%).
  • In Casale Monferrato, pleural neoplasm mortality relative risk was 13.0. The model estimated a 38.3% excess in lung cancer mortality there; the actual lung cancer SMR was 1.39, with 372 cases.
  • Urban municipalities had higher lung cancer mortality than rural ones (relative risk 1.21; 95% confidence interval 1.16 to 1.27).
  • Assuming zero background pleural neoplasm risk, 3.9% of lung cancer deaths (95% confidence interval 2.1 to 5.7%), or 1,016.7 cases (84.7 per year), were attributable to residence in municipalities with high pleural neoplasm mortality. The figure was 4.0% in urban areas and 3.2% in rural areas. It was 3.7% assuming a baseline pleural neoplasm rate of 2 cases per million per year.
  • Excluding Casale Monferrato, the regression coefficient changed to 0.037 (p = 0.02). Omitting Turin, it was 0.024 (p less than 0.01). An analysis using crude SMRs instead of Bayesian estimates gave a weaker association (coefficient 0.015; attributable risk 2.4%).
• The authors' conclusion: The authors suggested that the effects of asbestos exposure on lung cancer described in occupational epidemiology are also observable when studying the general population. They considered it likely that occupational and environmental asbestos exposure is one of the causes of the observed association.
• Limitations the authors state:
  • Population denominators from the 1981 census were applied throughout the 13-year period, so migration and changes in age structure reduce the reliability of the rates.
  • Error may come from misclassification of pleural neoplasm deaths and from residual error in small-area mortality estimates.
  • Asbestos exposure was assumed to be constant over each municipality, which may distort the observable effect.
  • Because latency periods differ, excess pleural neoplasm mortality may reflect asbestos exposure earlier than the period relevant for lung cancer.
  • Data on smoking and other important determinants of lung cancer were not available.
  • Without information on individual lung cancer cases, occupational and environmental components of the association could not easily be separated.
  • A few municipalities had strong leverage on the results; 46.9% of pleural neoplasm deaths were in Casale Monferrato, and Turin held a large share of the population.

Topics

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

Medical terms in this study, explained in our Lexicon: Lung Cancer, Neoplasm, Asbestos Exposure, confidence interval, Dose-response relationship.

Related studies

• Spatial variation of mortality for common and rare cancers in Piedmont, Italy, from 1980 to 2000: a Bayesian approach — M. Maule, F. Merletti, et al. (2006)
• Cohort mortality study of women compensated for asbestosis in Italy — D. Germani, S. Belli, et al. (1999)
• Pleural mesothelioma mortality and asbestos exposure mapping in Italy — L. Fazzo, M. De Santis, et al. (2012)
• Burden of Mortality from Asbestos-Related Diseases in Italy — L. Fazzo, A. Binazzi, et al. (2021)
• Cohort mortality study of rock salt workers in Italy — M. Tarchi, D. Orsi, et al. (1994)
• Epidemiology of pleural mesothelioma in North-western Italy (Piedmont) — G.F. Rubino, G. Scansetti, et al. (1972)

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