Mortality among sheet metal workers participating in a respiratory screening program

Our plain-English summary of Mortality among sheet metal workers participating in a respiratory screening program, by L. Welch, J. Dement, G. West (American Journal of Industrial Medicine, 2015).

The study

• Authors: L. Welch, J. Dement, G. West
• Journal: American Journal of Industrial Medicine
• Published: 2015;58(4):378-91
• Cited by: 14 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Background: The Sheet Metal Occupational Health Institute Trust (SMOHIT) established a screening program in 1985 to examine the health hazards of the sheet metal industry in the U.S. and Canada.

Methods: 17,345 individuals with over 20 years in the trade and who participated in the program were followed for causes of death between 1986 and 2010. Both SMRs and Cox proportional hazards models investigated predictors of death due to lung cancer, mesothelioma, and chronic obstructive pulmonary disease (COPD).

Results: Significant excess mortality was seen for mesothelioma and asbestosis. Controlling for smoking, a strong trend for increasing lung cancer risk with increasing chest x-ray profusion >0/0 was observed. With an profusion score <1/0, FEV1 /FVC <80% was associated with lung cancer risk. COPD risk increased with increasing profusion score.

Conclusions: This study demonstrates asbestos-related diseases among workers with largely indirect exposures and an increased lung cancer risk with low ILO scores.

© 2015 Wiley Periodicals, Inc.

In plain English

This study followed 17,345 sheet metal workers who had been union members for 20 or more years and were screened by chest x-ray at 62 sites nationwide between 1986 and 2004. The authors traced deaths through 2010, compared them with the U.S. population, and used models accounting for smoking. There were 6,636 deaths, and overall death rates were lower than the national population's. Lung cancer deaths (808) were not significantly raised overall. Mesothelioma deaths (85; SMR 7.34, 95% CI 5.86 to 9.08) and asbestosis deaths (48) were significantly raised. Lung cancer risk rose as lung-tissue shadowing on x-ray (the ILO profusion score) increased, and this was also seen in workers who never smoked. Among workers scoring below 1/0, an FEV1/FVC breathing-test ratio under 80% was linked to higher lung cancer risk. Pleural changes on x-ray were linked to mesothelioma deaths but not, after adjustment, to lung cancer. The authors concluded that these workers, whose exposures were largely indirect, showed asbestos-related disease. They noted a healthy survivor effect and that workers with under 20 years in the trade were not covered.

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: Cohort mortality study (an extended follow-up of an earlier analysis of the same screening cohort). Deaths were compared with U.S. national rates (standardized mortality ratios, SMRs), and Cox proportional hazards models were stratified by smoking status.
• Who was studied: 17,345 sheet metal workers who had been union members for 20 years or more as of January 1, 1986. All were screened by chest x-ray at 62 sites nationwide between 1986 and 2004 and followed for deaths through December 31, 2010. The cohort was 99.8% male and 99.2% White, with a mean age of 57.4 at intake. The Cox models were limited to White males (16,068 in the lung cancer model).
• Exposure: Asbestos exposure in sheet metal work. Fiber types are not named and no exposure measurements were made for the cohort. The article describes handling asbestos-containing gaskets and working near insulation workers applying asbestos for fireproofing and insulation. It also describes working on or around asbestos-fireproofed beams and renovating asbestos-insulated ventilation systems. Exposure was studied indirectly through years in the trade, time since entry into the trade, and x-ray and lung-function findings.
• Main results:
  • Among the 17,345 workers, 6,636 died by December 31, 2010. The all-causes SMR was 0.83 (95% CI 0.81 to 0.85), which the authors attributed to a healthy worker effect.
  • Lung cancer (trachea, bronchus, lung) in the whole cohort: 808 deaths against 784.86 expected, SMR 1.03 (95% CI 0.96 to 1.10), not significantly elevated.
  • Whole cohort, by cause: mesothelioma, 85 deaths against 11.58 expected, SMR 7.34 (95% CI 5.86 to 9.08); malignant neoplasm of the pleura, 11 deaths against 1.48 expected, SMR 7.42 (95% CI 3.70 to 13.27). Both were significantly elevated.
  • Asbestosis deaths in the whole cohort were significantly elevated: 48 deaths against 4.11 expected. Table II gives an SMR of 11.68 (95% CI 8.61 to 15.48); the text gives 11.74.
  • Lung cancer SMRs rose with x-ray profusion category: 0.94 (0.87 to 1.01) for categories 0/- to 0/1, 1.65 (1.39 to 1.94) for 1/0 to 1/2, and 3.27 (1.79 to 5.48) for 2/1 to 2/3.
  • In the Cox model controlling for smoking (16,068 White males), the lung cancer hazard ratio compared with profusion 0/- to 0/0 was 1.08 (0.85 to 1.37) for 0/1, 1.45 (1.19 to 1.76) for 1/0 to 1/2, and 2.72 (1.52 to 4.86) for 2/1 to 2/3. The trend was significant: the text gives P less than 0.001 and the table footnote gives P less than 0.0001.
  • In a model restricted to workers who reported never smoking (4,323 workers, 50 lung cancers), those with profusion 1/0 or higher had a lung cancer hazard ratio of 2.19 (95% CI 1.01 to 4.72). Pleural changes alone did not increase lung cancer risk in this group.
  • After adjusting for x-ray profusion, smoking and other factors, pleural abnormalities were not significantly linked to lung cancer death (HR 0.90, 95% CI 0.74 to 1.11). In a model not adjusting for age or years in the trade, the HR was 1.24 (95% CI 1.00 to 1.50).
  • Among workers with an ILO profusion score below 1/0, an FEV1/FVC ratio below 80% was associated with increased lung cancer risk, and the risk rose as the ratio fell. Low FEV1 or FVC (below 80% of predicted) did not reach statistical significance (P = 0.07 for FEV1 and 0.31 for FVC). The table with these figures is not in the supplied text.
  • In the Cox model for mesothelioma (15,355 White males, 82 cases), the hazard ratio for pleural abnormalities was 1.95 (95% CI 1.16 to 3.27), P = 0.0116. For profusion 1/0 or higher it was 1.67 (95% CI 0.93 to 3.02), of borderline significance (P = 0.0879). No spirometry measure predicted mesothelioma death.
  • COPD in the whole cohort: 461 deaths, SMR 0.95 (0.86 to 1.04). COPD SMRs were raised among workers with pleural changes (1.28, 95% CI 1.10 to 1.48), with profusion 1/0 to 1/2 (1.56, 1.26 to 1.92) and with profusion 2/1 to 2/3 (3.63, 1.87 to 6.33).
  • In the Cox model for COPD without spirometry terms (10,256 White males, 246 COPD deaths), profusion 1/0 or higher had a hazard ratio of 1.71 (95% CI 1.26 to 2.30). With spirometry included, the hazard ratio was 1.33 (0.98 to 1.80), of borderline significance (P = 0.07).
• The authors' conclusion: The authors concluded that the study shows asbestos-related diseases among workers with largely indirect asbestos exposures. They also concluded that it shows increased lung cancer risk even with low ILO x-ray scores. In the discussion they add that the findings suggest a relationship between asbestos exposure and death from COPD.
• Limitations the authors state:
  • X-ray readings and smoking status were recorded at the screening exam, which could have been more than a decade before death. Some workers classed as without asbestosis may have developed it later. The authors judged that this misclassification is unlikely to explain the results.
  • The study had a strong healthy survivor effect. Because it included only workers with 20 or more years in the trade, it could not address risks for those who worked fewer years, and the contrasts in exposure were reduced.
  • Other unmeasured risk factors, such as welding fumes or other occupational lung carcinogens, could have contributed. The authors judged that these are unlikely to explain the patterns seen by profusion score.
  • Requiring 20 or more years in the trade left no reference group with low cumulative asbestos exposure. The authors say this may partly explain why no pleural effect on lung cancer was found in the adjusted model.

Topics

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

Medical terms in this study, explained in our Lexicon: Lung Cancer, Mesothelioma, COPD, Chest X-Ray, FEV1.

Related studies

• Asbestos-related disease among sheet-metal workers. Preliminary results of the National Sheet Metal Worker Asbestos Disease Screening Program — L.S. Welch, D. Michaels, et al. (1991)
• The National Sheet Metal Worker Asbestos Disease Screening Program: radiologic findings. National Sheet Metal Examination Group — L.S. Welch, D. Michaels, et al. (1994)
• Asbestos disease in sheet metal workers: proportional mortality update — D. Michaels, S. Zoloth (1988)
• Asbestos-related disease among sheet metal workers 1986-2004: radiographic changes over time — L.S. Welch, E. Haile (2009)
• Change in prevalence of asbestos-related disease among sheet metal workers 1986 to 2004 — L.S. Welch, E. Haile, et al. (2007)
• Respiratory health of workers exposed to metal dusts and foundry fumes in a copper refinery — G. Ostiguy, C. Vaillancourt, et al. (1995)

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