ED^ United States Environmental Protection Agency Office of Chemical Safety and Pollution Prevention Final Risk Evaluation for Asbestos Part 1: Chrysotile Asbestos Systematic Review Supplemental File: Data Quality Evaluation for Epidemiological Studies of Ovarian and Laryngeal Cancers December 2020 ------- Table Listing l Acheson et al. 1982: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes 3 2 Acheson et al. 1982: Evaluation of Cancer for Lung Cancer Mortality Outcomes . 6 3 Newhouse and Sullivan 1989: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes ^ 4 Newhouse and Sullivan 1989: Evaluation of Cancer for Female Lung Cancer Mor- tality Outcomes 12 5 Newhouse and Sullivan 1989: Evaluation of Cancer for Male and Female Lung Cancer Mortality Outcomes 15 6 Newhouse and Sullivan 1989: Evaluation of Cancer for Laryngeal Cancer Mortal- ity Outcomes 18 7 Gardner and Powell 1986: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes 21 8 Gardner and Powell 1986: Evaluation of Cancer for Female Lung Cancer Mortality Outcomes 24 9 Gardner and Powell 1986: Evaluation of Cancer for Male and Female Lung Cancer Mortality Outcomes 27 10 Gardner and Powell 1986: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes 30 11 Pira et al. 2017: Evaluation of Cancer for Lung Cancer Mortality Outcomes ... 33 12 Pira et al. 2017: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes gg 13 Mamo and Costa 2004: Evaluation of Cancer for Ovarian Cancer Mortality Out- comes 39 14 Mamo and Costa 2004: Evaluation of Cancer for Female Lung Cancer Mortality Outcomes 42 15 Mamo and Costa 2004: Evaluation of Cancer for Male and Female Lung Cancer Mortality Outcomes 44 16 Mamo and Costa 2004: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes 46 17 Germani et al. 1999: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes 49 18 Germani et al. 1999: Evaluation of Cancer for Lung Cancer Mortality Outcomes 52 19 Germani et al. 1999: Evaluation of Cancer for Laryngeal Cancer Mortality Out- comes 55 20 Tarchi et al. 1994: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes 53 21 Tarchi et al. 1994: Evaluation of Cancer for Female Lung Cancer Mortality Outcomes 60 22 Tarchi et al. 1994: Evaluation of Cancer for Male and Female Lung Cancer Mortality Outcomes 62 23 Tarchi et al. 1994: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes 64 2 of 65 ------- Table 1: Acheson et al. 1982: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes Study Citation: Acheson, E. D., Gardner, M. J., Pippard, E. C., Grime, L. P. (1982). Mortality of two groups of women who manufactured gas masks from chrysotile and crocidolite asbestos: a 40-year follow-up Occupational and Environmental Medicine, 39(4,4), 344-348 Data Type: gasmask manufacturing women ovarian-Cancer HERO ID: 32 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 The study setting was described and details on the mask manufacturing process and products were pro- vided. Subjects were included based on residence in Blackburn in 1939 and indication of mask manu- facturing on their employment records from the Of- fice of Population Censuses and Surveys. There is some uncertainty whether employment records from this resource would be complete for all women in the area. High X 0.4 0.4 A small number of women (4) died before the follow- up period began (1951), but this is not expected to appreciably bias the results. No other indication of loss-to-follow-up was provided. Medium x 0.2 0.4 Rates of mortality for England and Wales (1951- 1980) were used to compare with women from Black- burn. Local mortality rates 1968-1978 were used to adjust for location. The study authors note that rates of cause-specific mortality for the relevant area (Blackburn, Leyland, and Preston) were only avail- able for this time period. Applying mortality rates from this specific time period to the full follow-up period may insert some uncertainty into the results. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated High x 0.5 NA x 0.5 1.5 NA 0.5 Exposure was assessed by employment at each of the asbestos gas mask manufacturing facilities only. Temporality was established. Women residing in Blackburn in 1939 with employment records showing gas mask manufacturing were followed until 1980. This is a sufficiently long follow-up period for deter- mining rates of cause-specific mortality. Domain 3: Outcome Assessment Continued on next page 3 of 65 ------- . continued from previous page Study Citation: Acheson, E. D., Gardner, M. J., Pippard, E. C., Grime, L. P. (1982). Mortality of two groups of women who manufactured gas masks from chrysotile and crocidolite asbestos: a 40-year follow-up Occupational and Environmental Medicine, 39(4,4), 344-348 Data Type: gasmask manufacturing women ovarian-Cancer HERO ID: 32 Domain Metric Rating^ MWF* Score Comments^ Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 Vital status was determined form 1951 to 1980 for all subjects. It was not expressly stated, however, vital status was presumably obtained from the Na- tional Health Service Central Register. Mortality from cancer was detailed using ICD-8 codes. His- tological confirmation of cancer deaths was not de- scribed. Metric 8: Reporting Bias Medium x 0.333 0.67 Outcomes prespecified in the abstract, introduction, and methods were provided either qualitatively or quantitatively in the results. SMR estimates were provided with observed cases, expected cases, and confidence intervals. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Medium x 0.5 1 Only women were included in the analysis. Age- and calendar period-specific (in 5-year bins) mor- tality rates were calculated, accounting for age and time period. Individual smoking information was not available and not accounted for in the analysis. Medium X 0.25 0.5 Covariate information was presumably obtained from the same source as vital status and employment—the National Health Service Central Register, Office of Population Censuses and Surveys. Medium X 0.25 0.5 Masks manufactured at the Blackburn facility were noted to contain only chrysotile, charcoal, and merino wool. Crocidolite was not reported to be used at the relevant manufacturing facility. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Metric 14: Reproducibility of analyses Medium X 0.4 0.8 Cause-specific mortality for women employed in the mask manufacturing industry in Blackburn were de- termined by calculating SMRs using the person- years method. This is an appropriate design for the study question. Medium X 0.2 0.4 A total of 570 women were included in the analy- sis and followed for a total of 14,324 person-years. This is a sufficiently large population to detect an effect. The number of women observed with ovarian cancer was relatively low (n=5) and may need to be interpreted with caution. Medium X 0.2 0.4 Analyses were adequately described and could be re- produced given original data. Continued on next page 4 of 65 ------- . continued from previous page Study Citation: Acheson, E. D., Gardner, M. J., Pippard, E. C., Grime, L. P. (1982). Mortality of two groups of women who manufactured gas masks from chrysotile and crocidolite asbestos: a 40-year follow-up Occupational and Environmental Medicine, 39(4,4), 344-348 Data Type: gasmask_manufacturing_women_ovarian-Cancer HERO ID: 32 Domain Metric Rating^ MWF* Score Comments^ Metric 15 Statistical models Medium x 0.2 0.4 The statistical methods used were tails on the area adjustment were this was not the primary outcome transparent. De- limited, however, of the study. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 1.9 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating, ft This metric met the criteria for high confidence as expected for this type of study 5 of 65 ------- Table 2: Acheson et al. 1982: Evaluation of Cancer for Lung Cancer Mortality Outcomes Study Citation: Acheson, E. D., Gardner, M. J., Pippard, E. C., Grime, L. P. (1982). Mortality of two groups of women who manufactured gas masks from chrysotile and crocidolite asbestos: a 40-year follow-up Occupational and Environmental Medicine, 39(4,4), 344-348 Data Type: gasmask manufacturing women lung-Cancer HERO ID: 32 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 The study setting was described and details on the mask manufacturing process and products were pro- vided. Subjects were included based on residence in Blackburn in 1939 and indication of mask manu- facturing on their employment records from the Of- fice of Population Censuses and Surveys. There is some uncertainty whether employment records from this resource would be complete for all women in the area. High X 0.4 0.4 A small number of women (4) died before the follow- up period began (1951), but this is not expected to appreciably bias the results. No other indication of loss-to-follow-up was provided. Medium x 0.2 0.4 Rates of mortality for England and Wales (1951- 1980) were used to compare with women from Black- burn. Local mortality rates 1968-1978) were used to adjust for location. The study authors note that rates of cause-specific mortality for the relevant area (Blackburn, Leyland, and Preston) were only avail- able for this time period. Applying mortality rates from this specific time period to the full follow-up period may insert some uncertainty into the results. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated High x 0.5 NA x 0.5 1.5 NA 0.5 Exposure was assessed by employment at each of the asbestos gas mask manufacturing facilities only. Temporality was established. Women residing in Blackburn in 1939 with employment records showing gas mask manufacturing were followed until 1980. This is a sufficiently long follow-up period for deter- mining rates of cause-specific mortality. Domain 3: Outcome Assessment Continued on next page 6 of 65 ------- . continued from previous page Study Citation: Acheson, E. D., Gardner, M. J., Pippard, E. C., Grime, L. P. (1982). Mortality of two groups of women who manufactured gas masks from chrysotile and crocidolite asbestos: a 40-year follow-up Occupational and Environmental Medicine, 39(4,4), 344-348 Data Type: gasmask manufacturing women lung-Cancer HERO ID: 32 Domain Metric Rating^ MWF* Score Comments^ Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 Vital status was determined form 1951 to 1980 for all subjects. It was not expressly stated, however, vital status was presumably obtained from the Na- tional Health Service Central Register. Mortality from cancer was detailed using ICD-8 codes. His- tological confirmation of cancer deaths was not de- scribed. Metric 8: Reporting Bias Low x 0.333 1.0 Outcomes prespecified in the abstract, introduction, and methods were provided either qualitatively or quantitatively in the results. SMR estimates were provided with observed cases, expected cases, and confidence intervals. Reported cases included one mesothelioma case Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Medium x 0.5 1 Only women were included in the analysis. Age- and calendar period-specific (in 5-year bins) mor- tality rates were calculated, accounting for age and time period. Individual smoking information was not available and not accounted for in the analysis. Medium X 0.25 0.5 Covariate information was presumably obtained from the same source as vital status and employment—the National Health Service Central Register, Office of Population Censuses and Surveys. Medium X 0.25 0.5 Masks manufactured at the Blackburn facility were noted to contain only chrysotile, charcoal, and merino wool. Crocidolite was not reported to be used at the relevant manufacturing facility. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Metric 14: Reproducibility of analyses Metric 15: Statistical models Medium X 0.4 0.8 Cause-specific mortality for women employed in the mask manufacturing industry in Blackburn were de- termined by calculating SMRs using the person- years method. This is an appropriate design for the study question. Medium X 0.2 0.4 A total of 570 women were included in the analysis and followed for a total of 14,324 person-years. This is a sufficiently large population to detect an effect. Medium X 0.2 0.4 Analyses were adequately described and could be re- produced given original data. Medium X 0.2 0.4 The statistical methods used were transparent. De- tails on the area adjustment were limited, however, this was not the primary outcome of the study. Continued on next page 7 of 65 ------- . continued from previous page Study Citation: Acheson, E. D., Gardner, M. J., Pippard, E. C., Grime, L. P. (1982). Mortality of two groups of women who manufactured gas masks from chrysotile and crocidolite asbestos: a 40-year follow-up Occupational and Environmental Medicine, 39(4,4), 344-348 Data Type: gasmask_manufacturing_women_lung-Cancer HERO ID: 32 Domain Metric Rating^ MWF* Score Comments^ Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 2.0 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 8 Of 65 ------- Table 3: Newhouse and Sullivan 1989: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos workers SMR_women ovary-Cancer HERO ID: 3082792 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 Retrospective occupational mortality cohort includ- ing 13 450 subjects (men and women) first employed between 1941 and 1979 in one factory located in the UK. Follow-up is until 1986. Nine men and one woman with unknown dates of birth have been ex- cluded from the original cohort. At the end of 1979, 2.6% of workers had emigrated. Setting of the plant is well described. No other details are provided on participants. High X 0.4 0.4 Little loss to follow-up due to emigration (2.6%). Medium X 0.2 0.4 Although not clearly spelled out in this reference, Berry and Newhouse 1983 (HERO ID 43) mentions use of national death rates as comparison group. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Medium x 0.5 NA x 0.5 1.5 NA Exposure assessed based solely on duration of em- ployment at the plant. Job titles were not available. Exposure levels are reported at the plant level (over 20 fibers/ml before 1931, 5-20 f/mL from 1931-1950, less than 5 f/mL after 1970). Adequate follow-up (43 years) for cancer mortality to be observed. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Low Metric 8: Reporting Bias Medium X 0.667 2 Death certificates from the National Health Service Central Registrar, were coded to the 8th revision of the International Classification of Disease (ICD). There are no other details provided on outcome assessment, and ICD codes are only re- ported for chronic respiratory diseases. X 0.333 0.67 Some cancer mortality results are reported for "other causes", some are spelled out in the text, but it is unclear if all are reported. Exposure levels are re- ported at the plant level (over 20 fibers/ml before 1931, 5-20 f/mL from 1931-1950, less than 5 f/mL after 1970). Domain 4: Potential Confounding/Variable Control Continued on next page 9 of 65 ------- .. . continued from previous page Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos workers SMR_women ovary-Cancer HERO ID: 3082792 Domain Metric Rating^ MWF* Score Comments^ Metric 9: Metric 10 Metric 11 Covariate Adjustment Covariate Characterization Co-exposure Confounding Medium Low Low x 0.5 x 0.25 x 0.25 1 0.75 0.75 Sex, age and period were accounted for. Smoking was not accounted for. There is no discussion of covariate characterization, but likely it came from employment records and death certificates. There is no specific discussion of potential co- exposures, however the occupational setting sug- gests potential for other exposures, which could be differential depending on the job titles of the work- ers. Domain 5: Analysis Metric 12 Metric 13 Metric 14 Metric 15 Study Design and Methods Statistical power Reproducibility of analyses Statistical models Medium Medium Medium Medium x 0.4 x 0.2 x 0.2 x 0.2 0.8 0.4 0.4 0.4 SMRs were calculated using the subject year method, with 90% confidence intervals (90% CI). Poisson or normal approximation were used depending on the number of observed deaths. The number of participants, number of observed deaths and length of follow-up are adequate. Basic details are provided in the report. Adequate models. SMRs based on 30 or fewer deaths confidence intervals were computed with the exact method based on the Pois- son distribution. The normal approximation was used for others. Domain 6: Other Considera Metric 16 Metric 17 Metric 18 Metric 19 Metric 20 Metric 21 Metric 22 ions for Biomarker Selection and Measurement Use of Biomarker of Exposure Effect biomarker Method Sensitivity Biomarker stability Sample contamination Method requirements Matrix adjustment NA NA NA NA NA NA NA NA NA NA NA NA NA NA Overall Quality Determination1" Low 2.3 Extracted Yes Continued on next page . .. 10 of 65 ------- . continued from previous page Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 D at a Type: Asbestos_workers_SMR_women_ovary- Cancer HERO ID: 3082792 Domain Metric Rating ^ MWF* Score Comments^ * MWF = Metric Weighting Factor t High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. + The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. if any metric is Unacceptable Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 11 of 65 ------- Table 4: Newhouse and Sullivan 1989: Evaluation of Cancer for Female Lung Cancer Mortality Outcomes Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos workers SMR_women lung-Cancer HERO ID: 3082792 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 Retrospective occupational mortality cohort includ- ing 13 450 subjects (men and women) first employed between 1941 and 1979 in one factory located in the UK. Follow-up is until 1986. Nine men and one woman with unknown dates of birth have been ex- cluded from the original cohort. At the end of 1979, 2.6% of workers had emigrated. Setting of the plant is well described. No other details are provided on participants. High X 0.4 0.4 Little loss to follow-up due to emigration (2.6%). Medium X 0.2 0.4 Although not clearly spelled out in this reference, Berry and Newhouse 1983 (HERO ID 43) mentions use of national death rates as comparison group. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Medium x 0.5 NA x 0.5 1.5 NA Exposure assessed based solely on duration of em- ployment at the plant. Job titles were not available. Exposure levels are reported at the plant level (over 20 fibers/ml before 1931, 5-20 f/mL from 1931-1950, less than 5 f/mL after 1970). Adequate follow-up (43 years) for cancer mortality to be observed. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Low Metric 8: Reporting Bias Medium X 0.667 2 Death certificates from the National Health Service Central Registrar, were coded to the 8th revision of the International Classification of Disease (ICD). There are no other details provided on outcome assessment, and ICD codes are only re- ported for chronic respiratory diseases. X 0.333 0.67 Some cancer mortality results are reported for "other causes", some are spelled out in the text, but it is unclear if all are reported. Exposure levels are re- ported at the plant level (over 20 fibers/ml before 1931, 5-20 f/mL from 1931-1950, less than 5 f/mL after 1970). Domain 4: Potential Confounding/Variable Control Continued on next page 12 of 65 ------- .. . continued from previous page Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos workers SMR_women lung-Cancer HERO ID: 3082792 Domain Metric Rating^ MWF* Score Comments^ Metric 9: Metric 10 Metric 11 Covariate Adjustment Covariate Characterization Co-exposure Confounding Medium Low Low x 0.5 x 0.25 x 0.25 1 0.75 0.75 Sex, age and period were accounted for. Smoking was not accounted for. There is no discussion of covariate characterization, but likely it came from employment records and death certificates. There is no specific discussion of potential co- exposures, however the occupational setting sug- gests potential for other exposures, which could be differential depending on the job titles of the work- ers. Domain 5: Analysis Metric 12 Metric 13 Metric 14 Metric 15 Study Design and Methods Statistical power Reproducibility of analyses Statistical models Medium Medium Medium Medium x 0.4 x 0.2 x 0.2 x 0.2 0.8 0.4 0.4 0.4 SMRs were calculated using the subject year method, with 90% confidence intervals (90% CI). Poisson or normal approximation were used depending on the number of observed deaths. The number of participants, number of observed deaths and length of follow-up are adequate. Basic details are provided in the report. Adequate models. SMRs based on 30 or fewer deaths confidence intervals were computed with the exact method based on the Pois- son distribution. The normal approximation was used for others. Domain 6: Other Considera Metric 16 Metric 17 Metric 18 Metric 19 Metric 20 Metric 21 Metric 22 ions for Biomarker Selection and Measurement Use of Biomarker of Exposure Effect biomarker Method Sensitivity Biomarker stability Sample contamination Method requirements Matrix adjustment NA NA NA NA NA NA NA NA NA NA NA NA NA NA Overall Quality Determination1" Low 2.3 Extracted Yes Continued on next page . .. 13 of 65 ------- . continued from previous page Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos_workers_SMR_women_lung-Cancer HERO ID: 3082792 Domain Metric Rating ^ MWF* Score Comments^ * MWF = Metric Weighting Factor t High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. + The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. if any metric is Unacceptable Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 14 of 65 ------- Table 5: Newhouse and Sullivan 1989: Evaluation of Cancer for Male and Female Lung Cancer Mortality Outcomes Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos workers SMR_combined lung-Cancer HERO ID: 3082792 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 Retrospective occupational mortality cohort includ- ing 13 450 subjects (men and women) first employed between 1941 and 1979 in one factory located in the UK. Follow-up is until 1986. Nine men and one woman with unknown dates of birth have been ex- cluded from the original cohort. At the end of 1979, 2.6% of workers had emigrated. Setting of the plant is well described. No other details are provided on participants. High X 0.4 0.4 Little loss to follow-up due to emigration (2.6%). Medium X 0.2 0.4 Although not clearly spelled out in this reference, Berry and Newhouse 1983 (HERO ID 43) mentions use of national death rates as comparison group. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Medium x 0.5 NA x 0.5 1.5 NA Exposure assessed based solely on duration of em- ployment at the plant. Job titles were not available. Exposure levels are reported at the plant level (over 20 fibers/ml before 1931, 5-20 f/mL from 1931-1950, less than 5 f/mL after 1970). Adequate follow-up (43 years) for cancer mortality to be observed. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Low Metric 8: Reporting Bias Medium X 0.667 2 Death certificates from the National Health Service Central Registrar, were coded to the 8th revision of the International Classification of Disease (ICD). There are no other details provided on outcome assessment, and ICD codes are only re- ported for chronic respiratory diseases. X 0.333 0.67 Some cancer mortality results are reported for "other causes", some are spelled out in the text, but it is unclear if all are reported. Exposure levels are re- ported at the plant level (over 20 fibers/ml before 1931, 5-20 f/mL from 1931-1950, less than 5 f/mL after 1970). Domain 4: Potential Confounding/Variable Control Continued on next page 15 of 65 ------- .. . continued from previous page Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos workers SMR_combined lung-Cancer HERO ID: 3082792 Domain Metric Rating^ MWF* Score Comments^ Metric 9: Metric 10 Metric 11 Covariate Adjustment Covariate Characterization Co-exposure Confounding Medium Low Low x 0.5 x 0.25 x 0.25 1 0.75 0.75 Sex, age and period were accounted for. Smoking was not accounted for. There is no discussion of covariate characterization, but likely it came from employment records and death certificates. There is no specific discussion of potential co- exposures, however the occupational setting sug- gests potential for other exposures, which could be differential depending on the job titles of the work- ers. Domain 5: Analysis Metric 12 Metric 13 Metric 14 Metric 15 Study Design and Methods Statistical power Reproducibility of analyses Statistical models Medium Medium Medium Medium x 0.4 x 0.2 x 0.2 x 0.2 0.8 0.4 0.4 0.4 SMRs were calculated using the subject year method, with 90% confidence intervals (90% CI). Poisson or normal approximation were used depending on the number of observed deaths. The number of participants, number of observed deaths and length of follow-up are adequate. Basic details are provided in the report. Adequate models. SMRs based on 30 or fewer deaths confidence intervals were computed with the exact method based on the Pois- son distribution. The normal approximation was used for others. Domain 6: Other Considera Metric 16 Metric 17 Metric 18 Metric 19 Metric 20 Metric 21 Metric 22 ions for Biomarker Selection and Measurement Use of Biomarker of Exposure Effect biomarker Method Sensitivity Biomarker stability Sample contamination Method requirements Matrix adjustment NA NA NA NA NA NA NA NA NA NA NA NA NA NA Overall Quality Determination1" Low 2.3 Extracted Yes Continued on next page . .. 16 of 65 ------- . continued from previous page Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos_workers_SMR_combined_lung-Cancer HERO ID: 3082792 Domain Metric Rating ^ MWF* Score Comments^ * MWF = Metric Weighting Factor t High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. + The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. if any metric is Unacceptable Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 17 Of 65 ------- Table 6: Newhouse and Sullivan 1989: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos workers SMR_men larynx-Cancer HERO ID: 3082792 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 Retrospective occupational mortality cohort includ- ing 13 450 subjects (men and women) first employed between 1941 and 1979 in one factory located in the UK. Follow-up is until 1986. Nine men and one woman with unknown dates of birth have been ex- cluded from the original cohort. At the end of 1979, 2.6% of workers had emigrated. Setting of the plant is well described. No other details are provided on participants. High X 0.4 0.4 Little loss to follow-up due to emigration (2.6%). Medium X 0.2 0.4 Although not clearly spelled out in this reference, Berry and Newhouse 1983 (HERO ID 43) mentions use of national death rates as comparison group. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Medium x 0.5 NA x 0.5 1.5 NA Exposure assessed based solely on duration of em- ployment at the plant. Job titles were not available. Exposure levels are reported at the plant level (over 20 fibers/ml before 1931, 5-20 f/mL from 1931-1950, less than 5 f/mL after 1970). Adequate follow-up (43 years) for cancer mortality to be observed. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Low Metric 8: Reporting Bias Medium X 0.667 2 Death certificates from the National Health Service Central Registrar, were coded to the 8th revision of the International Classification of Disease (ICD). There are no other details provided on outcome assessment, and ICD codes are only re- ported for chronic respiratory diseases. X 0.333 0.67 Some cancer mortality results are reported for "other causes", some are spelled out in the text, but it is unclear if all are reported. Exposure levels are re- ported at the plant level (over 20 fibers/ml before 1931, 5-20 f/mL from 1931-1950, less than 5 f/mL after 1970). Domain 4: Potential Confounding/Variable Control Continued on next page 18 of 65 ------- .. . continued from previous page Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 Data Type: Asbestos workers SMR_men larynx-Cancer HERO ID: 3082792 Domain Metric Rating^ MWF* Score Comments^ Metric 9: Metric 10 Metric 11 Covariate Adjustment Covariate Characterization Co-exposure Confounding Medium Low Low x 0.5 x 0.25 x 0.25 1 0.75 0.75 Sex, age and period were accounted for. Smoking was not accounted for. There is no discussion of covariate characterization, but likely it came from employment records and death certificates. There is no specific discussion of potential co- exposures, however the occupational setting sug- gests potential for other exposures, which could be differential depending on the job titles of the work- ers. Domain 5: Analysis Metric 12 Metric 13 Metric 14 Metric 15 Study Design and Methods Statistical power Reproducibility of analyses Statistical models Medium Medium Medium Medium x 0.4 x 0.2 x 0.2 x 0.2 0.8 0.4 0.4 0.4 SMRs were calculated using the subject year method, with 90% confidence intervals (90% CI). Poisson or normal approximation were used depending on the number of observed deaths. The number of participants, number of observed deaths and length of follow-up are adequate. Basic details are provided in the report. Adequate models. SMRs based on 30 or fewer deaths confidence intervals were computed with the exact method based on the Pois- son distribution. The normal approximation was used for others. Domain 6: Other Considera Metric 16 Metric 17 Metric 18 Metric 19 Metric 20 Metric 21 Metric 22 ions for Biomarker Selection and Measurement Use of Biomarker of Exposure Effect biomarker Method Sensitivity Biomarker stability Sample contamination Method requirements Matrix adjustment NA NA NA NA NA NA NA NA NA NA NA NA NA NA Overall Quality Determination1" Low 2.3 Extracted Yes Continued on next page . .. 19 of 65 ------- . continued from previous page Study Citation: Newhouse, M. L., Sullivan, K. R. (1989). A mortality study of workers manufacturing friction materials: 1941-86 British Journal of Industrial Medicine, 46(3,3), 176-179 D at a Type: Asbestos_workers_SMR_men_larynx- Cancer HERO ID: 3082792 Domain Metric Rating ^ MWF* Score Comments^ * MWF = Metric Weighting Factor t High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. + The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. if any metric is Unacceptable Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 20 of 65 ------- Table 7: Gardner and Powell 1986: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos workers SMR_ovary-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Low Metric 2: Attrition Metric 3: Comparison Group High Low X 0.4 1.2 Retrospective occupational cohort study of 2167 workers (1510 men and 657 women) employed be- tween 1941 and 1983 in chrysotile asbestos ce- ment products factories in England, South Wales and Sweden . This study combines three occupa- tional cohorts. Only the results for men are de- scribed. The periods of employment varied be- tween plants/studies. Eligibility criteria vary be- tween plants (employment for at least 6 months (plant 1), Thomas et al 1982 HERO ID 207), 3 months (plant 2/Swedish cohort, Ohlson and Hog- stedt 1985 HERO D: 3083459 ), none for plant 3 (Garner et al 1986 HERO ID 3083223) Follow-up for ascertainment of vital status was until December 1977 (plant 1), 1982 (plant 2) and Decem- ber 1984 for plant 3. Comparability of cohorts is not discussed. X 0.4 0.4 Loss to follow -up was minimal: 5% in Plant 1, 2% in Plant 2/Swedish cohort (1%), 30% for men in Plant 3 cohort. X 0.2 0.6 SMRs were estimated using England and Wales for comparison population (plants 1 and 3), and Swedish national death rates (Plant 2 cohort), is is unclear what the comparison population for total mortality rates were calculated. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Low x 0.5 NA x 0.5 1.5 NA 1.5 Exposures were based on employment at the re- spective plants, Industrial hygiene measures are de- scribed confirming exposure. For the combined anal- yses type of job was not considered. Follow-up sometimes ended the year after employ- ment, making it an inadequate length of time for cancer mortality to be observed. Domain 3: Outcome Assessment Continued on next page 21 of 65 ------- . continued from previous page Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos workers SMR_ovary-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Metric 7: Outcome measurement or characterization Low x 0.667 Metric 8: Reporting Bias Medium x 0.333 0.67 Sources were different by cohort. Plant 1 and 3 cohorts used death certificates from the National Health Service Central Register and the Department of Health and Social Security. ICD codes are listed for Plant 1 cohort. In the Plant 2 cohort, vital status was traced through the death and burial books of the parishes. Death certificates were checked with the Swedish National Central Bureau of Statistics and the officially determined underlying cause of death has been used. ICD codes are not listed for cohort 3. There is potential for outcome misclassification from death certificates. In addition, for the analy- sis combining the three cohorts, there is high poten- tial misclassification due to various sources, national practices, etc . Measured outcomes were reported, out unclear how the reference population was used for the combined cancers across cohorts. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Low X 0.5 1.5 Although not specifically stated, age and calendar year were accounted for in the Plant-specific analy- ses, making it likely that the total SMR would be adjusted similarly. Low X 0.25 0.75 No discussion of covariate characterization, likely from plant records and death certificates. Medium X 0.25 0.5 For plant 2: "The vast majority of asbestos used was chrysotile but 630 tons of amosite were used between 1949 and 1951 and 400 tons of crocidolite in 1962. The use of these amphiboles was estimated to be less than 1% of all asbestos used." Plant 3 only used chrysotile asbestos except for a "small" amount of amosite during 4 months in 1976. This is not expected to appreciably bias the results. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Medium X 0.4 0.8 Adequate design and methods. SMRs were esti- mated using Poisson distribution assumptions (plant 2), or by the person years method using a computer program (Plants 1 and 3). Medium X 0.2 0.4 2167 workers (1510 men and 657 women) were in- cluded in the analysis, however the short follow-up may limit the power to detect an effect. Continued on next page 22 of 65 ------- . continued from previous page Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos_workers_SMR_ovary-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Metric 14 Metric 15 Reproducibility of analyses Statistical models Medium Low x 0.2 x 0.2 © © Unclear how/what the referent group was used for the combined cohorts analysis. There is no description of how the combined SMRs were calculated. Domain 6: Other Considera Metric 16 Metric 17 Metric 18 Metric 19 Metric 20 Metric 21 Metric 22 ions for Biomarker Selection and Measurement Use of Biomarker of Exposure Effect biomarker Method Sensitivity Biomarker stability Sample contamination Method requirements Matrix adjustment NA NA NA NA NA NA NA NA NA NA NA NA NA NA Overall Quality Determination1" Low 2.6 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 23 of 65 ------- Table 8: Gardner and Powell 1986: Evaluation of Cancer for Female Lung Cancer Mortality Outcomes Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos workers SMR_lung-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Low Metric 2: Attrition Metric 3: Comparison Group High Low X 0.4 1.2 Retrospective occupational cohort study of 2167 workers (1510 men and 657 women) employed be- tween 1941 and 1983 in chrysotile asbestos ce- ment products factories in England, South Wales and Sweden . This study combines three occupa- tional cohorts. Only the results for men are de- scribed. The periods of employment varied be- tween plants/studies. Eligibility criteria vary be- tween plants (employment for at least 6 months (plant 1), Thomas et al 1982 HERO ID 207), 3 months (plant 2/Swedish cohort, Ohlson and Hog- stedt 1985 HERO D: 3083459 ), none for plant 3 (Garner et al 1986 HERO ID 3083223) Follow-up for ascertainment of vital status was until December 1977 (plant 1), 1982 (plant 2) and Decem- ber 1984 for plant 3. Comparability of cohorts is not discussed. X 0.4 0.4 Loss to follow -up was minimal: 5% in Plant 1, 2% in Plant 2/Swedish cohort (1%), 30% for men in Plant 3 cohort. X 0.2 0.6 SMRs were estimated using England and Wales for comparison population (plants 1 and 3), and Swedish national death rates (Plant 2 cohort), is is unclear what the comparison population for total mortality rates were calculated. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Low x 0.5 NA x 0.5 1.5 NA 1.5 Exposures were based on employment at the re- spective plants, Industrial hygiene measures are de- scribed confirming exposure. For the combined anal- yses type of job was not considered. Follow-up sometimes ended the year after employ- ment, making it an inadequate length of time for cancer mortality to be observed. Domain 3: Outcome Assessment Continued on next page 24 of 65 ------- . continued from previous page Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos workers SMR_lung-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Metric 7: Outcome measurement or characterization Low x 0.667 Metric 8: Reporting Bias Medium x 0.333 0.67 Sources were different by cohort. Plant 1 and 3 cohorts used death certificates from the National Health Service Central Register and the Department of Health and Social Security. ICD codes are listed for Plant 1 cohort. In the Plant 2 cohort, vital status was traced through the death and burial books of the parishes. Death certificates were checked with the Swedish National Central Bureau of Statistics and the officially determined underlying cause of death has been used. ICD codes are not listed for cohort 3. There is potential for outcome misclassification from death certificates. In addition, for the analy- sis combining the three cohorts, there is high poten- tial misclassification due to various sources, national practices, etc . Measured outcomes were reported, out unclear how the reference population was used for the combined cancers across cohorts. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Low X 0.5 1.5 Although not specifically stated, age and calendar year were accounted for in the Plant-specific analy- ses, making it likely that the total SMR would be adjusted similarly. Low X 0.25 0.75 No discussion of covariate characterization, likely from plant records and death certificates. Medium X 0.25 0.5 For plant 2: "The vast majority of asbestos used was chrysotile but 630 tons of amosite were used between 1949 and 1951 and 400 tons of crocidolite in 1962. The use of these amphiboles was estimated to be less than 1% of all asbestos used." Plant 3 only used chrysotile asbestos except for a "small" amount of amosite during 4 months in 1976. This is not expected to appreciably bias the results. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Medium X 0.4 0.8 Adequate design and methods. SMRs were esti- mated using Poisson distribution assumptions (plant 2), or by the person years method using a computer program (Plants 1 and 3). Medium X 0.2 0.4 2167 workers (1510 men and 657 women) were in- cluded in the analysis, however the short follow-up may limit the power to detect an effect. Continued on next page 25 of 65 ------- . continued from previous page Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos_workers_SMR_lung-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Metric 14 Metric 15 Reproducibility of analyses Statistical models Medium Low x 0.2 x 0.2 © © Unclear how/what the referent group was used for the combined cohorts analysis. There is no description of how the combined SMRs were calculated. Domain 6: Other Considera Metric 16 Metric 17 Metric 18 Metric 19 Metric 20 Metric 21 Metric 22 ions for Biomarker Selection and Measurement Use of Biomarker of Exposure Effect biomarker Method Sensitivity Biomarker stability Sample contamination Method requirements Matrix adjustment NA NA NA NA NA NA NA NA NA NA NA NA NA NA Overall Quality Determination1" Low 2.6 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 26 of 65 ------- Table 9: Gardner and Powell 1986: Evaluation of Cancer for Male and Female Lung Cancer Mortality Outcomes Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos workers SMR_lung combined-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Low Metric 2: Attrition Metric 3: Comparison Group High Low X 0.4 1.2 Retrospective occupational cohort study of 2167 workers (1510 men and 657 women) employed be- tween 1941 and 1983 in chrysotile asbestos ce- ment products factories in England, South Wales and Sweden . This study combines three occupa- tional cohorts. Only the results for men are de- scribed. The periods of employment varied be- tween plants/studies. Eligibility criteria vary be- tween plants (employment for at least 6 months (plant 1), Thomas et al 1982 HERO ID 207), 3 months (plant 2/Swedish cohort, Ohlson and Hog- stedt 1985 HERO D: 3083459 ), none for plant 3 (Garner et al 1986 HERO ID 3083223) Follow-up for ascertainment of vital status was until December 1977 (plant 1), 1982 (plant 2) and Decem- ber 1984 for plant 3. Comparability of cohorts is not discussed. X 0.4 0.4 Loss to follow -up was minimal: 5% in Plant 1, 2% in Plant 2/Swedish cohort (1%), 30% for men in Plant 3 cohort. X 0.2 0.6 SMRs were estimated using England and Wales for comparison population (plants 1 and 3), and Swedish national death rates (Plant 2 cohort), is is unclear what the comparison population for total mortality rates were calculated. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Low x 0.5 NA x 0.5 1.5 NA 1.5 Exposures were based on employment at the re- spective plants, Industrial hygiene measures are de- scribed confirming exposure. For the combined anal- yses type of job was not considered. Follow-up sometimes ended the year after employ- ment, making it an inadequate length of time for cancer mortality to be observed. Domain 3: Outcome Assessment Continued on next page 27 of 65 ------- . continued from previous page Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos workers SMR_lung combined-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Metric 7: Outcome measurement or characterization Low x 0.667 Metric 8: Reporting Bias Medium x 0.333 0.67 Sources were different by cohort. Plant 1 and 3 cohorts used death certificates from the National Health Service Central Register and the Department of Health and Social Security. ICD codes are listed for Plant 1 cohort. In the Plant 2 cohort, vital status was traced through the death and burial books of the parishes. Death certificates were checked with the Swedish National Central Bureau of Statistics and the officially determined underlying cause of death has been used. ICD codes are not listed for cohort 3. There is potential for outcome misclassification from death certificates. In addition, for the analy- sis combining the three cohorts, there is high poten- tial misclassification due to various sources, national practices, etc . Measured outcomes were reported, out unclear how the reference population was used for the combined cancers across cohorts. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Low X 0.5 1.5 Although not specifically stated, age and calendar year were accounted for in the Plant-specific analy- ses, making it likely that the total SMR would be adjusted similarly. Low X 0.25 0.75 No discussion of covariate characterization, likely from plant records and death certificates. Medium X 0.25 0.5 For plant 2: "The vast majority of asbestos used was chrysotile but 630 tons of amosite were used between 1949 and 1951 and 400 tons of crocidolite in 1962. The use of these amphiboles was estimated to be less than 1% of all asbestos used." Plant 3 only used chrysotile asbestos except for a "small" amount of amosite during 4 months in 1976. This is not expected to appreciably bias the results. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Medium X 0.4 0.8 Adequate design and methods. SMRs were esti- mated using Poisson distribution assumptions (plant 2), or by the person years method using a computer program (Plants 1 and 3). Medium X 0.2 0.4 2167 workers (1510 men and 657 women) were in- cluded in the analysis, however the short follow-up may limit the power to detect an effect. Continued on next page 28 of 65 ------- . continued from previous page Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos_workers_SMR_lung_combined-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Metric 14 Metric 15 Reproducibility of analyses Statistical models Medium Low x 0.2 x 0.2 © © Unclear how/what the referent group was used for the combined cohorts analysis. There is no description of how the combined SMRs were calculated. Domain 6: Other Considera Metric 16 Metric 17 Metric 18 Metric 19 Metric 20 Metric 21 Metric 22 ions for Biomarker Selection and Measurement Use of Biomarker of Exposure Effect biomarker Method Sensitivity Biomarker stability Sample contamination Method requirements Matrix adjustment NA NA NA NA NA NA NA NA NA NA NA NA NA NA Overall Quality Determination1" Low 2.6 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 29 of 65 ------- Table 10: Gardner and Powell 1986: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos workers SMR_larynx-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Low Metric 2: Attrition Metric 3: Comparison Group High Low X 0.4 1.2 Retrospective occupational cohort study of 2167 workers (1510 men and 657 women) employed be- tween 1941 and 1983 in chrysotile asbestos ce- ment products factories in England, South Wales and Sweden . This study combines three occupa- tional cohorts. Only the results for men are de- scribed. The periods of employment varied be- tween plants/studies. Eligibility criteria vary be- tween plants (employment for at least 6 months (plant 1), Thomas et al 1982 HERO ID 207), 3 months (plant 2/Swedish cohort, Ohlson and Hog- stedt 1985 HERO D: 3083459 ), none for plant 3 (Garner et al 1986 HERO ID 3083223) Follow-up for ascertainment of vital status was until December 1977 (plant 1), 1982 (plant 2) and Decem- ber 1984 for plant 3. Comparability of cohorts is not discussed. X 0.4 0.4 Loss to follow -up was minimal: 5% in Plant 1, 2% in Plant 2/Swedish cohort (1%), 30% for men in Plant 3 cohort. X 0.2 0.6 SMRs were estimated using England and Wales for comparison population (plants 1 and 3), and Swedish national death rates (Plant 2 cohort), is is unclear what the comparison population for total mortality rates were calculated. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Low x 0.5 NA x 0.5 1.5 NA 1.5 Exposures were based on employment at the re- spective plants, Industrial hygiene measures are de- scribed confirming exposure. For the combined anal- yses type of job was not considered. Follow-up sometimes ended the year after employ- ment, making it an inadequate length of time for cancer mortality to be observed. Domain 3: Outcome Assessment Continued on next page 30 of 65 ------- . continued from previous page Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos workers SMR_larynx-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Metric 7: Outcome measurement or characterization Low x 0.667 Metric 8: Reporting Bias Medium x 0.333 0.67 Sources were different by cohort. Plant 1 and 3 cohorts used death certificates from the National Health Service Central Register and the Department of Health and Social Security. ICD codes are listed for Plant 1 cohort. In the Plant 2 cohort, vital status was traced through the death and burial books of the parishes. Death certificates were checked with the Swedish National Central Bureau of Statistics and the officially determined underlying cause of death has been used. ICD codes are not listed for cohort 3. There is potential for outcome misclassification from death certificates. In addition, for the analy- sis combining the three cohorts, there is high poten- tial misclassification due to various sources, national practices, etc . Measured outcomes were reported, out unclear how the reference population was used for the combined cancers across cohorts. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Low X 0.5 1.5 Although not specifically stated, age and calendar year were accounted for in the Plant-specific analy- ses, making it likely that the total SMR would be adjusted similarly. Low X 0.25 0.75 No discussion of covariate characterization, likely from plant records and death certificates. Medium X 0.25 0.5 For plant 2: "The vast majority of asbestos used was chrysotile but 630 tons of amosite were used between 1949 and 1951 and 400 tons of crocidolite in 1962. The use of these amphiboles was estimated to be less than 1% of all asbestos used." Plant 3 only used chrysotile asbestos except for a "small" amount of amosite during 4 months in 1976. This is not expected to appreciably bias the results. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Medium X 0.4 0.8 Adequate design and methods. SMRs were esti- mated using Poisson distribution assumptions (plant 2), or by the person years method using a computer program (Plants 1 and 3). Medium X 0.2 0.4 2167 workers (1510 men and 657 women) were in- cluded in the analysis, however the short follow-up may limit the power to detect an effect. Continued on next page 31 of 65 ------- . continued from previous page Study Citation: Gardner, M. J., Powell, C. A. (1986). Mortality of asbestos cement workers using almost exclusively chrysotile fibre Journal of the Society of Occupational Medicine, 36(4,4), 124-126 Data Type: Asbestos_workers_SMR_larynx-Cancer HERO ID: 3083384 Domain Metric Rating^ MWF* Score Comments^ Metric 14 Metric 15 Reproducibility of analyses Statistical models Medium Low x 0.2 x 0.2 © © Unclear how/what the referent group was used for the combined cohorts analysis. There is no description of how the combined SMRs were calculated. Domain 6: Other Considera Metric 16 Metric 17 Metric 18 Metric 19 Metric 20 Metric 21 Metric 22 ions for Biomarker Selection and Measurement Use of Biomarker of Exposure Effect biomarker Method Sensitivity Biomarker stability Sample contamination Method requirements Matrix adjustment NA NA NA NA NA NA NA NA NA NA NA NA NA NA Overall Quality Determination1" Low 2.6 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 32 of 65 ------- Table 11: Pira et al. 2017: Evaluation of Cancer for Lung Cancer Mortality Outcomes Study Citation: Pira, E., Romano, C., Donat , F., Pelucchi, C., La Vecchia, C., BofFetta, P. (2017). Mortality from cancer and other causes among Italian chrysotile asbestos miners Occupational and Environmental Medicine, 74(8,8), 558-563 Data Type: Italian chrysotile miners lung cancer-Cancer HERO ID: 5060134 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Medium x 0.4 Metric 2: Attrition Metric 3: Comparison Group High x 0.4 Medium x 0.2 0.8 Subjects were drawn from the employment records of an Italian asbestos mine (n=1056 men). Those el- igible had worked for the mine for at least one year between 1930 and 1989. Description of the mine set- ting was described including some historical occupa- tional measurements of dust and asbestos. There is potential for some healthy worker effect in this pop- ulation. 0.4 The study authors note that the cause of death could not be determined for a small sample (n=6), and a small percentage (3.8%) of participants emigrated or were otherwise lost to follow-up. This level of attri- tion is not expected to appreciably bias the results. 0.4 Mortality rates for males from the Piedmont Region were used as a comparison group. This is a suf- ficiently similar group, however, the study authors note that mortality rates were not available for cer- tain periods (e.g., 1946-1954) and rates from adja- cent periods of time were used instead (1955-1959 rates applied to 1946-1954). Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Medium x 0.5 Not Rated NA High x 0.5 NA 0.5 Descriptions of the mine and occupational surveil- lance was described. Estimates of historical mean concentrations were 37 fibre/mL up to 1950 and 5 fibre/mL between 1971 and 1976. Exposure was de- termined by employment at the asbestos mine only. Subjects were followed until loss to follow-up, their death, 85th birthday, or through 2014. This is a sufficiently long follow-up period. Domain 3: Outcome Assessment Continued on next page 33 of 65 ------- .. . continued from previous page Study Citation: Pira, E., Romano, C., Donat , F., Pelucchi, C., La Vecchia, C., BofFetta, P. (2017). Mortality from cancer and other causes among Italian chrysotile asbestos miners Occupational and Environmental Medicine, 74(8,8), 558-563 Data Type: Italian chrysotile miners lung cancer-Cancer HERO ID: 5060134 Domain Metric Rating^ MWF* Score Comments^ Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 Cause specific mortalities were obtained from death certificates collected from population registers, mu- nicipal registration offices, and local health authori- ties. Causes of death were coded using ICD-9, how- ever, the study authors did not report whether can- cer cases were histologically confirmed. It's unclear if there may be any misclassification from obtaining vital status or cause of death from various sources. Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. SMRs were presented with observed cases, expected cases, and a confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Medium x 0.5 1 SMRs for male miners were calculated, stratified for age- and calendar-year-specific mortality rates. In- formation on individual smoking behavior was de- scribed, but was not accounted for in the analysis. Metric 10: Covariate Characterization Medium x 0.25 0.5 Covariates were presumably drawn from employ- ment records Metric 11: Co-exposure Confounding Low x 0.25 0.75 Co-exposures are not expressly discussed. There may be additional occupational exposures in this group, which may depend on job title or position. Domain 5: Analysis Metric 12: Study Design and Methods Medium x 0.4 0.8 SMRs were used to assess differences in cause- specific mortality rates between employees of an as- bestos mine compared to a reference population in the same region. This is an appropriate design for the study question. Metric 13: Statistical power Medium x 0.2 0.4 There were a sufficient number of employees in- cluded in the analysis. No concerns with low case numbers. Metric 14: Reproducibility of analyses Medium x 0.2 0.4 The analysis was well-described and could be repro- duced with original data. Metric 15: Statistical models Medium x 0.2 0.4 The method for calculating SMRs was transparent and appropriate. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16: Use of Biomarker of Exposure NA NA Metric 17: Effect biomarker NA NA Metric 18: Method Sensitivity NA NA Metric 19: Biomarker stability NA NA Continued on next page . .. 34 of 65 ------- .. . continued from previous page Study Citation: Pira, E., Romano, C., Donat , F., Pelucchi, C., La Vecchia, C., Boffetta, P. (2017). Mortality from cancer and other causes among Italian chrysotile asbestos miners Occupational and Environmental Medicine, 74(8,8), 558-563 Data Type: Italian chrysotile miners lung cancer-Cancer HERO ID: 5060134 Domain Metric Rating^ MWF* Score Comments^ Metric 20: Sample contamination NA NA Metric 21: Method requirements NA NA Metric 22: Matrix adjustment NA NA Overall Quality Determination1" Medium 1.8 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; X MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. This metric met the criteria for high confidence as expected for this type of study 35 of 65 ------- Table 12: Pira et al. 2017: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes Study Citation: Pira, E., Romano, C., Donat , F., Pelucchi, C., La Vecchia, C., BofFetta, P. (2017). Mortality from cancer and other causes among Italian chrysotile asbestos miners Occupational and Environmental Medicine, 74(8,8), 558-563 Data Type: Italian chrysotile miners laryngeal cancer-Cancer HERO ID: 5060134 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 Subjects were drawn from the employment records of an Italian asbestos mine (n=1056 men). Those el- igible had worked for the mine for at least one year between 1930 and 1989. Description of the mine setting was described including some historical oc- cupational measurements of dust and asbestos. High X 0.4 0.4 The study authors note that the cause of death could not be determined for a small sample (n=6), and a small percentage (3.8%) of participants emigrated or were otherwise lost to follow-up. This level of attri- tion is not expected to appreciably bias the results. Medium X 0.2 0.4 Mortality rates for males from the Piedmont Region were used as a comparison group. This is a suf- ficiently similar group, however, the study authors note that mortality rates were not available for cer- tain periods (e.g., 1946-1954) and rates from adja- cent periods of time were used instead (1955-1959 rates applied to 1946-1954). Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated High x 0.5 NA x 0.5 1.5 NA 0.5 Descriptions of the mine and occupational surveil- lance was described. Estimates of historical mean concentrations were 37 fibre/mL up to 1950 and 5 fibre/mL between 1971 and 1976. Exposure was de- termined by employment at the asbestos mine only. Subjects were followed until loss to follow-up, their death, 85th birthday, or through 2014. This is a sufficiently long follow-up period. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Medium X 0.667 1.33 Cause specific mortalities were obtained from death certificates collected from population registers, mu- nicipal registration offices, and local health authori- ties. Causes of death were coded using ICD-9, how- ever, the study authors did not report whether can- cer cases were histologically confirmed. It's unclear if there may be any misclassification from obtaining vital status or cause of death from various sources. Continued on next page 36 of 65 ------- .. . continued from previous page Study Citation: Pira, E., Romano, C., Donat , F., Pelucchi, C., La Vecchia, C., BofFetta, P. (2017). Mortality from cancer and other causes among Italian chrysotile asbestos miners Occupational and Environmental Medicine, 74(8,8), 558-563 Data Type: Italian chrysotile miners laryngeal cancer-Cancer HERO ID: 5060134 Domain Metric Rating^ MWF* Score Comments^ Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. SMRs were presented with observed cases, expected cases, and a confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Medium x 0.5 1 SMRs for male miners were calculated, stratified for age- and calendar-year-specific mortality rates. In- formation on individual smoking behavior was de- scribed, but was not accounted for in the analysis. Metric 10 Covariate Characterization Medium x 0.25 0.5 Covariates were presumably drawn from employ- ment records Metric 11 Co-exposure Confounding Low x 0.25 0.75 Co-exposures are not expressly discussed. There may be additional occupational exposures in this group, which may depend on job title or position. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 SMRs were used to assess differences in cause- specific mortality rates between employees of an as- bestos mine compared to a reference population in the same region. This is an appropriate design for the study question. Metric 13 Statistical power Medium x 0.2 0.4 There were a sufficient number of employees in- cluded in the analysis. No concerns with low case numbers. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The analysis was well-described and could be repro- duced with original data. Metric 15 Statistical models Medium x 0.2 0.4 The method for calculating SMRs was transparent and appropriate. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 1.9 Extracted Yes Continued on next page . .. 37 of 65 ------- . continued from previous page Study Citation: Pira, E., Romano, C., Donat , F., Pelucchi, C., La Vecchia, C., Boffetta, P. (2017). Mortality from cancer and other causes among Italian chrysotile asbestos miners Occupational and Environmental Medicine, 74(8,8), 558-563 Data Type: Italian_chrysotile_miners_laryngeal_cancer-Cancer HERO ID: 5060134 Domain Metric Rating^ MWF* Score Comments^ MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 38 of 65 ------- Table 13: Mamo and Costa 2004: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian asbestos product workers ovarian cancer-Cancer HERO ID: 6912534 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection High x 0.4 0.4 Subjects consisted of former employees of an asbestos-based products manufacturing plant in Grugliasco, Italy, hired before 1971. Subjects were followed from 1981 through 1995. Metric 2: Attrition Medium x 0.4 0.8 The initial study sample consisted of 1653 employ- ees. The study authors note that the analysis sample consisted of 967 individuals, selected as 18-74 y of age, with concern about dropping older individuals. Metric 3: Comparison Group Medium x 0.2 0.4 The study authors attempted to find a similar popu- lation to the asbestos-products employees by using a reference population consisting of residents from the same area (Turin) whom were manual workers—this was an attempt to mitigate healthy worker effect. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Low x 0.5 1.5 Exposure was determined by occupation only. Ba- sic descriptions of the asbestos product facility were provided. Metric 5: Exposure levels Not Rated NA NA Metric 6: Temporality Medium x 0.5 1 Subjects employed in 1971 were followed-up from 1981 to 1995 which is a sufficiently long follow-up period to detect an effect. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 Causes of death were obtained from the Italian Na- tional Registry of Deaths (ISTAT) and consulting death certificates at the Piedmont Cancer Registry. It was not reported whether cancer cases were his- tologically confirmed, and it appears that causes of death were coded using ICD-9 . Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in results. SMRs were provided with the number of cases and 95% confi- dence intervals. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Medium x 0.5 1 Individual smoking behaviors were not able to be controlled for. Continued on next page . 39 of 65 ------- . continued from previous page Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian asbestos product workers ovarian cancer-Cancer HERO ID: 6912534 Domain Metric Rating^ MWF* Score Comments^ Metric 10 Covariate Characterization Medium x 0.25 0.5 Covariates were largely not collected except sex, which was presumably taken from employment records. Metric 11 Co-exposure Confounding Medium x 0.25 0.5 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 SMRs were calculated to estimate the effect of oc- cupational asbestos exposure on cancer mortality rates. This is an appropriate study design for the study question. Metric 13 Statistical power Medium x 0.2 0.4 A total of 967 individuals were included in the anal- ysis sample, which is sufficiently large to detect an effect. There is some concern for the limited number of cases of laryngeal and ovarian cancer mortality (3 and 1, respectively) which may make estimates un- stable. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The analysis was sufficiently described and could be reproduced given original data. Metric 15 Statistical models Medium x 0.2 0.4 The calculation of sex-specific SMRs was detailed and transparent. No concerns. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 2.0 Extracted Yes Continued on next page . .. 40 of 65 ------- . continued from previous page Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian_asbestos_product_workers_ovarian_cancer-Cancer HERO ID: 6912534 Domain Metric Rating ^ MWF* Score Comments^ * MWF = Metric Weighting Factor t High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. + The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. if any metric is Unacceptable Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 41 Of 65 ------- Table 14: Mamo and Costa 2004: Evaluation of Cancer for Female Lung Cancer Mortality Outcomes Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian asbestos product workers lung cancer female-Cancer HERO ID: 6912534 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection High x 0.4 0.4 Subjects consisted of former employees of an asbestos-based products manufacturing plant in Grugliasco, Italy, hired before 1971. Subjects were followed from 1981 through 1995. Metric 2: Attrition Medium x 0.4 0.8 The initial study sample consisted of 1653 employ- ees. The study authors note that the analysis sample consisted of 967 individuals, selected as 18-74 y of age, with concern about dropping older individuals. Metric 3: Comparison Group Medium x 0.2 0.4 The study authors attempted to find a similar popu- lation to the asbestos-products employees by using a reference population consisting of residents from the same area (Turin) whom were manual workers—this was an attempt to mitigate healthy worker effect. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Low x 0.5 1.5 Exposure was determined by occupation only. Ba- sic descriptions of the asbestos product facility were provided. Metric 5: Exposure levels Not Rated NA NA Metric 6: Temporality Medium x 0.5 1 Subjects employed in 1971 were followed-up from 1981 to 1995 which is a sufficiently long follow-up period to detect an effect. It's not clear whether individuals were followed between 1971 and 1981. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 Causes of death were obtained from the Italian Na- tional Registry of Deaths (ISTAT) and consulting death certificates at the Piedmont Cancer Registry. It was not reported whether cancer cases were his- tologically confirmed, and it appears that causes of death were coded using ICD-9 . Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in results. SMRs were provided with the number of cases and 95% confi- dence intervals. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Medium x 0.5 1 Individual smoking behaviors were not able to be controlled for. Continued on next page . 42 of 65 ------- ... continued from previous page Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian_asbestos_product_workers_lung_cancer_female-Cancer HERO ID: 6912534 Domain Metric Rating^ MWF* Score Comments^ Metric 10 Covariate Characterization Medium x 0.25 0.5 Covariates were largely not collected except sex, which was presumably taken from employment records. Metric 11 Co-exposure Confounding Medium x 0.25 0.5 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 SMRs were calculated to estimate the effect of oc- cupational asbestos exposure on cancer mortality rates. This is an appropriate study design for the study question. Metric 13 Statistical power Medium x 0.2 0.4 A total of 967 individuals were included in the anal- ysis sample, which is sufficiently large to detect an effect. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The analysis was sufficiently described and could be reproduced given original data. Metric 15 Statistical models Medium x 0.2 0.4 The calculation of sex-specific SMRs was detailed and transparent. No concerns. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 2.0 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 43 of 65 ------- Table 15: Mamo and Costa 2004: Evaluation of Cancer for Male and Female Lung Cancer Mortality Outcomes Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian asbestos product workers lung cancer combined-Cancer HERO ID: 6912534 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection High x 0.4 0.4 Subjects consisted of former employees of an asbestos-based products manufacturing plant in Grugliasco, Italy, hired before 1971. Subjects were followed from 1981 through 1995. Metric 2: Attrition Medium x 0.4 0.8 The initial study sample consisted of 1653 employ- ees. The study authors note that the analysis sample consisted of 967 individuals, selected as 18-74 y of age, with concern about dropping older individuals. Metric 3: Comparison Group Medium x 0.2 0.4 The study authors attempted to find a similar popu- lation to the asbestos-products employees by using a reference population consisting of residents from the same area (Turin) whom were manual workers—this was an attempt to mitigate healthy worker effect. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Low x 0.5 1.5 Exposure was determined by occupation only. Ba- sic descriptions of the asbestos product facility were provided. Metric 5: Exposure levels Not Rated NA NA Metric 6: Temporality Medium x 0.5 1 Subjects employed in 1971 were followed-up from 1981 to 1995 which is a sufficiently long follow-up period to detect an effect. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 Causes of death were obtained from the Italian Na- tional Registry of Deaths (ISTAT) and consulting death certificates at the Piedmont Cancer Registry. It was not reported whether cancer cases were his- tologically confirmed, and it appears that causes of death were coded using ICD-9 . Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in results. SMRs were provided with the number of cases and 95% confi- dence intervals. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Medium x 0.5 1 Individual smoking behaviors were not able to be controlled for. Continued on next page . 44 of 65 ------- ... continued from previous page Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian_asbestos_product_workers_lung_cancer_combined-Cancer HERO ID: 6912534 Domain Metric Rating^ MWF* Score Comments^ Metric 10 Covariate Characterization Medium x 0.25 0.5 Covariates were largely not collected except sex, which was presumably taken from employment records. Metric 11 Co-exposure Confounding Medium x 0.25 0.5 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 SMRs were calculated to estimate the effect of oc- cupational asbestos exposure on cancer mortality rates. This is an appropriate study design for the study question. Metric 13 Statistical power Medium x 0.2 0.4 A total of 967 individuals were included in the anal- ysis sample, which is sufficiently large to detect an effect. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The analysis was sufficiently described and could be reproduced given original data. Metric 15 Statistical models Medium x 0.2 0.4 The calculation of sex-specific SMRs was detailed and transparent. No concerns. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 2.0 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 45 of 65 ------- Table 16: Mamo and Costa 2004: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian asbestos product workers laryngeal cancer-Cancer HERO ID: 6912534 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection High x 0.4 0.4 Subjects consisted of former employees of an asbestos-based products manufacturing plant in Grugliasco, Italy, hired before 1971. Subjects were followed from 1981 through 1995. Metric 2: Attrition Medium x 0.4 0.8 The initial study sample consisted of 1653 employ- ees. The study authors note that the analysis sample consisted of 967 individuals, selected as 18-74 y of age, with concern about dropping older individuals. Metric 3: Comparison Group Medium x 0.2 0.4 The study authors attempted to find a similar popu- lation to the asbestos-products employees by using a reference population consisting of residents from the same area (Turin) whom were manual workers—this was an attempt to mitigate healthy worker effect. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Low x 0.5 1.5 Exposure was determined by occupation only. Ba- sic descriptions of the asbestos product facility were provided. Metric 5: Exposure levels Not Rated NA NA Metric 6: Temporality Medium x 0.5 1 Subjects employed in 1971 were followed-up from 1981 to 1995 which is a sufficiently long follow-up period to detect an effect. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 Causes of death were obtained from the Italian Na- tional Registry of Deaths (ISTAT) and consulting death certificates at the Piedmont Cancer Registry. It was not reported whether cancer cases were his- tologically confirmed, and it appears that causes of death were coded using ICD-9 . Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in results. SMRs were provided with the number of cases and 95% confi- dence intervals. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Medium x 0.5 1 Individual smoking behaviors were not able to be controlled for. Continued on next page . 46 of 65 ------- . continued from previous page Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian asbestos product workers laryngeal cancer-Cancer HERO ID: 6912534 Domain Metric Rating^ MWF* Score Comments^ Metric 10 Covariate Characterization Medium x 0.25 0.5 Covariates were largely not collected except sex, which was presumably taken from employment records. Metric 11 Co-exposure Confounding Medium x 0.25 0.5 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 SMRs were calculated to estimate the effect of oc- cupational asbestos exposure on cancer mortality rates. This is an appropriate study design for the study question. Metric 13 Statistical power Medium x 0.2 0.4 A total of 967 individuals were included in the anal- ysis sample, which is sufficiently large to detect an effect. There is some concern for the limited number of cases of laryngeal and ovarian cancer mortality (3 and 1, respectively) which may make estimates un- stable. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The analysis was sufficiently described and could be reproduced given original data. Metric 15 Statistical models Medium x 0.2 0.4 The calculation of sex-specific SMRs was detailed and transparent. No concerns. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 2.0 Extracted Yes Continued on next page . .. 47 of 65 ------- . continued from previous page Study Citation: Mamo, C., Costa, G. (2004). Mortality experience in an historical cohort of chrysotile asbestos textile workers Global asbestos congress 2004 in Tokyo, Data Type: Italian_asbestos_product_workers_laryngeal_cancer-Cancer HERO ID: 6912534 Domain Metric Rating ^ MWF* Score Comments^ * MWF = Metric Weighting Factor t High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. + The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. 4 if any metric is Unacceptable Overall rating = J]. (Metric Score; x MWF;) / J] . MWFj (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 48 of 65 ------- Table 17: Germani et al. 1999: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian women asbestosis ovarian cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 Study participants were drawn from the Italian Na- tional Institute for Insurance of Occupational Acci- dents (INAIL). Eligibility criteria were not expressly stated beyond membership in the asbestosis com- pensation cohort and being alive at the study's onset (1979). Most women entered the asbestosis compen- sation cohort around 50 years of age; some healthy worker effect may be present. High X 0.4 0.4 The number of individuals lost to follow-up was pro- vided and appeared minimal for all groups (total cohort = 0.6 percent). The study authors do not suggest any other missing data. Low X 0.2 0.6 Rates of mortality in the asbestosis compensation cohort were compared with Italian national mortal- ity rates, stratifying for gender-, age-, and calendar period-specific rates. This represents consideration of potential covariates. The study group of those compensated for asbestosis may have been different from national comparison group. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated High x 0.5 NA x 0.5 1.5 NA 0.5 Exposure was assigned only by the women's occupa- tion. No further characterization of the occupational exposure was provided. Asbestosis compensation cohort members were eligi- ble if they were alive and enrolled at the end of 1979. Women were followed until October 1997, allowing for a sufficient follow-up period. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 The cause of death was obtained for deceased sub- jects from the Registry Office of the municipality of residence or death. Causes of death were coded ac- cording to ICD-9 codes, but no details were provided about histological confirmation. Continued on next page 49 of 65 ------- . continued from previous page Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian women asbestosis ovarian cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. The re- sults provide the number of observations, the SMR, and a confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Medium x 0.5 1 SMRs were stratified for gender, age, and calendar period. Individual smoking data was not available and may be of concern for outcomes including lung cancer. Medium X 0.25 0.5 Covariates were drawn from information collected by INAIL for asbestosis compensation. It is not entirely clear how INAIL may have collected the covariates. Low X 0.25 0.75 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Metric 14: Reproducibility of analyses Metric 15: Statistical models Medium X 0.4 0.8 SMRs were used to achieve the study objective of comparing cause-specific mortality rates of an occu- pational cohort to the Italian national population (stratifying for relevant characteristics). Medium x 0.2 0.4 A total of 631 women compensated for asbestosis were included in the overall cohort, with 276 and 278 women in the textile and asbestos cement industry, respectively. Medium X 0.2 0.4 The analysis is well described and transparent. Given the same data, the analysis could be easily reproduced. Medium X 0.2 0.4 SMRs and 95% confidence intervals were provided, and the method to calculate the SMRs was suffi- ciently explained. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Continued on next page 50 of 65 ------- . continued from previous page Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian_women_asbestosis_ovarian_cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Overall Quality Determination1" Medium 1.9 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 51 Of 65 ------- Table 18: Germani et al. 1999: Evaluation of Cancer for Lung Cancer Mortality Outcomes Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian women asbestosis lung cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 Study participants were drawn from the Italian Na- tional Institute for Insurance of Occupational Acci- dents (INAIL). Eligibility criteria were not expressly stated beyond membership in the asbestosis com- pensation cohort and being alive at the study's onset (1979). Most women entered the asbestosis compen- sation cohort around 50 years of age; some healthy worker effect may be present. High X 0.4 0.4 The number of individuals lost to follow-up was pro- vided and appeared minimal for all groups (total cohort = 0.6 percent). The study authors do not suggest any other missing data. Low X 0.2 0.6 Rates of mortality in the asbestosis compensation cohort were compared with Italian national mortal- ity rates, stratifying for gender-, age-, and calendar period-specific rates. This represents consideration of potential covariates. The study group of those compensated for asbestosis may have been different from national comparison group. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated High x 0.5 NA x 0.5 1.5 NA 0.5 Exposure was assigned only by the women's occupa- tion. No further characterization of the occupational exposure was provided. Asbestosis compensation cohort members were eligi- ble if they were alive and enrolled at the end of 1979. Women were followed until October 1997, allowing for a sufficient follow-up period. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 The cause of death was obtained for deceased sub- jects from the Registry Office of the municipality of residence or death. Causes of death were coded ac- cording to ICD-9 codes, but no details were provided about histological confirmation. Continued on next page 52 of 65 ------- . continued from previous page Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian women asbestosis lung cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. The re- sults provide the number of observations, the SMR, and a confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Medium x 0.5 1 SMRs were stratified for gender, age, and calendar period. Individual smoking data was not available and may be of concern for outcomes including lung cancer. Medium X 0.25 0.5 Covariates were drawn from information collected by INAIL for asbestosis compensation. It is not entirely clear how INAIL may have collected the covariates. Low X 0.25 0.75 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Metric 14: Reproducibility of analyses Metric 15: Statistical models Medium X 0.4 0.8 SMRs were used to achieve the study objective of comparing cause-specific mortality rates of an occu- pational cohort to the Italian national population (stratifying for relevant characteristics). Medium x 0.2 0.4 A total of 631 women compensated for asbestosis were included in the overall cohort, with 276 and 278 women in the textile and asbestos cement industry, respectively. Medium X 0.2 0.4 The analysis is well described and transparent. Given the same data, the analysis could be easily reproduced. Medium X 0.2 0.4 SMRs and 95% confidence intervals were provided, and the method to calculate the SMRs was suffi- ciently explained. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Continued on next page 53 of 65 ------- . continued from previous page Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian_women_asbestosis_lung_cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Overall Quality Determination1" Medium 1.9 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 54 Of 65 ------- Table 19: Germani et al. 1999: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian women asbestosis larynx cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group Medium X 0.4 0.8 Study participants were drawn from the Italian Na- tional Institute for Insurance of Occupational Acci- dents (INAIL). Eligibility criteria were not expressly stated beyond membership in the asbestosis com- pensation cohort and being alive at the study's onset (1979). Most women entered the asbestosis compen- sation cohort around 50 years of age; some healthy worker effect may be present. High X 0.4 0.4 The number of individuals lost to follow-up was pro- vided and appeared minimal for all groups (total cohort = 0.6 percent). The study authors do not suggest any other missing data. Low X 0.2 0.6 Rates of mortality in the asbestosis compensation cohort were compared with Italian national mortal- ity rates, stratifying for gender-, age-, and calendar period-specific rates. This represents consideration of potential covariates. The study group of those compensated for asbestosis may have been different from national comparison group. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated High x 0.5 NA x 0.5 1.5 NA 0.5 Exposure was assigned only by the women's occupa- tion. No further characterization of the occupational exposure was provided. Asbestosis compensation cohort members were eligi- ble if they were alive and enrolled at the end of 1979. Women were followed until October 1997, allowing for a sufficient follow-up period. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization Medium x 0.667 1.33 The cause of death was obtained for deceased sub- jects from the Registry Office of the municipality of residence or death. Causes of death were coded ac- cording to ICD-9 codes, but no details were provided about histological confirmation. Continued on next page 55 of 65 ------- . continued from previous page Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian women asbestosis larynx cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Metric 8: Reporting Bias High x 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. The re- sults provide the number of observations, the SMR, and a confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Metric 11: Co-exposure Confounding Medium x 0.5 1 SMRs were stratified for gender, age, and calendar period. Individual smoking data was not available and may be of concern for outcomes including lung cancer. Medium X 0.25 0.5 Covariates were drawn from information collected by INAIL for asbestosis compensation. It is not entirely clear how INAIL may have collected the covariates. Low X 0.25 0.75 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12: Study Design and Methods Metric 13: Statistical power Metric 14: Reproducibility of analyses Metric 15: Statistical models Medium X 0.4 0.8 SMRs were used to achieve the study objective of comparing cause-specific mortality rates of an occu- pational cohort to the Italian national population (stratifying for relevant characteristics). Medium x 0.2 0.4 A total of 631 women compensated for asbestosis were included in the overall cohort, with 276 and 278 women in the textile and asbestos cement industry, respectively. Medium X 0.2 0.4 The analysis is well described and transparent. Given the same data, the analysis could be easily reproduced. Medium X 0.2 0.4 SMRs and 95% confidence intervals were provided, and the method to calculate the SMRs was suffi- ciently explained. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Continued on next page 56 of 65 ------- . continued from previous page Study Citation: Germani, D., Belli, S., Bruno, C., Grignoli, M., Nesti, M., Pirastu, R., Comba, P. (1999). Cohort mortality study of women compensated for asbestosis in Italy American Journal of Industrial Medicine, 36(1,1), 129-134 Data Type: Italian_women_asbestosis_larynx_cancer-Cancer HERO ID: 709605 Domain Metric Rating^ MWF* Score Comments^ Overall Quality Determination1" Medium 1.9 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 57 of 65 ------- Table 20: Tarchi et al. 1994: Evaluation of Cancer for Ovarian Cancer Mortality Outcomes Cohort mortality study of rock salt Study Citation: Tarchi, M., Orsi, D., Comba, P., De Santis, M., Pirastu, R., Battista, G., Valiani, M. (1994). workers in Italy American Journal of Industrial Medicine, 25(2,2), 251-256 Data Type: rock salt workers ovarian cancer mortality-Cancer HERO ID: 2739094 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group High X 0.4 0.4 The setting and activities conducted in the rock salt mine were described. All employees of the rock salt mine for any duration of length between 1965 and the end of 1989 were eligible (n=487). High X 0.4 0.4 Vital status was determined for all cohort members, indicating no loss to follow-up. The cause of death could not be determined for two individuals, which is not expected to appreciably bias the results. Medium x 0.2 0.4 Rates of mortality were compared for employees in a rock salt mine compared to those of the Tuscany region (where the mine is located). It was not clear whether age was considered for stratification when calculating SMRs. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Low x 0.5 NA x 0.5 1.5 NA 1.5 Exposure was based on employment at the rock salt mine only. Participants were included if they were employed be- tween 1965 and 1989, and followed only until 1989. Those becoming eligible later during the eligibility period may have had less time for onset of disease. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization High Metric 8: Reporting Bias High X 0.667 0.67 Vital status and cause of death were determined at the registrar's office of the individual subject's place of residence or death using ICD-9 codes. The study authors note that classification was carried out in- dependently by two physicians. X 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. Effect estimates were provided as SMRs, observed cases, expected cases, and a 90% confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Medium Medium x 0.5 x 0.25 0.5 Sex-specific SMRs were provided, however, individ- ual smoking rates did not appear to be considered. Covariates were presumably taken from employment records. Continued on next page 58 of 65 ------- ... continued from previous page Study Citation: Tarchi, M., Orsi, D., Comba, P., De Santis, M., Pirastu, R., Battista, G. workers in Italy American Journal of Industrial Medicine, 25(2,2), 251-256 Data Type: rock_salt_workers_ovarian_cancer_mortality-Cancer HERO ID: 2739094 Valiani, M. (1994). Cohort mortality study of rock salt Domain Metric Rating^ MWF* Score Comments^ Metric 11 Co-exposure Confounding Low x 0.25 0.75 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 Standardized mortality ratios were used to deter- mine excess risk of various cause-specific mortalities related to cancer. This is an appropriate study de- sign for the study question. Metric 13 Statistical power Medium x 0.2 0.4 A total of 487 individuals (367 men, 120 women) were in the analysis sample. There is some concern due to the low number of cases of ovarian and laryn- geal cancers, making the estimates unstable. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The description of the analysis was sufficient that, given original data, the analysis could be repro- duced. Metric 15 Statistical models Medium x 0.2 0.4 The calculation of SMRs was transparent and model assumptions appear to be met. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium —> Low§ Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study § Evaluator's explanation for rating change: "Low number of mortality cases for ovarian cancer." 59 of 65 ------- Table 21: Tarchi et al. 1994: Evaluation of Cancer for Male and Female Lung Cancer Mortality Outcomes Cohort mortality study of rock salt Study Citation: Tarchi, M., Orsi, D., Comba, P., De Santis, M., Pirastu, R., Battista, G., Valiani, M. (1994). workers in Italy American Journal of Industrial Medicine, 25(2,2), 251-256 Data Type: rock salt workers lung cancer mortality-Cancer HERO ID: 2739094 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group High X 0.4 0.4 The setting and activities conducted in the rock salt mine were described. All employees of the rock salt mine for any duration of length between 1965 and the end of 1989 were eligible (n=487). High X 0.4 0.4 Vital status was determined for all cohort members, indicating no loss to follow-up. The cause of death could not be determined for two individuals, which is not expected to appreciably bias the results. Medium x 0.2 0.4 Rates of mortality were compared for employees in a rock salt mine compared to those of the Tuscany region (where the mine is located). It was not clear whether age was considered for stratification when calculating SMRs. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Low x 0.5 NA x 0.5 1.5 NA 1.5 Exposure was based on employment at the rock salt mine only. Participants were included if they were employed be- tween 1965 and 1989, and followed only until 1989. Those becoming eligible later during the eligibility period may have had less time for onset of disease. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization High Metric 8: Reporting Bias High X 0.667 0.67 Vital status and cause of death were determined at the registrar's office of the individual subject's place of residence or death using ICD-9 codes. The study authors note that classification was carried out in- dependently by two physicians. X 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. Effect estimates were provided as SMRs, observed cases, expected cases, and a 90% confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Medium Medium x 0.5 x 0.25 0.5 Sex-specific SMRs were provided, however, individ- ual smoking rates did not appear to be considered. Covariates were presumably taken from employment records. Continued on next page 60 of 65 ------- .. . continued from previous page Study Citation: Tarchi, M., Orsi, D., Comba, P., De Santis, M., Pirastu, R., Battista, G., Valiani, M. (1994). Cohort mortality study of rock salt workers in Italy American Journal of Industrial Medicine, 25(2,2), 251-256 Data Type: rock salt workers lung cancer mortality-Cancer HERO ID: 2739094 Domain Metric Rating^ MWF* Score Comments^ Metric 11 Co-exposure Confounding Low x 0.25 0.75 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 Standardized mortality ratios were used to deter- mine excess risk of various cause-specific mortalities related to cancer. This is an appropriate study de- sign for the study question. Metric 13 Statistical power Medium x 0.2 0.4 A total of 487 individuals (367 men, 120 women) were in the analysis sample. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The description of the analysis was sufficient that, given original data, the analysis could be repro- duced. Metric 15 Statistical models Medium x 0.2 0.4 The calculation of SMRs was transparent and model assumptions appear to be met. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 1.9 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; X MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 61 Of 65 ------- Table 22: Tarchi et al. 1994: Evaluation of Cancer for Female Lung Cancer Mortality Outcomes Study Citation: Tarchi, M., Orsi, D., Comba, P., De Santis, M., Pirastu, R., Battista, G. workers in Italy American Journal of Industrial Medicine, 25(2,2), 251-256 Data Type: rock salt workers female lung cancer mortality-Cancer HERO ID: 2739094 Valiani, M. (1994). Cohort mortality study of rock salt Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group High X 0.4 0.4 The setting and activities conducted in the rock salt mine were described. All employees of the rock salt mine for any duration of length between 1965 and the end of 1989 were eligible (n=487). High X 0.4 0.4 Vital status was determined for all cohort members, indicating no loss to follow-up. The cause of death could not be determined for two individuals, which is not expected to appreciably bias the results. Medium x 0.2 0.4 Rates of mortality were compared for employees in a rock salt mine compared to those of the Tuscany region (where the mine is located). It was not clear whether age was considered for stratification when calculating SMRs. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Low x 0.5 NA x 0.5 1.5 NA 1.5 Exposure was based on employment at the rock salt mine only. Participants were included if they were employed be- tween 1965 and 1989, and followed only until 1989. Those becoming eligible later during the eligibility period may have had less time for onset of disease. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization High Metric 8: Reporting Bias High X 0.667 0.67 Vital status and cause of death were determined at the registrar's office of the individual subject's place of residence or death using ICD-9 codes. The study authors note that classification was carried out in- dependently by two physicians. X 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. Effect estimates were provided as SMRs, observed cases, expected cases, and a 90% confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Medium Medium x 0.5 x 0.25 0.5 Sex-specific SMRs were provided, however, individ- ual smoking rates did not appear to be considered. Covariates were presumably taken from employment records. Continued on next page 62 of 65 ------- .. . continued from previous page Study Citation: Tarchi, M., Orsi, D., Comba, P., De Santis, M., Pirastu, R., Battista, G., Valiani, M. (1994). Cohort mortality study of rock salt workers in Italy American Journal of Industrial Medicine, 25(2,2), 251-256 Data Type: rock salt workers female lung cancer mortality-Cancer HERO ID: 2739094 Domain Metric Rating^ MWF* Score Comments^ Metric 11 Co-exposure Confounding Low x 0.25 0.75 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 Standardized mortality ratios were used to deter- mine excess risk of various cause-specific mortalities related to cancer. This is an appropriate study de- sign for the study question. Metric 13 Statistical power Medium x 0.2 0.4 A total of 487 individuals (367 men, 120 women) were in the analysis sample. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The description of the analysis was sufficient that, given original data, the analysis could be repro- duced. Metric 15 Statistical models Medium x 0.2 0.4 The calculation of SMRs was transparent and model assumptions appear to be met. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium 1.9 Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; X MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study 63 of 65 ------- Table 23: Tarchi et al. 1994: Evaluation of Cancer for Laryngeal Cancer Mortality Outcomes Cohort mortality study of rock salt Study Citation: Tarchi, M., Orsi, D., Comba, P., De Santis, M., Pirastu, R., Battista, G., Valiani, M. (1994). workers in Italy American Journal of Industrial Medicine, 25(2,2), 251-256 Data Type: rock salt workers larynx cancer mortality-Cancer HERO ID: 2739094 Domain Metric Rating^ MWF* Score Comments^ Domain 1: Study Participation Metric 1: Participant selection Metric 2: Attrition Metric 3: Comparison Group High X 0.4 0.4 The setting and activities conducted in the rock salt mine were described. All employees of the rock salt mine for any duration of length between 1965 and the end of 1989 were eligible (n=487). High X 0.4 0.4 Vital status was determined for all cohort members, indicating no loss to follow-up. The cause of death could not be determined for two individuals, which is not expected to appreciably bias the results. Medium x 0.2 0.4 Rates of mortality were compared for employees in a rock salt mine compared to those of the Tuscany region (where the mine is located). It was not clear whether age was considered for stratification when calculating SMRs. Domain 2: Exposure Characterization Metric 4: Measurement of Exposure Metric 5: Exposure levels Metric 6: Temporality Low Not Rated Low x 0.5 NA x 0.5 1.5 NA 1.5 Exposure was based on employment at the rock salt mine only. Participants were included if they were employed be- tween 1965 and 1989, and followed only until 1989. Those becoming eligible later during the eligibility period may have had less time for onset of disease. Domain 3: Outcome Assessment Metric 7: Outcome measurement or characterization High Metric 8: Reporting Bias High X 0.667 0.67 Vital status and cause of death were determined at the registrar's office of the individual subject's place of residence or death using ICD-9 codes. The study authors note that classification was carried out in- dependently by two physicians. X 0.333 0.33 Outcomes specified in the abstract, introduction, and methods were provided in the results. Effect estimates were provided as SMRs, observed cases, expected cases, and a 90% confidence interval. Domain 4: Potential Confounding/Variable Control Metric 9: Covariate Adjustment Metric 10: Covariate Characterization Medium Medium x 0.5 x 0.25 0.5 Sex-specific SMRs were provided, however, individ- ual smoking rates did not appear to be considered. Covariates were presumably taken from employment records. Continued on next page 64 of 65 ------- ... continued from previous page Study Citation: Tarchi, M., Orsi, D., Comba, P., De Santis, M., Pirastu, R., Battista, G. workers in Italy American Journal of Industrial Medicine, 25(2,2), 251-256 Data Type: rock_salt_workers_larynx_cancer_mortality-Cancer HERO ID: 2739094 Valiani, M. (1994). Cohort mortality study of rock salt Domain Metric Rating^ MWF* Score Comments^ Metric 11 Co-exposure Confounding Low x 0.25 0.75 This is an occupational cohort and co-exposures were not addressed. Co-exposures may be present depending on specific job titles and positions. Domain 5: Analysis Metric 12 Study Design and Methods Medium x 0.4 0.8 Standardized mortality ratios were used to deter- mine excess risk of various cause-specific mortalities related to cancer. This is an appropriate study de- sign for the study question. Metric 13 Statistical power Medium x 0.2 0.4 A total of 487 individuals (367 men, 120 women) were in the analysis sample. There is some concern due to the low number of cases of ovarian and laryn- geal cancers, making the estimates unstable. Metric 14 Reproducibility of analyses Medium x 0.2 0.4 The description of the analysis was sufficient that, given original data, the analysis could be repro- duced. Metric 15 Statistical models Medium x 0.2 0.4 The calculation of SMRs was transparent and model assumptions appear to be met. Domain 6: Other Considerations for Biomarker Selection and Measurement Metric 16 Use of Biomarker of Exposure NA NA Metric 17 Effect biomarker NA NA Metric 18 Method Sensitivity NA NA Metric 19 Biomarker stability NA NA Metric 20 Sample contamination NA NA Metric 21 Method requirements NA NA Metric 22 Matrix adjustment NA NA Overall Quality Determination1" Medium —> Low§ Extracted Yes MWF = Metric Weighting Factor High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value. The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High. Overall rating = ]T\ (Metric Score; x MWF;) / J] . MWFj if any metric is Unacceptable (round to the nearest tenth) otherwise where High => 1 to < 1.7; Medium => 1.7 to < 2.3; Low => 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is crossed out and an arrow points to the new rating. ^ This metric met the criteria for high confidence as expected for this type of study § Evaluator's explanation for rating change: "Low number of mortality cases for laryngeal cancer." 65 of 65 ------- |