Summary Statement

The Sheet Metal Occupational Health Institute Trust (SMOHIT) was formed in 1985 to examine the health hazards of the sheet metal industry in the U.S. and Canada through an asbestos disease screening program. This investigation updates findings concerning mortality patterns among screening program participants. A cohort of 17,345 individuals with 20 or more years in the trade and who participated in the asbestos disease screening program were followed for vital status and causes of death between 1986 and 2010. Data from the screening program included chest x-ray results, spirometry, and smoking history. Standardized Mortality Ratios (SMR) by cause were generated using U.S. death rates and Cox proportional hazards models were used to investigate predictors of death due to lung cancer, mesothelioma, and COPD. A significantly reduced SMR of 0.83 (95% CI=0.81-0.85) was observed for all causes combined. Statistically significant excess mortality was observed for pleural cancers,mesothelioma, and asbestosis in the SMR analyses. In Cox models, which controlled for smoking, increased lung cancer risk was observed among workers with ILO scores of 0/1 (RR=1.10, 95% CI=0.86-1.39), with a strong trend for increasing lung cancer risk with increasing ILO profusion score greater than 0/0. Among workers with an ILO profusion score < 1/0, an FEV1 /FVC ratio less than 80% was associated with an increased risk for lung cancer. COPD mortality was predicted by increased interstitial markings on the chest x-ray, FVC below 70% of predicted, FEV1 below 70% of predicted, and an FEV1 /FVC ratio below 70%. The risk of death from cancers of the pleura and mesothelioma also were significantly increased among workers without radiological evidence of asbestosis or pleural abnormalities. Conclusions Sheet metal workers are at increased risk for asbestos-related diseases. This study contributes to the literature demonstrating asbestos-related diseases among workers with largely indirect exposures and supports an increased lung cancer risk among workers with low ILO profusion scores.
April 2014

© 2014, CPWR – The Center for Construction Research and Training. CPWR, the research and training arm of the Building and Construction Trades Dept., AFL-CIO, is uniquely situated to serve construction workers, contractors, practitioners, and the scientific community. This report was prepared by the authors noted. Funding for this research study was made possible by a cooperative agreement with the National Institute for Occupational Safety and Health, NIOSH (OH009762). The contents are solely the responsibility of the authors and do not necessarily represent the official views of NIOSH or CPWR.

RUNNING TITLE: Mortality among Sheet Metal Workers

GRANT SPONSOR: National Institute for Occupational Safety and Health

GRANT NUMBER: Cooperative Agreement Number (U60 OH009762, CFDA # 93.262)

CPWR SMALL STUDY NUMBER: 13-1-PS

Correspondence address: John M. Dement, Ph.D., CIH, Professor
Division of Occupational & Environmental Medicine, Department of Community & Family Medicine
Duke University Medical Center
2200 West Main Street, Suite 400
Durham, NC 27710
tel (919)684-8136, fax (919)286-1021, John.Dement@Duke.edu

Draft: April 23, 2014

ABSTRACT

Background

The Sheet Metal Occupational Health Institute Trust (SMOHIT) was formed in 1985 to examine the health hazards of the sheet metal industry in the U.S. and Canada through an asbestos disease screening program. This investigation updates findings concerning mortality patterns among screening program participants was undertaken and further investigates predictors of increased mortality due to lung cancer, mesothelioma, and COPD.

Methods

A cohort of 17,345 individuals with 20 or more years in the trade and who participated in the asbestos disease screening program were followed for vital status and causes of death between 1986 and 2010. Data from the screening program included chest x-ray results by ILO criteria, spirometry, and smoking history. Standardized Mortality Ratios (SMR) by cause were generated using U.S. death rates and Cox proportional hazards models were used to investigate predictors of death due to lung cancer, mesothelioma, and COPD

Results

A significantly reduced SMR of 0.83 (95% CI=0.81-0.85) was observed for all causes combined. Statistically significant excess mortality was observed for pleural cancers, mesothelioma, and asbestosis in the SMR analyses. In Cox models, which controlled for smoking, increased lung cancer risk was observed among workers with ILO scores of 0/1 (RR=1.10, 95% CI=0.86-1.39), with a strong trend for increasing lung cancer risk with increasing ILO profusion score greater than 0/0. Among workers with an ILO profusion score < 1/0, an FEV1 /FVC ratio less than 80% was associated with an increased risk for lung cancer. COPD mortality was predicted by increased interstitial markings on the chest x-ray, FVC below 70% of predicted, FEV1 below 70% of predicted, and an FEV1 /FVC ratio below 70%. The risk of death from cancers of the pleura and mesothelioma also were significantly increased among workers without radiological evidence of asbestosis or pleural abnormalities.

Conclusions

Sheet metal workers are at increased risk for asbestos-related diseases. This study contributes to the literature demonstrating asbestos-related diseases among workers with largely indirect exposures and supports an increased lung cancer risk among workers with low ILO profusion scores.

KEY WORDS: sheet metal worker, construction, trades, mortality, cancer, lung cancer

KEY FINDINGS

  • Statistically significant excess mortality was observed for pleural cancers, mesothelioma, and asbestosis among workers participating in the Sheet Metal Occupational Health Institute Trust (SMOHIT) medical screening program, diseases associated with a history of asbestos exposure.
  • After controlling for smoking, increased lung cancer risk was observed among workers with ILO profusion scores of 0/1, with a strong trend for increasing lung cancer risk with increasing ILO profusion score greater than 0/0.
  • Among workers with an ILO profusion score < 1/0, an FEV1 /FVC ratio less than 80% was associated with an increased risk for lung cancer.
  • COPD risk was predicted by increased interstitial markings on the chest x-ray, FVC below 70% of predicted, FEV1 below 70% of predicted, and an FEV1 /FVC ratio below 70%.
  • The risk of cancers of the pleura and mesothelioma also were significantly increased among workers without radiological evidence of asbestosis or pleural abnormalities.

BACKGROUND

Numerous studies have documented the health effects of occupational exposure to asbestos [Becklake ,1976; Nicholson et al., 1982; Selikoff et al., 1978; IARC, 2009; American Thoracic Society, 2004]. Based on the results of studies undertaken in the 1980s [Zoloth and Michaels, 1985; Selikoff and Lilis, 1991], the Sheet Metal Workers International Association (SMWIA) and the Sheet Metal and Air Conditioning National Association formed The Sheet Metal Occupational Health Institute Trust (SMOHIT) to examine the health impact of asbestos exposure in the sheet metal industry. This investigation updates prior reports on findings concerning mortality patterns among screening program participants, and further investigates predictors of increased mortality due to lung cancer, mesothelioma, and COPD.

Sheet metal work involves fabrication or installation of metal products, such as ventilation systems, metal roofing, and metal facades, as well as large-scale production of metal products, such as refrigerators and air conditioners. Sheet metal workers are primarily employed in the construction industry but have also worked in the railroad industry and shipyards, as well as in specialized sheet metal production shops. The craft of sheet metal work handled asbestoscontaining materials through the use of gaskets, but at the same time for many years sheet metal workers in construction were exposed to asbestos while working in close proximity to insulation workers applying asbestos containing materials for fireproofing and insulation, by working on or around beams that had been previously fireproofed with asbestos, and by renovating asbestosinsulated metal ventilation systems [Williams et al., 2007]. Very high levels of airborne asbestos fibers were measured during spray application of asbestos before 1973 [Paik et al., 1973], when this application method was banned.

Prior analyses have examined the prevalence of asbestos-related diseases among this cohort of sheet metal workers [Welch et al., 1991; Welch et al., 1994; Welch et al., 2007]. Welch and colleagues [2007] reported that among 18,211 individuals 10.4% had radiographic findings consistent with asbestosis and 21.7% had pleural scarring; the prevalence of asbestos-related radiographic change increased with years worked and with any shipyard work. In a study of mortality in this same cohort [Dement et al., 2009] statistically significant excess mortality was observed for pleural cancers, mesothelioma, and asbestosis. Standardized mortality ratios for both lung cancer and COPD increased consistently and strongly with increasing ILO profusion score. Here we report an extended follow-up of this same group of workers.

MATERIALS AND METHODS

Cohort Definition

We have previously presented mortality results for this cohort with follow-up through 2004 [Dement et al., 2009]. The current analyses extended the mortality follow-up of this cohort through 2010 and readers are referred to the prior publication for details of the cohort and follow-up methods. Briefly, individuals who were members of the Sheet Metal Workers International Union for 20 years or more as of January 1, 1986 were invited to participate in a medical examination program provided by the Sheet Metal Occupational Health Institute Trust (SMOHIT). Previous reports on this population describe the methods in more detail and present the prevalence of asbestos-related disease on chest radiographs [Welch et al., 1991; Welch et al., 1994; Welch et al., 2007]. Components of the screening program included: 1) completion of an occupational and medical questionnaire, 2) a limited physical examination (blood pressure determination, examination of the heart and lungs, and examination for digit clubbing), 3) spirometry, performed according to American Thoracic Society (ATS) guidelines [American Thoracic Society, 1987], and 4) PA and lateral chest radiograph, interpreted using the International Labour Office (ILO) classification for pneumoconiosis [International Labour Office, 1980]. Each chest x-ray was classified by one reader who was an A-reader, a B-reader, or a physician with proficiency in the use of the ILO classification but who was neither an A nor a B reader [Welch et al., 2007]. For the current analyses percent predicted FVC and FEV1/FVC ratio were calculated based on the prediction equations of Hankinson et al. [1999].)

The current study cohort included 17,345 individuals who were screened for asbestos related disease by chest x-ray at 62 sites nationwide, between 1986 and 2004. We excluded 1582 workers without sufficient demographic data vital status determination.

Two sources of information were used to obtain vital status and cause of death data: 1) records in the Sheet Metal National Pension Fund (SMNPF), and 2) the National Death Index (NDI). Records of the SMNPF were searched and members covered by this plan and those receiving pension benefits or still actively contributing toward their pension as of December 31, 2010 were considered alive. Workers with unknown vital status by match with the SMNPF and workers identified as deceased were followed to identify additional deaths and causes of deaths through December 31, 2010 using the NDI Plus system [Bilgrad, 1995], maintained by the National Center for Health Statistics (NCHS). Record linkage with the NDI was accomplished using probabilistic scores assigned by the NDI and recommended cut-off scores by class for records without a perfect match [Horm, 1996; NCHS, 2013]. We assumed that workers not identified as deceased by the SMNPF or the NDI were still alive as of December 31, 2010 as the NDI provides virtually complete ascertainment of deaths among men and among employed women [Boyle and Decouflė, 1990; Stampfer et al., 1984].

The study was conducted in accord with the recommendations of the Helsinki Declaration (World Medical Association 1975) and was approved by the IRB of CPWR. Each participant signed an informed consent to permit this use of data collected in the examinations.

Cohort Mortality Analyses

The Life Table Analysis System (LTAS.Net Version 3.0.3) developed by the National Institute for Occupational Safety and Health (NIOSH) [Steenland et al., 1990; Robinson et al., 2006] was utilized to compute cause-specific Standardized Mortality Ratios (SMRs), comparing the mortality experience of the cohort to that of the U.S. national population for 119 causes of death, were calculated adjusting for age, race, sex, and calendar year. The chest x-ray date was selected as the starting point for person-years accumulation for each cohort member and person-years accumulated until death or the study cut-off date of December 31, 2010. SMRs were calculated as the ratio of observed to expected deaths and 95% confidence intervals for SMRs were computed assuming that the observed number of deaths in the cohort is a Poisson random variable. The NIOSH LTAS uses the Byar approximation when the number of cases is six or more and the exact Poisson confidence interval when the number of cases was fewer than five [Rothman and Boise, 1979; NIOSH, 2011]

In addition to overall results for the entire cohort, we investigated mortality for selected causes by time since entry into the sheet metal trade, ILO parenchymal profusion category, and presence or absence of pleural changes. Parenchymal profusion categories were group into four categories as was done by Cullen et al. [2005]. A pleural abnormality was defined as bilateral pleural thickening or plaques, with or without calcification [Cullen et al., 2005]. .

Multivariate Modeling of Lung Cancer, Mesothelioma, and COPD Mortality Predictors

Within the overall cohort, further analyses were undertaken to examine the association between chest x-ray readings, spirometry, work history, and smoking and mortality due to lung cancer, mesothelioma, and COPD. Analyses of the relationship between chest film changes and lung cancer mortality were restricted to Caucasian males with 20 or more years of work in the sheet metal trade and having data on other covariates considered in the models. Females and other race groups were excluded from these analyses due to small numbers. In the main analyses, only workers having spirometry meeting ATS criteria for repeatability between maneuvers were included. Other investigators have suggested that exclusion of workers not meeting reproducibility criteria could introduce bias through exclusion of workers with accelerated loss of lung function [Eisen et al., 1984]. To investigate this issue, we conducted additional sensitivity analyses to determine effects of eliminating workers with spirometry not meeting ATS criteria.

Stratified Cox proportional hazards models were used to obtain lung cancer, mesothelioma, and COPD relative risks and 95 percent confidence intervals. The time axis in these models was the time from the initial screening examination to death due to diseases being studied or the last date that the worker was known alive. Stratified Cox models were fit with strata defined by smoking status at examination (never, past, and current) and covariates considered in the baseline models for each disease included pack-years of smoking (0, 1-19, 20-39, or ≥ 40 pack-years), age at start of follow-up (<50, 50-54, 55-59, 60-64, 65-69, 70-74, or ≥ 75 years), time since last work in the sheet metal trade (<5, 5-9, or ≥ 10 years), years of sheet metal work (20-24, 25-29, 30-34, or ≥ 35 years). Age, pack-years, and years of sheet metal work were modeled as grouped linear variables, constructed by assigning ordinal scores to categories based on category median values, and fitted as continuous variables in a manner similar to the study by Cullen et al. [2005]. Asbestos related chest radiographic predictors (presence of pleural abnormalities, and ILO parenchymal profusion category) and spirometry classifications for percent predicted FVC, percent predicted FEV1, and FEV1/FVC ratio (≥80, 70-79, 60-69, and <60) were investigated in the models as categorical variables and tests for trends across covariate category were performed by entering the covariate in the model as a grouped linear variable [Rothman and Greenland, 1998]. Pleural abnormalities were defined as bilateral pleural thickening or plaques, with or without calcification [Cullen et al., 2005].

After assessing the magnitude of the effect of each single variable on outcome, covariates that were significant in a univariate models (likelihood ratio p-values <0.1) were considered candidate variables for inclusion in multivariate models for each disease. Covariates were retained in the final models if statistically significant based on their Wald Type 3 chi-square or if their inclusion in the multivariate models changed the parameter estimates for pleural changes, parenchymal changes, or spirometry more than 10 percent. We restricted our analyses of spirometry predictors of lung cancer mortality to a sub-cohort with spirometry and with an ILO profusion score <1/0. For analyses of mesothelioma risk, cases included mesotheliomas (ICD- 10) and pleural cancers (ICD-9) and parenchymal profusion was dichotomized into <1/0 and ≥ 1/0 due to small numbers in higher profusion categories. Analyses of COPD risks were restricted to the sub-cohort with spirometry and parenchymal profusion was dichotomized into <1/0 and ≥ 1/0 due to small numbers in higher profusion categories.

Cox proportional hazards models were fit using PROC PHREG in SAS Version 9.3 [SAS, 2011]. The EXACT method of handing ties in PROC PHREG was used and the assumption of proportional hazards over the follow-up period was assessed with time-dependent covariates (the product of log-transformed time and the factor of interest). The ASSESS option for testing the proportional hazard assumption available in SAS Version 9.3 also was used for this purpose.

RESULTS

There were 6,636 deaths as of December 31, 2010 among the 17,345 workers in the cohort (Table I). The cohort was almost entirely male and Caucasian with a mean age of 57.4 years at intake exam. Twenty-six percent of the cohort had never smoke cigarettes, and 25.4% were still smoking at the time of their intake exam. Radiographic parenchymal changes (profusion > 1/0) were observed in 10.4% of workers and 21.7% had radiographic pleural changes. There were 808 deaths from lung cancer, 85 deaths from mesothelioma with an additional 11 deaths coded to malignant neoplasm of the pleura, and 461 deaths from COPD.

Standardized mortality ratio analyses showed a significant deficit for all causes of death (Table II). The SMR for malignant neoplasm of the trachea, bronchus, and lung was 1.03 which was not significantly elevated. The SMR for mesothelioma and for malignant neoplasms of the pleura were both significantly elevated. The SMR was significantly decreased for a number of causes of death, including heart disease and diseases of the respiratory system, with the exception of a significantly elevated SMR of 11.74 for asbestosis.

Table III displays mortality for selected causes of death by time since entry into the sheet metal trade, a variable that reflects both duration of exposure and latency. The SMR for lung cancer did not increase with time since entry into the trade. The vast majority of deaths from asbestosis, mesothelioma and malignant neoplasm of the pleura occurred after 40 years from first exposure. Table IV displays mortality by increasing amount of parenchymal disease on chest x-ray. The SMRs for lung cancer and asbestosis increased significantly with increasing profusion of parenchymal abnormalities. Table V shows that deaths from asbestosis were significantly higher among workers with pleural changes, but that for the other conditions pleural changes were not associated with a higher SMR.

Tables VI-IX present results of the Cox models for lung cancer, mesothelioma, and COPD. The relative risk for lung cancer death increased significantly as the profusion category on chest x-ray increased, and also increased with, age, smoking, years in sheet metal trade, and years since last sheet metal trade work at exam. A test for linear trend for lung cancer risk by profusion category was highly significant (P<0.001). There is no significant relationship between lung cancer deaths and the presence or absence of pleural abnormalities after parenchymal changes, smoking and other covariates were entered into the model. Among workers with an ILO profusion score < 1/0, an FEV1 /FVC ratio less than 80% was associated with an increased risk for lung cancer and the relative risk increased as the ratio decreased; there was no such relationship with FVC or FEV1.

Parenchymal profusion changes ≥1/0 and pleural changes were significantly associated with increased risk of mesothelioma mortality (Table VIII). Table IX death from COPD was predicted by increased interstitial markings on the chest x-ray, FVC below 70% of predicted, FEV1 below 70% predicted, and an FEV1 /FVC ratio below 70%. Age, smoking, years in the sheet metal trade, and years since last sheet metal trade work at exam were also predictive of COPD mortality.

DISCUSSION

Sheet metal workers who participated in this nationwide screening program had a reduced SMR overall compared to the US population, consistent with a healthy survivor effect. No overall increase in lung cancer mortality was observed among this cohort when compared to the US population; however, SMR analyses revealed excess mortality for mesothelioma, malignant neoplasm of the pleura, and asbestosis. Additionally, the SMR analyses demonstrated significant excess risk for lung cancer and COPD among workers with parenchymal changes ≥ 1/0 in profusion. The SMR was significantly elevated for pleural cancers, mesothelioma, and asbestosis among workers who did not have parenchymal changes. Cox proportional hazards models controlling for smoking confirmed the excess risk of lung cancer among workers with a profusion score ≥ 1/0 and provided compelling evidence for excess lung cancer risk among workers with parenchymal profusion scores <1/0 on the ILO scale.

In addition to smoking and abnormal pulmonary function, which are known risk factors for COPD mortality, both increased interstitial markings on chest x-ray and years in the sheet metal trade were also predictive of death from COPD. These findings suggest a relationship between asbestos exposure and death from COPD. Previous research has shown that asbestos exposure is associated with obstructive disease on lung function testing [ATS, 2004, Dement, et al., 2010]. Exposure to dust, fumes, gases, and vapors is now recognized as a cause of COPD, and of mortality from COPD among construction workers [Bergdahl, 2006]; asbestos is an important component of the dust exposure.

It is well accepted that workers with exposure to asbestos sufficient to cause radiographic changes consistent with clinical asbestosis are at extremely high risk of lung cancer [Coutts et al., 1987; Berry, 1981; Roggli, 1990; Cookson et al., 1985; Liddell and McDonald, 1980; Huuskonen, 1978; Finkelstein et al., 1981; Reid et al., 2005, Reid 2006; Oksa et al., 1997; Karjalainen et al., 1999; Markowitz et al., 2013]. Consistent with the current study, increased lung cancer risk also has been observed among workers without radiological evidence of asbestosis [Markowitz, et al., 2013; Reid, 2006; Wilkinson et al., 1995; Finkelstein, 2010; Anttila et al. 1993; Cullen et al., 2005]. Cullen et al. [2005] conducted a follow-up study of 4060 men with heavy asbestos exposure who participated in a β-carotene and retinol efficiency trial. Parenchymal changes on radiograph were associated with progressively increasing lung cancer risk and workers with a parenchymal profusion of 0/1 were found to have a lung cancer relative risk of 1.48 (0.99-2.22) compared to workers without parenchymal changes. Additionally, the risk of lung cancer increased steadily by duration of heavy asbestos exposure among workers without chest x-ray evidence of asbestosis. Reid reported similar findings in a study of former workers and residents of the Wittenoom mine, finding that both radiographic asbestosis and asbestos exposure were significantly associated with an increased risk of lung cancer, and there was an increased risk of lung cancer reported specifically in those without asbestosis. The most recent study to report on this issue comes from an updated analysis of the North American insulator cohort [Markowitz et al., 2013]. The authors reported that asbestos exposure without radiographic evidence of asbestosis in non-smokers raise the risk of lung cancer by 3.6 fold, and that asbestosis further doubled the lung cancer mortality risk.

Although prior research had found a relationship between the presence of pleural plaque and lung cancer mortality [Loomis et al., 1989; Hillerdal, 1994; Karjalainen et al., 1999; Cullen et al., 2005; Ameile et al, 2011] our study did not find an excess lung cancer risk among workers with pleural changes after adjustment for other model covariates including duration of sheet metal trade work, smoking, presence of parenchymal changes, and time since last sheet metal work at exam. Our failure to find an independent effect of pleural changes on the risk of lung cancer is most likely due to the nature of our cohort, consisting of older workers with 20 or more years of sheet metal work. The prevalence of pleural changes increased markedly with age in our cohort with a low of 9.5% among workers less than 55 years of age to 43.8% among workers older than 70 years. A logistic model (not shown) found both age and duration of sheet metal work to be strong predictors of pleural changes (p<0.001) thus control for both age and years of sheet metal work in our Cox lung cancer models diminished the effects of pleural changes. The presence of pleural plaques may serve as a marker for a higher cumulative asbestos exposure, thus our inability to detect a pleural effect may be due to the requirement that workers have 20 or more years in the sheet metal trade for entry into our cohort, thereby diminishing availability of a reference group with low cumulative asbestos exposure.

One limitation of this study is that the radiologic interpretation and the classification of smoking status were determined at the time of the clinical exam, which could have preceded the date of death by more than a decade. It is possible that some of the workers who were categorized as without asbestosis could have developed parenchymal disease in the interim period. Prior analysis within this same cohort found that only 5% had progression from a normal chest x-ray to one classified as 1/0 or higher over a mean of 9 years [Welch et al., 2007], so misclassification of asbestosis is unlikely to explain the results.

The current study provides additional evidence that workers who experienced largely intermittent and indirect exposure to asbestos are at increased risk of asbestos-related diseases and at risk for COPD. The risk of lung cancer risks increased sharply with parenchymal profusion score; however, the study provided compelling evidence for increased risk among workers without radiographic asbestosis. The risk of cancers of the pleura and mesothelioma also were significantly increased among workers without radiological evidence of asbestosis or pleural abnormalities. Our study has several strengths including a large population with chest xray data classified by ILO criteria, and smoking histories on each member of the cohort. Nonetheless, our study is limited by a strong healthy survivor effect and an inability to address risks for workers who worked less than 20 years. While our analyses controlled for smoking, it is impossible to entirely exclude a contribution by other unmeasured risk factors such as welding fumes or other occupational lung carcinogens. However, confounding by these unmeasured exposures is unlikely to explain the steep and consistent patterns observed by profusion score.

ACKNOWLEDGEMENTS

The National Institute for Occupational Safety and Health supported this study through a cooperative agreement with the Center for Construction Research and Training and the National Institute for Occupational Safety and Health (Cooperative Agreement Number OH009762). We express our appreciation to the Sheet Metal Workers International Association and Sheet Metal Occupational Health Institute Trust who provided the data for this project and assisted with cohort follow-up. Opinions expressed are those of the authors and do not necessarily represent the official views of CPWR or NIOSH.

REFERENCES

Allison PD 1995. Survival Analysis Using the SAS System – A Practical Guide. , Cary , NC: SAS Institute Inc. 1995

American Thoracic Society. Standardization of Spirometry -1987 update. 1987. Am Rev Respir Dis .136:1285-1298.

American Thoracic Society. (2004. Diagnosis and initial management of nonmalignant diseases related to asbestos.Am J Respir Crit Care Med 170(6): 691-715.

Ameille, J., P. Brochard, Letourneux M, Paris C, Pairon,JC . 2011. Asbestos-related cancer risk in patients with asbestosis or pleural plaques. Rev Mal Respir 28(6): e11-e17.

Anttila S, Karjalainen A, Taikina-aho O, Kyyronen P, Vainio H. 1993. Lung Cancer in the Lower Lobe is Associated with Pulmonary Asbestos Fiber Count and fiber Size. Environmental Health Perspectives. 110(2):166-170.

Becklake MR. 1976. Asbestos-related diseases of the lung and other organs: their epidemiology and implications for clinical practice. Am Rev Respir Dis 114(1):187-227.

Bergdahl IA, Torėn K, Eriksson K, Hedlund U, Nilssonz T, Flodin R, Järvholm B (2004) Increased mortality in COPD among construction workers exposed to inorganic dust. Eur Respir J 23:402-406.

Baker EL, Dagg T, Greene RE.1985. Respiratory illness in the construction trades: 1. The significance of asbestos-associated pleural disease among sheet metal workers. J Occup Med 27:483-489.

Berry G. 1981. Mortality of workers certified by pemumoconiosis medical panels as having asbestosis. British Journal of Industrial Medicine. 38:130-137

Bilgrad R. 1995. National Death Index Plus: Coded Causes of Death Supplement to the National Death Index User’s Manual, Centers for Disease Control and Prevention.

Boyle CA, Decouflė P. 1990. National sources of vital status information: Extent of coverage and possible selectivity in reporting. Am J Epidemiol. 131:160- 168.

Cookson WO, Musk AW, Glancy JJ; de Klerk NH, Yin R, Mele R, Carr NG, Armstrong BK, Hobbs MS. 1985. Compensation, radiographic changes, and survival in applicants for asbestosis compensation. Br J Ind Med. 42(7):461-8.

Coutts II JC, Turner-Warwich M. 1987. Mortality in cases of asbestosis diagnosed by pneumoconiosis medical panel. Thorax. 42:111-116.

Cullen M, Barnett MJ, Balmes JR, Cartmel B, Redlich CA, Brodkin C et al. 2005. Predictors of Lung Cancer among Asbestos-exposed Men in the Beta-Carotene and Retinol Efficacy Trial. Am J Epidemiol.161(3):260-270.

Dement J, Welch L, Haile E, Myers D. 2009. Mortality among sheet metal workers participating in a medical screening program. Am J Ind Med. 52(8):603-13.

Eisen EA, Robins AM, Greaves IA, Wegman DH. 1984. Selection effects of repeatability criteria applied to lung spirometry. Am J Epid. 120:734-742.

Finkelstein MM. 2010. Absence of radiographic asbestosis and the risk of lung cancer among asbestos-cement workers: Extended follow-up of a cohort. Am J Ind Med 53(11):1065-1069.

Finkelstein M, Kusiak R, Suranyl G. 1981. Mortality among workers receiving compensation for asbestosis in Ontario. CMA Journal.125:259-262.

Hankinson JL, Odencrantz JR, Fedan KB. 1999 Spirometric reference values from a sample of the General U.S. population. Am J Respir Crit Care Med. 159:179–187.

Hillerdal G. 1994. Pleural plaques and risk for bronchial carcinoma and mesothelioma: a prospective study. Chest. 105:144–50.

Horm J. 1996. Assignment of Probabilistic Scores to National Death Index Record Matches NDI User’s Manual Addendum, Centers for Disease Control and Prevention.

Huuskonen MS. 1978. Clinical Features, Mortality and Survival of Patients with Asbestosis. Scand. Journ. Work Enviro. 4:265-274.

IARC 2009. Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 100 C, Arsenic, Metals, Fibres, and Dusts volume 100 C, A Review of Human Carcinogens, International Agency for Research on Cancer, 150 cours Albert Thomas, 69372 Lyon Cedex 08, France

ILO U/C. 1980. International Classification of Radiographs of Pneumoconiosis. Geneva: ILO.

Karjalainen A, Pukkala E, Kauppinen T, Partanen T. 1999. Incidence of cancer among Finnish patients with asbestos-related pulmonary or pleural fibrosis. Cancer Causes Control 10(1): 51-57.

Kipen HM, Lilis R, Suzuki Y, Valciukas JA, Selikoff IJ. 1987. Pulmonary fibrosis in asbestos insulation workers with lung cancer: a radiological and histopathological evaluation. Br J Ind Med. 44(2):96-100.

Liddell FD, McDonald JC. 1980. Radiological findings as predictors of mortality in Quebec asbestos workers. Br J Ind Med. 37(3):257-67.

Loomis DP, Collman GW, Rogan WJ. 1989. Relationship of mortality, occupation, and pulmonary diffusing capacity to pleural thickening in the First National Health and Nutrition Examination Survey. Am J Ind Med. 16(5):477-84.

Markowitz, SB, Levin, SM, Miller, A Morabia, A. 2013. Asbestos, asbestosis, smoking, and lung cancer. New findings from the North American insulator cohort. Am.J.Respir.Crit Care Med. 188(1): 90-96.

Nicholson WJ, Perkel G, Selikoff IJ. 1982. Occupational exposure to asbestos: population at risk and projected mortality--1980-2030. Am J Ind Med. 3(3):259-311.

NCHS. 2013. National Death Index User’s Guide. National Center for Health Statistics Hyattsville, MD. 2013.

NIOSH. 2011. LTAS Manual, LTAS.Net Version 3.0.3. Downloaded November 8, 2011. http://www.cdc.gov/niosh/LTAS/news.html.

Oksa, P, Pukkala, E, Karjalainen, A, Ojajarvi, A Huuskonen, MS. 1997. Cancer incidence and mortality among Finnish asbestos sprayers and in asbestosis and silicosis patients. Am J Ind Med 31(6): 693-698.

Paik NW, Walcott RJ, Brogan PA. 1983. Worker exposure to asbestos during removal of sprayed material and renovation activity in buildings containing sprayed material. Am Ind Hyg Assoc J 44:428-432.

Reid A, de Klerk NH, Ambrosini GL, Olsen N, Pang SC, Berry G, Musk AW. 2005. The effect of asbestosis on lung cancer risk beyond the dose related effect of asbestos alone. Occup Environ Med. 62:885–889.

Reid A, de Klerk NH, Ambrosini GL, Berry G, Musk, AW. 2006. The risk of lung cancer with increasing time since ceasing exposure to asbestos and quitting smoking. Occup Environ Med 63(8):509-512.

Robinson CF, Schnorr TM, Cassinelli RT, Calvert GM, Steenland NK, Gersic CM, Schubauer- Berigan MK. 2006. Tenth Revision U.S. Mortality Rates for Use With the NIOSH Life Table Analysis System . J Occup Environ Med 48:662–667.

Roggli VL 1990. Human Disease Consequences of Fiber Exposures: A Review of Human Lung Pathology and Fiber Burden Data. Enviromental Health Perspectives. 88:295-303.

Rothman KJ, Boice JD. 1979. Epidemiologic Analysis With A Programmable Calculator, Washington, DC, US Government Printing Office.

Rothman KJ, Greenland S. 1998 Modern epidemiology. 2nd ed. Philadelphia, PA: Lippincott- Raven.

SAS Institute Inc. 2011. SAS/STAT® 9.3 User’s Guide. Cary, NC: SAS Institute Inc.

Selikoff IJ, Hammond EC. 1978. Asbestos-associated disease in United States shipyards. CA Cancer J Clin 28(2):87-99.

Selikoff IJ, Lilis R. 1991. Radiological abnormalities among sheet-metal workers in the construction industry in the United States and Canada: relationship to asbestos exposure. Arch Environ Health 46(1):30-36.

Stampfer MJ, Willett WC, Speizer FE, Dysert DC, Lipnick R, Rosner R, Hennekens CM. 1984. Test of the National Death Index. Am J Epidemiol 119:837-839.

Steenland K, Beaumont J, Spaeth S, Brown D, Okun A, Jurcenko L, et al. 1990. New developments in the Life Table Analysis System of the National Institute for Occupational Safety and Health. J Occup Med 32:1091-1098

Wilkinson P, Hansell DM, Jassens J, Rubens M, Rudd RM, Taylot AN, McDonald JC. 1995. Is lung cancer associated with asbestos exposure when there are no small opacities on the chest radiograph. Lancet. 345:1074-1078.

Welch LS, Michaels D, Zoloth SR. 1991. Asbestos-related disease among sheet-metal workers: preliminary results of the National Sheet Metal Worker Asbestos Disease Screening Program. Ann NY Acad Sci. 643:287-295.

Welch LS, Michaels D, Zoloth SR. 1994. The National Sheet Metal Worker Asbestos Disease Screening Program: Radiologic findings. National Sheet Metal Examination Group. Am J Ind Med. 25(5):635-648.

Welch LS, Haile E, Dement J, Michaels D. 2007. Change in Prevalence of Asbestos-Related Disease among Sheet Metal Workers 1988-2004. Chest. 131:863-869

Williams PRD, Phelka AD, Paustenbach DJ. 2007. A Review of Historical Exposures to Asbestos among Skilled Craftsmen (1940-2006). J Tox Env Hlth, Part B, 10:319-377.

Zoloth S, Michaels D. 1985. Asbestos disease in sheet metal workers: the results of proportional mortality analysis. Am J Ind Med. 7:315-321.

TABLES

Table I: Sheet Metal Cohort Demographics and Vital Status
Characteristic Total
Cohort
Lung
Cancer
Deaths
Number of Workers  17,345 808 
Total Deaths, December 31, 2010  6,636 --- 
Percent Male  99.8%  100% 
Percent Caucasian  99.2%  99.4 
Age at Intake Exam (Mean, (SD))  57.4 (8.7)  61.2 (7.7) 
Smoking Status at Intake Exam (No. (%))1    
___ Never Smoked  26.0%  6.6% 
___ Past Smoker  48.6%  42.7% 
___ Current Smoker  25.4%  50.7% 
Smoking Pack-Years for Ever Smoked (Mean, (SD))  32.3 (22.4)  44.3 (24.4) 
Years of Sheet Metal Trade Work (Mean, (SD))  31.9 (7.2)  33.5 (7.7) 
Prevalence of Radiographic Parenchymal Changes2 10.4%  19.7% 
Prevalence of Radiographic Pleural Changes3 21.7%  26.1% 
1 24 workers missing smoking data
2 ILO profusion of small irregular shadows in the lung parenchyma >= category 1/ 0
3 Any notations of positive findings in sections 3A-D of the NIOSH ILO coding form

 

Table II: Sheet Metal Worker Overall Mortality
Cause of Death  Obs.  Exp.  SMR  95% CI
Lower
95% CI
Upper
All Causes   6636 7972.54 0.83** 0.81 0.85
All Cancers  2224 2221.73 0.98 0.94 1.02
__MN buccal & pharynx  31 38.33 0.81 0.55 1.15
____MN lip 1 0.45 2.23 0.06 12.45
____MN tongue 6 8.92 0.67 0.25 1.46
____MN other buccal  12 9.77 1.23 0.63 2.14
____MN pharynx  12 19.19 0.63 0.32 1.09
__MN digestive & peritoneum  475 540.23 0.88** 0.80 0.96
____MN esophagus  60 73.81 0.81 0.62 1.05
____MN stomach  51 51.82 0.98 0.73 1.29
____MN intestine  139 184.96 0.75** 0.63 0.89
____MN rectum 28 37.04 0.76 0.50 1.09
____MN biliary, liver, gall bladder  58 68.52 0.85 0.64 1.09
____MN pancreas  131 117.79 1.11 0.93 1.32
____MN peritoneum, other & unspecified sites  8 6.29 1.27 0.55 2.51
__MN respiratory  839 812.78 1.03 0.96 1.10
____MN larynx 15 23.21 0.65 0.36 1.07
____MN trachea, bronchus, lung  808 784.86 1.03 0.96 1.10
____MN pleura  11 1.48 7.42** 3.70 13.27
____MN other respiratory  5 3.23 1.55 0.50 3.62
__MN breast   5 3.07 1.63 0.53 3.80
__MN female genital organs  0 0.20 0.00 0.00 18.28
__MN male genital organs  179 225.30 0.79** 0.68 0.92
__MN urinary  110 132.48 0.83 0.68 1.00
____MN kidney  49 60.02 0.82 0.60 1.08
____MN bladder & other urinary site  61 72.46 0.84 0.64 1.08
__MN other & unspecified sites   350 291.14 1.20** 1.08 1.33
____MN bone  2 3.38 0.59 0.07 2.14
____MN melanoma  30 37.20 0.81 0.54 1.15
____MN other skin  14 13.22 1.06 0.58 1.78
____MN mesothelioma  85 11.58 7.34** 5.86 9.08
____MN connective tissues  7 11.57 0.60 0.24 1.25
____MN brain & other nervous  48 48.47 0.99 0.73 1.31
____MN eye  1 1.02 0.98 0.02 5.45
____MN thyroid  3 4.34 0.69 0.14 2.02
____MN other & unspecified sites  160 160.35 1.00 0.85 1.17
__MN lymphatic & hematopoietic   235 228.19 1.03 0.90 1.17
____Hodgkin's disease  5 4.21 1.19 0.39 2.77
____Non-Hodgkin's lymphoma  97 92.40 1.05 0.85 1.28
____Multiple myeloma  43 42.73 1.01 0.73 1.36
____Leukemia  90 88.86 1.01 0.81 1.24
Benign & unspecified nature neoplasms 18 27.05 0.67 0.39 1.05
Diseases blood & blood-forming organs 31 39.17 0.79 0.54 1.12
Diabetes mellitus  112 220.61 0.51** 0.42 0.61
Mental & psychiatric disorders  107 136.79 0.78** 0.64 0.95
Alcoholism  13 22.64 0.57* 0.31 0.98
Other mental disorders  94 114.15 0.82 0.67 1.01
Nervous system disorders  249 263.28 0.95 0.83 1.07
Heart diseases  1871 2486.51 0.75**  0.72 0.79
Other diseases of the circulatory system 509 645.79 0.79** 0.72 0.86
Diseases respiratory system  743 839.70 0.88** 0.82 0.95
__Acute resp. infection, except. flu, pneumonia  3 1.22 2.46 0.51 7.19
__Influenza  3 2.76 1.09 0.22 3.18
__Pneumonia  129 196.57 0.66** 0.55 0.78
__COPD  461 485.42 0.95 0.86 1.04
__Asthma  5 7.28 0.69 0.22 1.60
__Asbestosis  48 4.11 11.68** 8.61 15.48
__Silicosis  0 0.68 0.00 0.00 5.44
__Other pneumoconiosis  1 4.00 0.25 0.01 1.39
__Other respiratory diseases  93 137.67 0.68** 0.55 0.83
Diseases digestive system  194 267.43 0.73** 0.63 0.84
Diseases skin & subcutaneous  5 8.21 0.61 0.20 1.42
Diseases musculoskeletal & connective 14 23.02 0.61 0.33 1.02
Diseases genito-urinary system  109 162.66 0.67** 0.55 0.81
Symptoms & ill-defined conditions  39 59.32 0.66** 0.47 0.90
Transportation injuries  62 71.80 0.86 0.66 1.11
Falls  63 56.42 1.12 0.86 1.43
Other injury  56 75.67 0.74* 0.56 0.96
Violence  65 90.62 0.72** 0.55 0.91
Other & unspecified causes   164 206.96 0.79** 0.68 0.92
           
* Two-Sided P < 0.05
** Two-Sided P < 0.01


Table III: Sheet Metal Worker Mortality by Time since Entry into Sheet Metal Trade
Disease Category
Time Since
Trade Entry
(years)
Obs.
Exp.

SMR

95% CI
Lower
95% CI
Upper
Lung Cancer 
20 to 29 
31
28.18
1.10
0.75
1.56
30 to 39
142
144.32
0.98
0.83
1.16
40 to 49
309
307.80
1.00
0.90
1.12
 
50+ 
326
304.47
1.07
0.96
1.19
MN Pleura 
20 to 29 
1
0.08
12.15
0.31
67.71
30 to 39
1
0.32
3.10
0.08
17.25
40 to 49
6
0.62
9.73**
3.57
21.17
 
50+ 
3
0.46
6.51*
1.34
19.02
Mesothelioma
20 to 29 
1
0.13
7.54
0.19
42.02
30 to 39
7
1.09
6.43**
2.58
13.24
40 to 49
36
3.66
9.84**
6.89
13.63
 
50+ 
41
6.70
6.12**
4.39
8.31
COPD 
20 to 29 
7
11.67
0.60
0.24
1.24
30 to 39
29
53.80
0.54**
0.36
0.77
40 to 49
131
157.35
0.83**
0.70
0.99
 
50+ 
294
262.54
1.12
1.00
1.26
Asbestosis 
20 to 29 
2
0.08
24.83**
3.01
89.69
30 to 39
4
0.36
11.13**
3.03
28.50
40 to 49
11
1.19
9.23**
4.60
16.52
 
50+ 
31
2.48
12.51**
8.50
17.75
             
* Two-Sided P < 0.05
** Two-Sided P < 0.01

 

Table IV: Sheet Metal Worker Mortality by Chest X-Ray Parenchymal Category
Disease
Category 
Parenchymal
Change 
Category 
Obs. 
Exp. 
SMR 
95% CI
Lower
95% CI
Upper
Lung Cancer 
0/- to 0/1 
649
693.46
0.94
0.87
1.01
1/0 to 1/2
143
86.85
1.65
1.39
1.94
2/1 to 2/3
14
4.28
3.27
1.79
5.48
3/2 to 3/+
2
0.26
7.58
0.92
27.38
MN Pleura 
0/- to 0/1 
10
1.27
7.89
3.78
14.51
1/0 to 1/2
1
0.20
4.92
0.12
27.40
2/1 to 2/3
0
0.01
0.00
0.00
300.99
3/2 to 3/+
0
<0.01 
0.00
0.00
6087.99
Mesothelioma 
0/- to 0/1 
69
10.37
6.65
5.18
8.42
1/0 to 1/2
16
1.16
13.85
7.91
22.49
2/1 to 2/3
0
0.04
0.00
0.00
82.29
3/2 to 3/+
0
<0.01 
0.00
0.00
1050.38
COPD 
0/- to 0/1 
356
423.03
0.84
0.76
0.93
1/0 to 1/2
92
58.90
1.56
1.26
1.92
2/1 to 2/3
12
3.31
3.63
1.87
6.33
3/2 to 3/+
1
0.18
5.52
0.14
30.74
Asbestosis 
0/- to 0/1 
27
3.58
7.54
4.96
10.96
1/0 to 1/2
18
0.50
35.96
21.30
56.84
2/1 to 2/3
3
0.03
116.75
24.08
341.20
3/2 to 3/+
0
<0.01 
0.00
0.00
2446.02
             
* Two-Sided P < 0.05
** Two-Sided P < 0.01

 

Table V: Sheet Metal Worker Mortality by Chest X-Ray Pleural Category
Disease
Category
Pleural
Changes
Obs.
Exp.
SMW
95% CI
Lower
95% CI
Upper
Lung Cancer 
No
597
585.39
1.02
0.94
1.11
Yes
211
199.47
1.06
0.92
1.21
MN Pleura 
No
7
1.05
6.66**
2.68
13.73
Yes
4
0.43
9.25**
2.52
23.68
Mesothelioma
No
53
8.72
6.08**
4.55
7.95
Yes
32
2.86
11.19**
7.65
15.79
COPD 
No
287
349.36
0.82**
0.73
0.92
Yes
174
136.06
1.28**
1.10
1.48
Asbestosis 
No
21
2.94
7.15**
4.42
10.93
Yes
27
1.17
23.03**
15.17
33.51
             
* Two-Sided P < 0.05
** Two-Sided P < 0.01

 

Table VI: Cox Model Chest Radiograph Predictors of Lung Cancer Mortality1
Risk Predictor Number in
Model2
No. of
Cancer
Cases
Relative
Risk2
95% CI
Lower
>95% CI
Upper
Profusion Categories4
0/- - 0/0 
13066
526
1.00
Ref 
Ref 
0/1 
1341
78
1.10
0.86
1.39
1/0 - 1/2 
1559
131
1.48
1.22
1.80
2/1 - 2/3 
95
12
2.83
1.59
5.04
3/2 - 3/+ 
7
2
3.96
0.98
16.05
Pleural Abnormalities
Negative 
13997
635
1.00
Ref 
Ref 
Positive 
2071
114
0.91
0.74
1.11
Years Since Last Sheet Metal Trade Work at Exam
<1 
9880
366
1.00
Ref 
Ref 
1-5 
2942
189
1.33
1.09
1.62
5-10 
1884
102
1.21
0.91
1.60
>10 
1362
92
2.00
1.43
2.79
Age3 
16068
749
1.045
1.028
1.061
Smoking Pack-Years3
16068
749
1.026
1.021
1.031

Years in Sheet Metal Trade3

16068
749
1.026
1.01
1.042

1 Cox proportional hazard analyses based on 16068 Caucasian males with 20 or more years in sheet metal trade and having data on other model covariates.

2 Stratified Cox model with strata defined by smoking status (never, past, and current) and adjusted for pack-years of smoking (0, 1-19, 20-39, or ≥ 40 pack-years), age at start of follow-up (<50, 50-54, 55-59, 60-64, 65-69, 70-74, or ≥ 75 years), time since last work in the sheet metal trade at start of follow-up (<5, 5-9, or ≥ 10 years), years of sheet metal work beyond 20 years (20-24, 25-29, 30-34, or ≥ 35 years), presence of pleural abnormalities, and profusion category. A pleural abnormality was defined as bilateral pleural thickening or plaques, with or without calcification.

3 Age, pack-years of smoking, and years of sheet metal work entered as grouped continuous variables.

4 Test for trend across profusions categories, p<0.0001.

 

Table VII: Cox Model Spirometry Predictors of Lung Cancer Mortality1
Risk Predictor Number in
Model2
No. of
Cancer
Cases
Relative
Risk2
>95% CI
Lower
95% CI
Upper
FVC Percent Predicted4
>=80 
7354 
226 
1.00
Ref 
Ref 
70-79 
1159 
66 
1.16
0.92
1.65
60-69 
471 
43 
1.54
0.94
2.50
<60 
240 
28
1.73
0.95
3.16
FEV1 Percent Predicted4
>=80 
7116 
176 
1.00
Ref 
Ref 
70-79 
1020 
70 
1.58
1.13
2.22
60-69 
47 
556 
1.38
0.88
2.17
<60 
532 
70 
1.57
0.84
2.92
FEV1/FVC4
>=80 
3230 
51 
1.00
Ref 
Ref 
70-79 
4110 
137 
1.44
1.03
2.00
60-69 
1290 
100 
2.16
1.47
3.18
<60 
594 
75 
2.79
1.69
4.63
Years Since Last Sheet Metal Trade Work at Exam 
<1 
6088 
186 
1.00
Ref 
Ref 
1-5 
1510 
93 
1.50
1.13
2.00
5-10 
973 
43 
1.12
0.74
1.70
>10 
653 
41 
2.01
1.24
3.25
Age3
9224 
363 
1.033
1.009
1.057
Smoking Pack-Years3
9224 
363 
1.017
1.009
1.025
Years in Sheet Metal Trade3
9224 
363 
1.021
0.999
1.044

1 Cox proportional hazard analyses based on 9224 Caucasian males with 20 or more years in sheet metal trade and having spirometry data and data on other model covariates. Only workers with spirometry meeting ATS reproducibility criteria were included.

2 Stratified Cox model with strata defined by smoking status (never, past, and current) and adjusted for pack-years of smoking (0, 1-19, 20-39, or ≥ 40 pack-years), age at start of follow-up (<50, 50-54, 55-59, 60-64, 65-69, 70-74, or ≥ 75 years), time since last work in the sheet metal trade at start of follow-up (<5, 5-9, or ≥ 10 years), years of 1 sheet metal work beyond 20 years (20-24, 25-29, 30-34, or ≥ 35 years). Pleural abnormalities were not significantly associated with lung cancer in the adjusted model (p=0.38).

3 Age, pack-years of smoking, and years of sheet metal work entered as grouped continuous variables.

4 Test for trend across categories for percent predicted FVC (p=0.0472), FEV1 (p=0.0971) and FEV1/FVC (p<0.0001).

 

Table VIII: Cox Model Predictors of Mesothelioma Mortality
Risk Predictor Number in
Model1
No. of
Cancer
Cases
Relative
Risk2
95% CI
Lower
95% CI
Upper
Profusion Categories
< 1/0 
14407
72
1.00
Ref 
Ref 
• 1/0 
1661
17
1.77
1.03
3.05
Pleural Abnormalities
Negative 
13997
67
1.00
Ref 
Ref 
Positive 
2071
22
1.79
1.09
2.94
Age3 
16068
89
1.063
1.034
1.092

1 Cox proportional hazard analyses based on 16068 Caucasian males with 20 or more years in sheet metal trade and having data on other model covariates. Smoking (p=0.24), years of sheet metal work beyond 20 years (p=0.73), years since last sheet metal work (p=41), percent predicted FVC (p=0.16), and percent predicted FEV1/FVC (p=0.19) were not significant predictors of mesothelioma mortality and were eliminated from the final model.

2 An alternate model with pleural abnormalities defined as any indication of pleural change on the ILO recording form resulted in slightly greater risk (RR= 1.97, 95% CI=1.36-3.06).

3 Age entered as grouped continuous variables.

 

Table IX: Cox Model Predictors of COPD Mortality1
Risk Predictor Number in
Model2
No. of
COPD Cases
Relative
Risk2
95% CI
Lower
95% CI
Upper
Profusion Categories 
< 1/0  9224 187 1.00 Ref  Ref 
• 1/0  1032 59 1.39 1.03 1.88
FVC Percent Predicted4 
>=80  7979 102 1.00 Ref  Ref 
70-79  1353 50 1.17 0.76 1.78
60-69  597 48 1.83 1.12 2.97
<60  327 46 1.61 0.96 2.70
FEV1 Percent Predicted4 
>=80  7678 58 1.00 Ref  Ref 
70-79  1208 28 1.44 0.88 2.37
60-69  674 36 1.87 1.10 3.17
<60  696 124 325 1.71 6.19
FEV1/FVC4 
>=80  3512 14 1.00 Ref  Ref 
70-79  4542 45 1.49 0.81 2.75
60-69  1471 54 3.36 1.78 6.35
<60  731 133 10.61 5.35 21.02
Years Since Last Sheet Metal Trade Work at Exam 
<1  6588 77 1.00 Ref  Ref 
1-5  1723 61 1.75 1.21 2.54
5-10  1152 61 2.17 1.34 3.53
>10  746 47 3.29 1.89 5.75
Age3  10256 246 1.072 1.041 1.104
Smoking Pack-Years3  10256 246 1.022 1.012 1.033
Years in Sheet Metal Trade3  10256 246 1.042 1.013 1.071

1 Cox proportional hazard analyses based on 10256 Caucasian males with 20 or more years in sheet metal trade, spirometry, and having data on other model covariates. Only workers with spirometry meeting ATS reproducibility were included. Pleural changes (p=0.73) were not significantly associated with COPD mortality and this parameter was dropped from the final model.

2 Stratified Cox model with strata defined by smoking status (never, past, and current) and adjusted for pack-years of smoking (0, 1-19, 20-39, or ≥ 40 pack-years), age at start of follow-up (<50, 50-54, 55-59, 60-64, 65-69, 70-74, or ≥ 75 years), time since last work in the sheet metal trade at start of follow-up (<5, 5-9, or ≥ 10 years), years of sheet metal work beyond 20 years (20-24, 25-29, 30-34, or ≥ 35 years), and presence of pleural abnormalities.

3 Age, pack-years of smoking, and years of sheet metal work entered as grouped continuous variables.

4 Test for trend across categories for percent predicted FVC (p=0.0244), FEV1 (p=<0.0001) and FEV1/FVC (p<0.0001).