br We stratified mortality outcomes by gender occupational
We stratified mortality outcomes by gender, occupational history, and cumulative ambient asbestos concentration strata to examine the differences. Environmental Research 175 (2019) 449–456
Counts of underlying cause of death of follow-up cohort members, 2001–2011.
Cause of death ICD-10 codes N
ICD, International Classification of Diseases. NMRD, nonmalignant respiratory disease.
3.1. Descriptive statistics
Demographics of the cohort participants known to be alive at the start of 1988 are provided in Table 1. Death records showed that 847 were deceased as of December 31, 2011. The study cohort was evenly split between males (50%) and females (50%), 57% reported they were current or former (ever) smokers and 35% (n = 2043) were born prior to 1950. There were 101 former WM/WRG workers in the cohort and another 245 (4%) who reported living with a worker. Although most reported no direct contact with the vermiculite waste, a history of moving the waste was reported by 255 (4%) and 640 (11%) said they had played on the waste piles as children.
The geometric mean (GM) of cumulative ambient asbestos fiber exposure for the entire cohort was 0.0005 fibers/cc x months (Table 1). As expected, ambient exposure was highest in older individuals, parti-cularly for persons born prior to 1930 for whom the geometric mean exposure was 0.02 fibers/cc x months and greater; this is likely due to longer duration of residence in the study area during peak operations of the plant when emissions were highest. WM/WRG workers and household members of workers, and people who reported that they had moved or used waste or played on waste piles as children also had higher cumulative ambient exposure.
3.2. Cancer incidence
Over the PD98059 1988–2010, we observed a total of 636 incident cancers in the cohort, with 343 observed in males and 293 in females. There were seven mesotheliomas, three in males and four in females, and 118 lung cancers, 66 in males and 52 in females (Table 2). Com-pared with the overall Minnesota population, the PIR for mesothelioma
in females (PIR = 11.76; 95% CI: 3.17, 30.12) was significantly ele-vated, but no significant elevation was observed in males or both genders combined. This elevation in females held true even after the PIR was adjusted for the overall cancer deficit in the cohort (PIR = 11.11; 95% CI: 2.99, 28.45). We found a statistically significant increased PIR for lung cancer for both males and females, even after adjusting for the overall cancer deficit.
Study participants who reported a history of working in occupations with possible asbestos exposure, a category which included WM/WRG workers, were compared to those with no known source of occupational asbestos exposure (Fig. 1).
Lung cancer and all respiratory cancer were elevated in both groups. All-cancers combined were elevated only in the group with no known
Proportional incidence ratios and adjusted PIR for mesothelioma and lung cancer, 1988–2010.
Observed Expected PIR
PIR, Proportional incidence ratios.
* Statistically significant.
a PIR was adjusted for low overall cancer using adjustment method from Wong et al.
Cause of death ICD-10 codes Males
ICD, International Classification of Diseases. SMR, Standard mortality ratios. COPD, chronic obstructive pulmonary disease.
occupational asbestos exposure. When occupational histories were ex-amined by gender, two mesothelioma deaths occurred among females with no history of occupational asbestos exposure (Table 5).
Stratified analyses of the SMRs did not demonstrate increased risk with increasing log cumulative ambient exposure for any of the out-comes. However, all four of the mesothelioma deaths were found in the > 75th percentile, the highest cumulative exposure category (Table 6).
To our knowledge, hypothesis is the only community-based cohort study to document excess cancer incidence and mortality associated with non-occupational, low dose exposure to Libby asbestos from vermiculite processing operations outside of Libby, Montana. We found
mesothelioma incidence to be significantly elevated in females and lung cancer incidence was significantly elevated in both genders, even after adjusting for an overall cancer deficit in the area. Deaths due to COPD, mesothelioma, and lung cancer were significantly elevated in only fe-males, but deaths from all respiratory cancers combined were elevated in both genders. The results of this analysis are consistent with the findings of Horton et al. (2008) who reviewed cancer registry data and mortality data from 70 vermiculite processing sites in 23 states (Minnesota not included). Of the 53 sites, where cancer incidence was analyzed, seven had a sig-nificant excess of mesothelioma and investigators found a mesothe-lioma excess in females at three sites (Illinois, Jersey City, NJ and Jefferson Parish, LA). At these sites, the authors suggest mesothelioma in women may be due to take-home exposure among family members with occupational exposure. However, our analysis and case review