Environmental Research 130 (2014) 43–50
Contents lists available at ScienceDirect
Environmental Research journal homepage: www.elsevier.com/locate/envres
Leukemia, lymphoma and multiple myeloma mortality (1950–1999) and incidence (1969–1999) in the Eldorado uranium workers cohort Lydia B. Zablotska a,n, Rachel S.D. Lane b, Stanley E. Frost c, Patsy A. Thompson b a
Department of Epidemiology and Biostatistics, School of Medicine, University of California, San Francisco, CA 94118, USA Radiation and Health Sciences Division, Directorate of Environmental and Radiation Protection and Assessment, Canadian Nuclear Safety Commission, Ottawa, ON, Canada K1P 5S9 c Frost & Frost Consultants, Saskatoon, SK, Canada S7H 0A1 b
art ic l e i nf o
a b s t r a c t
Article history: Received 18 July 2013 Received in revised form 3 December 2013 Accepted 9 January 2014 Available online 28 February 2014
Uranium workers are chronically exposed to low levels of radon decay products (RDP) and gamma (γ) radiation. Risks of leukemia from acute and high doses of γ-radiation are well-characterized, but risks from lower doses and dose-rates and from RDP exposures are controversial. Few studies have evaluated risks of other hematologic cancers in uranium workers. The purpose of this study was to analyze radiation-related risks of hematologic cancers in the cohort of Eldorado uranium miners and processors first employed in 1932–1980 in relation to cumulative RDP exposures and γ-ray doses. The average cumulative RDP exposure was 100.2 working level months and the average cumulative whole-body γ-radiation dose was 52.2 millisievert. We identified 101 deaths and 160 cases of hematologic cancers in the cohort. Overall, male workers had lower mortality and cancer incidence rates for all outcomes compared with the general Canadian male population, a likely healthy worker effect. No statistically significant association between RDP exposure or γ-ray doses, or a combination of both, and mortality or incidence of any hematologic cancer was found. We observed consistent but non-statistically significant increases in risks of chronic lymphocytic leukemia (CLL) and Hodgkin lymphoma (HL) incidence and non-Hodgkin lymphoma (NHL) mortality with increasing γ-ray doses. These findings are consistent with recent studies of increased risks of CLL and NHL incidence after γradiation exposure. Further research is necessary to understand risks of other hematologic cancers from low-dose exposures to γ-radiation. & 2014 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
Keywords: Hematologic cancers Uranium miners Gamma radiation Leukemia Chronic lymphocytic leukemia Lymphoma
1. Introduction Significantly increased risks of leukemia associated with exposure to high acute doses of ionizing radiation are well-studied and recognized (UNSCEAR, 2008). The evidence of radiation-associated risks of other hematologic cancers, including Hodgkin lymphoma (HL), multiple myeloma (MM) and non-Hodgkin lymphoma (NHL), is mixed (Boice, 1992; Ron, 1998), with some studies showing significantly increased risks of NHL in some subgroups (Hsu et al., 2013; Richardson et al., 2009), and little evidence of increased risks of MM or HL. The risks of protracted low-dose radiation are also not clear. Recent studies of nuclear workers occupationally exposed to low doses of radiation showed increased radiation risks of leukemia, NHL and MM, but little evidence of radiation-associated risks of HL or
n Correspondence to: Department of Epidemiology and Biostatistics, School of Medicine, University of California, San Francisco, 3333 California Street, Suite 280, San Francisco, CA 94118, USA. Fax: þ1 415 563 4602. E-mail address: [emailprotected] (L.B. Zablotska).
chronic lymphocytic leukemia (CLL) (Cardis et al., 2007; Muirhead et al., 2009; Vrijheid et al., 2008). Ionizing radiation was thought to have little evidence of an effect on CLL in the study of atomic bomb (A-bomb) survivors from Hiroshima and Nagasaki, based mainly on the first 40-years of follow-up (Preston et al., 1994). However, CLL is very rare in the Japanese population ( 6%) (Tamura et al., 2001), making risk assessment difficult (Richardson et al., 2005). Recent results of the A-bomb survivors reported a significant increased risk of CLL, albeit based on 12 cases diagnosed over 55 years of follow-up (Hsu et al., 2013). A significantly increased radiation risk of CLL, similar in size to risks of other types of leukemia, was also reported in a recent study of Chornobyl cleanup workers exposed to protracted low-dose external irradiation (Zablotska et al., 2013). Once inhaled, radon decay products (RDP) can be distributed to the red bone marrow and consequently may increase the risk of leukemia (Laurier et al., 2001). Since uranium miners are also exposed to gamma (γ) radiation doses, there is growing interest in understanding the risks of leukemia and other hematologic
http://dx.doi.org/10.1016/j.envres.2014.01.002 0013-9351 & 2014 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
44
L.B. Zablotska et al. / Environmental Research 130 (2014) 43–50
cancers in these workers. Past mortality-based cohort and casecontrol studies of uranium miners and processing workers (Darby et al., 1995; Laurier et al., 2001; Polednak and Frome, 1981), as well as recent ecological studies (Boice et al., 2009, 2007, 2010), provide little evidence of an association between radon and leukemia. However, recent studies based on cancer incidence reported elevated risks of CLL and NHL (attributed to radon, RDP, and γ-rays) (Mohner et al., 2006, 2010; Rericha et al., 2006). The updated Eldorado cohort consists of 17,660 uranium mine, mill and processing workers, first employed in 1932–1980, with relatively recent mortality (1950–1999) and cancer incidence (1969–1999) follow-up (Lane et al., 2010). The purpose of the current analysis is to assess the radiation risks (RDP exposures and γ-ray doses) of hematologic cancers (combined), malignant lymphoma (HL and NHL), leukemia (CLL and non-CLL) and MM among Eldorado workers.
2. Materials and methods A detailed description of the Eldorado cohort, record linkages, outcomes and exposures has been published previously (Lane et al., 2010).
2.1. Cohort The cohort included uranium mine and mill workers employed at two mine sites (Port Radium, Northwest Territories, and Beaverlodge, northern Saskatchewan), and workers employed at the radium and uranium refining and processing plant (Port Hope, Ontario), with a small number of individuals employed at “other sites” including head office, aviation, research and development, and exploration (N ¼ 19,855). The sub-cohort membership was based on the employment site where a worker spent the longest period of time working for Eldorado. Cohort members were first employed between 1932 and 1980, were between the ages of 15 and 75 years, and alive at start of follow-up in 1950 (mortality analysis) or 1969 (cancer incidence analysis).
2.2. Record linkage An internal linkage and data processing led to the exclusion of 2195 (11.0%) records with missing information (i.e., sex, birth year, age out of range, no occupational record, and no exposure data). The final cohort for mortality analysis included 17,660 subjects (88.9% of the original cohort). Information on “fact of death” was used for “death clearance” between 1950 and 1969 (n ¼ 886) and 4 additional subjects were excluded because their RDP exposure occurred after cancer diagnosis. The incidence analysis is therefore based on 16,770 subjects. The cohort was linked to the Historic Summary Tax File (HSTF), the Canadian Mortality Database (CMDB) and the Canadian Cancer Database (CCDB) to ascertain mortality and cancer incidence until the end of 1999. In total, the vital status of 15,580 (78.5%) cohort subjects was ascertained. Individuals (N ¼4307, 21.8%) who could not be linked to the HSTF or the CMDB had their termination date at work as the last date known alive. The HSTF includes identifiers and the minimal amount of data required to ascertain the vital status and location of individuals since 1984. The HSTF was used to enhance the Eldorado cohort linkage by filling in data gaps, to determine the vital status of the cohort members at the end of the follow-up period, and to evaluate the results of the mortality linkage. The CMDB (1950 to present) is based on the vital statistics program at Statistics Canada, that routinely collects demographic and cause of death information from all provincial and territorial vital statistics registries on all deaths in Canada. Some data are also collected on Canadian residents who died in some states of the United States of America. Registration of deaths is a legal requirement through the Vital Statistics Acts (or equivalent legislation) in each Canadian province and territory, so reporting is virtually complete. Under-coverage is thought to be minimal (1% or less) (Goldberg et al., 1993). Cancer incidence records from provincial and territorial cancer registries are sent to Statistics Canada where they are edited, standardized and transformed into a format suitable for record linkage, thereby creating the CCDB (1969 to present). Cancer reporting is virtually complete and of high quality, since it is routinely checked for accuracy through regular assessments by Statistics Canada and the cancer registries. The cancer data are also linked with mortality data to ensure the completeness and correctness of vital status information and to capture missed cancer cases. Both linking procedures optimize the accuracy of the data (Canadian Cancer Society, 2013).
2.3. Outcomes For the mortality and cancer incidence analyses, the underlying causes of death and cancer diagnoses were recoded from the original International Classification of Disease (ICD) code in use at the time of death or diagnosis to the International Classification of Diseases, Ninth Revision (ICD-9) (WHO, 1998) or the International Classification of Diseases for Oncology: Morphology of Neoplasms, Third Edition (ICD-O-3) (Fritz et al., 2000). This included all hematologic cancers combined (ICD-9: 200.0–208.9) and all malignant lymphoma (ICD-9: 200.0–202.9), Hodgkin lymphoma (HL, ICD-9: 201.0–201.9), non-Hodgkin lymphoma (NHL, ICD-9: 200.0– 200.9 & 202.0–202.9), multiple myeloma (MM, ICD-9: 203.0–203.9), all leukemia (ICD-9: 204.0–208.9), CLL (ICD-9: 204.1, ICD-O-3: M9823/3) and non-CLL, which were evaluated separately. 2.4. Exposures The detailed work history file was obtained from Eldorado company records. A detailed description of how RDP exposure was estimated is available (Howe et al., 1986, 1987; Lane et al., 2010). In brief, the annual mean RDP exposure in WLM1 was calculated by summing over the WL measurements available for each type of workplace, the proportion of workers in each occupation and the proportion of time spent in each type of workplace by workers in each occupation. Workplace RDP concentration measurements were supplemented by data on seasonal mine ventilation rates, building air volumes and air exchange rates. Other studies of uranium miners used similar methods to assign personal exposures based on the job-exposure matrix (NRC, 1999). Exposures for workers with additional mining experience in early non-Eldorado Western Canadian mines were estimated based on the Beaverlodge WL data. Any additional RDP exposures (i.e., from Ontario uranium mines) were obtained from the National Dose Registry (NDR2). The current study also has information on individual γ-ray doses for all cohort subjects. Personal γ-ray doses were calculated from the average dose-rates and time on the job and expressed in millisieverts (mSv) for workers who did not wear a personal dosimeter. 2.5. Statistical analysis Each individual contributed person-years at risk from the later of the date of hire or the start date of follow-up, to the exit date. ‘Start date’ was defined as January 1st, 1950, for the mortality analysis and January 1st, 1969, for the incidence analysis. ‘Exit date’ was defined as December 31st, 1999, the date of cancer diagnosis or death, or the last date known alive (defined as date of last employment or contact), whichever occurred earlier. The initial set of analyses was based on external comparisons of the cohort with the general Canadian population. Observed (O) and expected (E) values were used to estimate standardized mortality ratios (SMR) and standardized incidence ratios (SIR) by means of indirect standardization. Expected values were derived from Canadian population mortality (1950–1999) and cancer incidence (1969– 1999) rates, adjusted for sex, age and calendar year at risk. Incidence and mortality for leukemia subtypes were not available for the general Canadian population, so SMRs and SIRs were provided for all leukemia combined. The second set of analyses was based on internal comparisons of the cohort and used grouped Poisson regression (Breslow and Day, 1987; Preston et al., 1993) to estimate risks using a simple linear relative risk (RR) model: Relative risk ¼ 1:0 þ ðβ XÞexpðΣ i γ i zi Þ
ð1Þ
where X represents factors such as RDP exposure or γ-ray dose, zi are potential modifying factors such as age at first γ-ray dose, and β and γi are coefficients, estimated using maximum likelihood techniques. The beta coefficient (β) is referred to as the excess relative risk (ERR) per unit of exposure; by adding 1.0 to the ERR one obtains the relative risk (RR) per 100 WLM for RDP exposure and RR per one sievert (Sv) for γ-ray dose. The summary person-years at risk were crossclassified by age at risk (15–19, 20–24… 85–100 years), calendar year at risk3
1 The concentration of RDP in workplace air was expressed in Working Levels (WL), where 1 WL is the concentration of RDP per liter of air that would result in the ultimate release of 1.3 105 MeV of potential alpha energy. Occupational exposure to RDP is the product of time in the workplace and the concentration of RDP in the workplace air, measured in Working Level Months (WLM), where 1 WLM is equivalent to one working month (170 h) in a concentration of 1 WL. 2 The NDR had no early records from Eldorado or other early Western Canadian mines. For all other non-Eldorado radiation exposures, the cohort was linked to the NDR. Miners’ γ-ray doses only became available in the NDR from 1981 onward. Miners were not included in the IARC15 country study of nuclear workers (Cardis et al., 2007); there is no evidence of a problem in the transfer of mining company records to the NDR. 3 Calendar year at risk for the cancer incidence analysis was (1969–1974,… 1995–1999).
L.B. Zablotska et al. / Environmental Research 130 (2014) 43–50
Table 1 Basic characteristics of the Eldorado uranium workers cohort. Characteristic
Mortality cohort (1950–1999)
Incidence cohort (1969–1999)
Number of subjects total Men (%) Beaverlodge Port Radium Port Hope Other sites Women (%) Meana RDP exposure, WLM (SD, range) Men Beaverlodge Port Radium Port Hope Other sites Women Meanb γ-ray dose, mSv (SD, range) Men Beaverlodge Port Radium Port Hope Other sites Women
17,660 16,236 (91.9) 9,498 3,047 2,652 1,039 1,424 (8.1)
16,770 15,366 (91.6) 9,231 2,687 2,416 1,032 1,404 (8.4)
100.2 (254.4, 0–2,569.0) 84.8 (203.4, 0–2,441.0) 180.1 (349.5, 0–2,569.0) 14.2 (54.1, 0–626.0) 14.9 (85.8, 0–984.2) 4.6 (10.1, 0–67.6)
88.6 (229.1, 0–2,663.0) 60.9 (143.5, 0–1,617.0) 199.2 (362.0, 0–2,663.0) 10.4 (43.0, 0–590.3) 7.9 (56.2, 0–984.2) 3.5 (9.0, 0–67.6)
52.2 (152.4, 0–3,420.0) 25.6 (39.4, 0–393.1) 46.8 (82.2, 0–897.3) 121.5 (306.8, 0–3,420.0) 23.4 (42.3, 0–250.7) 34.4 (77.4, 0–619.8)
41.5 (116.4, 0–2,921.0) 25.2 (39.5, 0–393.1) 43.7 (84.4, 0–1,076.0) 101.7 (257.0, 0–2,921.0) 18.5 (37.0, 0–250.7) 24.9 (66.6, 0–619.8)
Table 2 SMRs and SIRs and 95% CIs for various cancers and causes of death compared to Canadian national incidence (1969–1999) and mortality (1950–1999) rates for males. Cause of death
Abbreviations: mSv, millisieverts; RDP, radon decay product; SD, standard deviation; WLM, working level months. a b
Weighted by person-years and lagged by 5 years. Weighted by person-years and lagged by 5 years.
(1950–54, 1955–59… 1995–1999), employment site (Port Hope, Port Radium, Beaverlodge, “other sites”), total duration of employment ( o 6 months, 6 monthsþ )4, age at first exposure, cumulative exposure, and years since first exposure. This was done separately for RDP exposures and γ-ray doses. In contrast to Lane et al. (2010), both RDP exposures and γ-ray doses were lagged by 5 years to account for the latency period between exposure and cancer incidence and mortality. Finally, an exploratory analysis was performed for both RDP exposures and γ-ray doses, lagged by 10 years, to compare our findings with a previously published study (Richardson et al., 2009). The person-year weighted mean RDP exposures and γ-ray doses in each cross-classified cell were used in the regression analysis. We also estimated relative risks (RR) for γ-ray dose categories and evaluated linear trend tests based on mean doses for dose categories (0–14, 15– 49 and 50–3420 mSv). Regression parameters, 95% confidence intervals (95% CI) and p-values were estimated using the maximum likelihood method in the AMFIT module of the EPICURE software (Preston et al., 1993). Tests of statistical significance were based on the likelihood ratio test and all p-values were two-sided. Because of the form of Eq. (1), the possible values of β were limited by the requirement that the corresponding relative risks were not negative. If the likelihood of a point or bound estimate did not converge, the minimum value for β was given by o0. The number of cancer cases in this paper may differ between the external and internal analyses. In the external analysis, the population incidence rates, used to calculate expected cases, are based on newly diagnosed cancer cases. Thus, an individual can contribute more than one cancer case to the rates. In the internal analysis, the earliest diagnosed case is used so an individual can only contribute one cancer case to the analysis.
3. Results Thirty percent of Eldorado workers (O¼5332, N ¼17,660) died from 1950 to 1999 and 23% (O¼ 2210, N ¼16,770) developed cancer from 1969 to 1999. Table 1 presents the basic characteristics of the cohort. Most workers were male (N ¼16,236, 91.9% of the cohort), so all further analyses were restricted to “all male workers.” The mean RDP exposures and γ-ray doses (weighted by person-years) are presented in Table 1 separately for the mortality and incidence analyses. Mean RDP and γ-ray doses were higher in 4 Total duration of employment was split at 6 months, as risk drops after 6 months but then remains constant, as previously observed (Howe et al., 1988).
45
Cancer mortality Observed
Expected SMRa and 95% CI p-Value
Hodgkin lymphoma Non-Hodgkin lymphoma Multiple myeloma All leukemia
7 42 18 34
9.24 46.28 22.12 49
Cancer site
Cancer incidence
Hodgkin lymphoma Non-Hodgkin lymphoma Multiple myeloma All Leukemia
0.76 0.91 0.81 0.69
(0.30, 1.56) (0.65, 1.23) (0.48, 1.29) (0.48, 0.97)
0.594 0.59 0.45 0.031
Observedb Expected SIRa and 95% CI
p-Value
14 80 20 53
0.935 0.312 0.055 0.088
14.98 90.09 30.67 67.13
0.93 0.89 0.65 0.79
(0.51, 1.57) (0.70, 1.11) (0.40, 1.01) (0.59, 1.03)
Abbreviations: CI, confidence interval; SIR, standardized incidence ratio; SMR, standardized mortality ratio. a
Adjusted for age and calendar year at risk by stratification. The number of observed cases reflects the incidence of newly diagnosed cancer cases where a single individual can contribute more than one case of cancer. b
the mortality analysis cohort, probably because workers with the higher exposures (i.e., Port Radium workers in the 1930s and 1940s) died prior to 1969. Women had lower RDP exposures than men, although γ-ray doses were comparable. Port Hope workers had the highest γ-ray doses; 2-fold and 4-fold higher than Port Radium and Beaverlodge workers, respectively. Table 2 shows that the SMRs for HL, NHL, MM, and all leukemia were lower than the Canadian general population; however, the SMR was only statistically significant for all leukemia (SMR ¼0.69, 95% CI: 0.48, 0.97, p-value ¼0.031). Similar results were observed for the SIR analysis, but all leukemia (SIR ¼0.79, 95% CI: 0.59, 1.03, p-value ¼ 0.088) was no longer statistically significant. Only the SIR for MM approached statistical significance (SIR ¼0.65, 95% CI: 0.40, 1.01, p-value ¼0.055). Age at risk, calendar year at risk, employment site and total duration of employment significantly affected risk estimates so all subsequent analyses were controlled for these variables using stratification. There were 101 deaths from hematologic cancers (7 HD, 42 NHL, 18 MM, and 34 all leukemia, Table 3) and 553,492 personyears at risk in the male Eldorado mortality cohort. During 1969– 1999 incidence follow-up of this cohort, 160 cases of hematologic cancers were diagnosed (10 HD, 78 NHL, 20 MM, and 52 all leukemia, Table 3) and 373,477 person-years accumulated. Table 3 shows no relationship between RDP exposure and mortality or cancer incidence for any of the hematologic cancers (Table 3). HL incidence was borderline significant (N ¼10, ERR/100 WLM ¼20.7, 95% CI: o0, 324, p ¼0.081); however, this was based on few cases and a very wide confidence interval. In the γ-ray dose analysis, a statistically non-significant excess risk was also found for HL incidence (N ¼10, ERR/Sv ¼13.0, 95% CI: o0, 139, p-value ¼ 0.504) and CLL incidence (N ¼22, ERR/Sv ¼7.52, 95% CI: o0, 57.1, p-value ¼ 0.375) and NHL mortality (N ¼ 42, ERR/Sv ¼3.54, 95% CI: o0, 29.5, p-value ¼0.593). In the categorical analysis, risks of HL and CLL incidence increased with γ-ray dose, but linear trend tests were not statistically significant (p for linear trend¼0.672 and 0.271, respectively, not shown). Table 4 shows that HL incidence was significantly associated with γ-ray dose among Beaverlodge workers, but the finding was based on few cases (N ¼7, ERR/Sv¼ 91.3, 95% CI: o 0, 415, p-value ¼ 0.145). A high, although not statistically significant
46
L.B. Zablotska et al. / Environmental Research 130 (2014) 43–50
Table 3 ERRs for various causes of death and cancer in relation to RDP exposures and γ-ray doses. γ-Ray doses
RDP exposures Cause of death
Mortality Number of deaths
ERR/100 WLM (95% CI)a
p-Valueb
ERR/Sv (95% CI)c
p-Valueb
All malignant lymphoma Hodgkin lymphoma Non-Hodgkin lymphoma Multiple myeloma All Leukemia All hematological cancers
49 7 42 18 34 101
0 (n.e.) 0.2 ( o 0, 5.79) 0.03 (n.e.) 0.04 (o 0, 1.07) 0.02 ( o 0, 0.46) 0.02 ( o 0, 0.23)
0.9 0.576 0.768 0.845 0.808 0.799
0.29 (n.e.) 0.29 (n.e.) 3.54 ( o 0, 29.5)d 0.29 (n.e.) 0.29 (n.e.) 0.29 (n.e.)
0.825 0.752 0.593 0.639 0.824 0.572
Cancer site
Cancer incidence Number of cases
ERR/100 WLM (95% CI)a
p-Valueb
ERR/Sv (95% CI)c
p-Valueb
88 10 78 20 22 30 160
0.06 (o 0, 0.52) 20.7 ( o 0, 324) 0.04 (o 0, 0.46) 0.01 ( o0, 0.66) 0.04 (n.e.) 0.04 (n.e.) 0.01 ( o 0, 0.17)
0.568 0.081 0.676 0.935 0.59 0.489 0.823
0.34 (n.e.) 13.0 (o 0, 139)d 0.34 (n.e.) 0.34 (n.e.) 7.52 ( o 0, 57.1) 0.34 (n.e.) 0.34 (n.e.)
0.764 0.504 0.715 0.593 0.375 0.739 0.536
All malignant lymphoma Hodgkin lymphoma Non-Hodgkin lymphoma Multiple myeloma CLL Leukemia, excluding CLL All hematological cancers
Abbreviations: CLL, chronic lymphocytic leukemia; ERR, excess relative risk; n.e., not estimated; RDP, radon decay products; Sv, sievert; WLM, working level months. a
Model adjusted for employment site, age at risk, calendar year at risk and duration of employment by stratification. The γ-ray doses were not included in the model. p-Values from the likelihood ratio test comparing nested model with and without the continuous exposure term. c Model adjusted for employment site, age at risk, calendar year at risk and duration of employment by stratification. RDP exposures were not included in the model. d ERR estimate increased appreciably ( Z 10%), but remained not statistically significant in the analyses with lag times 10 years; all other estimates remained unchanged. b
Table 4 Associations between RDP exposures and γ-ray doses and incidence of HL, NHL and CLL (1969–1999) by site. Characteristic
Correlation between WLM and γ-ray dose Hodgkin lymphoma Number of cases γ-Ray doses ERR/Sv (95% CI) p-Valuea RDP exposures ERR/100 WLM (95% CI) p-Value Non-Hodgkin lymphoma Number of cases γ-Ray doses ERR/Sv (95% CI) p-Value RDP exposures ERR/100 WLM (95% CI) p-Value CLL Number of cases γ-Ray doses ERR/Sv (95% CI) p-Value RDP exposures ERR/100 WLM (95% CI) p-Value
Site Beaverlodge
Port radium
Port hope
Other sites
0.45
0.24
0.93
0.05
7
2
1
91.3 (o 0, 415) 0.145
n.e.
n.e.
n.e.
23.2 ( o 0, 365) 0.19
n.e.
n.e.
n.e.
37
21
15
5
2.54 (n.e.) 0.388
2.19 ( o 0, 27.6) 40.50
0.34 (n.e.) 40.50
n.e.
0.062 (n.e.) 40.50
0.14 ( o 0, 0.99) 0.372
0.17 (n.e.) 40.50
n.e.
15
3
4
11.8 ( o 0, 84.3) 0.268
10.2 ( o 0, 731) 40.50
0.34 (n.e.) 40.50
n.e.
0.17 ( o 0, 1.89) 40.50
0.04 (n.e.) 40.50
0.17 (n.e.) 40.50
n.e.
Abbreviations: CLL, chronic lymphocytic leukemia; ERR, excess relative risk; n.e., not estimated; RDP, radon decay products; SD, standard deviation; Sv, sievert; WLM, working level months. a
p-Values from the likelihood ratio test comparing nested model with and without the continuous exposure term.
increased risk of HL incidence related to RDP exposures was also observed among workers from this mine site. A moderate correlation between RDP exposures and γ-ray dose (Pearson0 s r ¼0.45) was observed among Beaverlodge workers, so the effects of both exposures were examined in one model. The entire increase in risk of HL was primarily due to γ-ray dose since the addition of the
RDP exposure term to the model did not significantly improve the model fit (p 40.50, not shown). The risk of NHL incidence related to γ-ray dose was statistically non-significantly increased among Port Radium workers (N¼ 21, ERR/ Sv¼2.19, 95% CI: o0, 27.6, p-value 40.50). Risks increased with γ-ray dose: 15–49 mSv (RR¼1.35, 95% CI: 0.46, 3.98) and Z50 mSv
L.B. Zablotska et al. / Environmental Research 130 (2014) 43–50
(RR¼1.53, 95% CI: 0.47, 5.00) compared to o15 mSv (p for linear trend¼ 0.537, not shown). The addition of the RDP exposure term to the model did not significantly improve the model fit (p40.50, not shown). The risk of CLL incidence related to γ-ray dose was statistically non-significantly increased among Beaverlodge workers (N ¼15, ERR/Sv ¼11.8, 95% CI: o 0, 84.3, p-value ¼0.268) and Port Radium workers (N ¼3, ERR/Sv ¼10.2, 95% CI: o0, 731, p-value 4 0.50). Again, risks increased with γ-ray dose and the addition of the RDP exposure terms to the models did not improve the models’ fit (both p-values 40.50, not shown). For Beaverlodge and Port Radium workers, risks increased with increasing γ-ray doses, but the risks for Port Hope workers, who had the highest γ-ray doses, were negative. Radiation-related risks of CLL did not differ across facilities (p heterogeneity across facilities 40.50, not shown). Three of the NHL cases in our study were small cell lymphomas (ICD-O-3: M9670.3), a type of B-cell lymphoma of which CLL is considered one of the stages (Harris et al., 1999). When these were analyzed together with CLL cases, the combined estimate was ERR/ Sv¼ 4.91, 95% CI: o0, 41.2, p ¼0.497. The risk among Beaverlodge and Port Radium workers was ERR/Sv ¼7.69, 95% CI: o0, 57.5, p ¼0.387 and ERR/Sv ¼8.74, 95% CI: o0, 657, p 40.50, respectively (not shown).
4. Discussion This report presents the analysis of 50 years of mortality (1950– 1999) and 31 years of cancer incidence (1969–1999) in a cohort of Eldorado uranium workers known to have worked sometime between 1932 and 1980 and followed-up until the end of 1999. Overall, workers had lower rates of mortality and cancer incidence of hematologic cancers compared to the age- and calendar year at risk adjusted rates for the general Canadian male population, a likely healthy worker effect (Howe et al., 1988). No statistically significant relationship was observed between RDP exposures or γ-ray doses, or a combination of both, and mortality or incidence of any hematologic cancers. However, for γ-ray doses, there was suggestive, though statistically non-significant, evidence of increased risk of CLL and HL incidence and NHL mortality with increasing dose. One of the main strengths of this study is its long-term follow-up using very high quality cancer incidence and mortality populationbased databases (Canadian Cancer Society, 2013; Goldberg et al., 1993). Comparatively high rates of follow-up were achieved through substantial improvements to the completeness to the nominal roll and work history files, multiple internal linkages to eliminate duplicate records, linkages to the HSTF, CMDB, CCDB, and the manual resolution of potential computer links. The follow-up of the entire cohort in terms of both incidence and mortality is unique and provided a complimentary view of the effects of RDP exposures and γ-ray doses on the risk of hematologic cancers. The total number of hematologic cancer cases (n ¼160) was significantly larger than the number of deaths (n ¼101), even though incidence follow-up was 19 years shorter than mortality follow-up. This is most likely because people diagnosed with HL, NHL and CLL, the most indolent of hematologic cancers, often do not die from these conditions (Canadian Cancer Society, 2013); thus workers with these conditions will often not have them recorded as the underlying cause of death on death certificates. If any advances in treatment occur during the study period, mortality would become a less sensitive outcome, whereas incidence would be unaffected. Likewise, cancers with high survival rates, such as HL, NHL and CLL, would not be detected by mortality statistics.
47
Hematologic cancers, in particular leukemia, are among the most radiosensitive cancers (NRC, 2006). They originate in the mutated cells of the bone marrow and from the mutated cells of the immune system. Whole-body γ-ray doses have been found to significantly increase risks of leukemia in A-bomb survivors from Japan (Hsu et al., 2013), risks of leukemia, NHL and MM in the U.K. radiation workers (Muirhead et al., 2009), and risks of leukemia and NHL in Chornobyl cleanup workers (Kesminiene et al., 2008; Zablotska et al., 2013). Less is known about the effects of radon decay alpha particles on the bone marrow, but a recent study (Harley and Robbins, 2009) reported that the bronchial mucosa has an abundance of circulating lymphocytes, thus suggesting that RDP exposures could be associated with hematologic cancers originating from these cells. Several limitations should be considered when interpreting the above results. First, the radiation exposure and corresponding doses to red bone marrow in uranium workers come from exposures to radon gas, RDP and long-lived radionuclides present in the uranium ore dust and to external gamma radiation (Marsh et al., 2012). Similar to other studies (NRC, 1999), the current analysis only evaluated RDP exposures, which were calculated as whole-body doses, while bone marrow doses would be a more relevant measure (NRC, 2006). The impact of this is to possibly underestimate radiation risks. Another important limitation is the study0 s limited statistical power due to the low RDP exposures and γ-ray doses, and the rareness of hematologic cancers. This could be addressed through further follow-up and pooling of this cohort with other cohorts of uranium workers. Effects of other carcinogens in ore require careful consideration. Beaverlodge ore was relatively clean with minimal amounts of other carcinogens, but Port Radium ore contained arsenic and cobalt. Arsenic, a known human carcinogen (IARC, 1980), was recently shown to increase lung cancer among uranium miners (Taeger et al., 2009); however, no such relationship was reported for lymphoma, MM or leukemia. It is unlikely that the suggested increased risk of CLL reported here is confounded by these carcinogens. In addition to ore, Port Hope workers were exposed to processing chemicals (i.e., ammonia, fluoride) and a variety of uranium compounds at higher concentration and of greater solubility (i.e., UO3, UF6) than that found in the ore. Port Hope workers were also exposed to radium compounds, uranium metal and some enriched uranium, which were recently shown to increase the risks of hematologic cancers (Guseva Canu et al., 2011). However, Port Hope workers provided no evidence of elevated risk of hematologic cancers. A recent comprehensive report (Preston et al., 2013) evaluated radiation exposure assessment errors and their impact on health risks associated with exposure to ionizing radiation and listed numerous sources of uncertainty associated with exposure to radon-decay products. One of the main sources of exposure errors would be shared dose errors, common to all individuals within a group, as well as unshared errors, unique to an individual within a cohort. Most RDP exposure measurement errors in the current study are likely to be non-differential. As pointed out by Blair et al. (2007), misclassification of exposure in studies of the association between airborne measurements of radon gas and cancer in underground miners is unlikely and the estimates of relative risk are generally unbiased. Measurement errors in exposure estimation almost certainly decreased with calendar time; thus, the Port Radium cohort had greater measurement errors than the Beaverlodge cohort, and recent workers had lower mean errors than earlier workers. The impact of such measurement errors depends on a number of factors, in particular, the quantitative nature of the error and the risk function considered. Further studies are necessary to better
48
L.B. Zablotska et al. / Environmental Research 130 (2014) 43–50
understand the effects of uncertainties in dose estimation on risk estimates. A further consideration is that residential radon exposure likely had a greater relative contribution to total RDP exposure in recent times when occupational exposures were lower. It is known that many houses in Uranium City, Port Hope, and the Eldorado town site required remedial work for reduction of radon concentrations. It is likely Eldorado workers lived in these communities at some point; however, information is not available to assign individual residential radon exposures to workers, so it is not possible to estimate residential radon risks. However, if there is no relationship between residential radon and total occupational RDP exposure, the risk estimates would be unbiased. In general, random misclassification of residential radon exposure would bias relative risk estimates towards the null. Tobacco smoking is a recognized risk factor for some leukemia subtypes (Musselman et al., 2013), but the evidence for association of smoking with NHL (Gibson et al., 2013) and multiple myeloma (Psaltopoulou et al., 2013) is less consistent. For smoking to modify RDP- or gamma-related risks of hematologic cancers it should also be correlated with RDP exposure and γ-ray doses. A case-control study of the Beaverlodge miners (L’Abbe et al., 1991), which accounted for close to 60% of uranium miners in the Eldorado cohort, provided no evidence that RDP exposure was correlated with cigarette smoking among miners. Even though smoking was banned at the Port Hope facility in the 1940s and 1950s and was allowed on a very limited basis thereafter, and was banned in the workplace at Beaverlodge in 1975, people still smoked outside the workplace. Although smoking data were not available, our previous analysis of smoking-related cancers other than lung cancer in the Eldorado cohort (Lane et al., 2010) showed that generally they were not elevated, suggesting that smoking was not substantially elevated relative to the general Canadian male population. Recent nested case-control studies of German (Schnelzer et al., 2010), French (Leuraud et al., 2007) and Czech (Tomasek, 2011) uranium miners, as well as the pooled analysis of these cohorts (Leuraud et al., 2011), showed that the RDP-associated risks of lung cancer persist even after adjustment for smoking. Similarly, smoking did not modify the γ-ray associated risks of cardiovascular outcomes in the Japanese A-bomb survivors (Shimizu et al., 2010) or workers of the Mayak Production Association in the Southern Urals region of the Russian Federation (Azizova et al., 2010). This cohort presented a unique opportunity to investigate the effects of RDP exposures and γ-ray doses in the same subjects. In our cohort, RDP exposures and γ-ray doses generally were not correlated (Pearson0 s r¼0.17) except for the Port Hope site, where they were strongly correlated (Pearson0 s r¼0.93). Comparatively high γ-ray doses can increase the risk of several cancer sites (UNSCEAR, 2008), with leukemia being particularly sensitive. However, in the present context, the mean γ-ray dose was fairly low (mean γ-ray dose¼ 52.2 mSv for men), so it is certainly possible that the study had low statistical power to detect any effect. Risks of CLL recently received attention after two incidencebased studies of occupationally exposed uranium workers suggested that the increased risk of CLL may be due to γ-ray doses rather than RDP exposures. One study (Rericha et al., 2006) reported a significant increase in CLL among Czech uranium miners (RR ¼1.96, 95%CI: 1.12, 3.42, p¼ 0.02 comparing the 80th percentile of exposure (20 mGy) to the 20th percentile of exposure (1 mGy)). The other study (Mohner et al., 2006) reported an elevated risk of CLL in East German Uranium Miners only when γ-radiation dose was examined alone. The current study also did not observe an increase in CLL risk with RDP exposures, but found a statistically non-significant increase in risk for γ-ray doses (ERR ¼ 7.52 per Sv, 95% CI: o0,
57.1, p ¼0.375). Relative risks increased with increasing dose, but a linear trend test was not significant (15–49 mSv (RR ¼1.97, 95% CI: 0.65, 5.99) and Z50 mSv (RR ¼2.25, 95% CI: 0.66, 7.64) compared to o15 mSv, p for linear trend ¼ 0.271). Furthermore, radiation risks of CLL were increased both in Beaverlodge and Port Radium workers (although based on few cases and with wide confidence intervals) and there was no heterogeneity in risks across all facilities (p 4 0.50). The larger magnitude of radiation risks of CLL in our analysis compared to other studies may reflect the greater γ-ray doses observed in our cohort compared to other cohorts of uranium miners. While a recent pooled analysis of nuclear power industry workers from 15 countries found little evidence for an association between low doses of external ionizing radiation, primarily γ-ray dose, and CLL mortality (Vrijheid et al., 2008), recent studies of incident CLL among Chornobyl cleanup workers reported increased risks in relation to bone marrow doses of γ-radiation (Kesminiene et al., 2008; Romanenko et al., 2008). Twenty-year follow up of Chornobyl cleanup workers from Ukraine provided evidence of significantly increased risks of CLL due to protracted gamma radiation exposures (ERR/Gy ¼2.93, 95% CI: 0.07, 9.64, p¼ 0.041) (Zablotska et al., 2013), as did a fifty-five year follow up of the A-bomb survivors (Hsu et al., 2013). Richardson et al. (2009) reported elevated risk of NHL mortality associated with whole-body radiation dose in both a cohort of male nuclear workers and in the A-bomb survivors with γ-ray doses lagged by 5 and 10 years. We observed statistically nonsignificantly increased radiation risks of NHL with γ-ray doses lagged by 5 years. Similar results were observed when we repeated our analyses using a 10-year lag (not shown). The results for radiation risks of NHL incidence were inconsistent across employment sites (negative for Beaverlodge workers; increased for Port Radium workers). Finally, MM incidence and mortality were not related to RDP exposures or γ-ray doses under various lag assumptions, consistent with a prior study of Czech uranium miners (Rericha et al., 2006).
5. Conclusions Overall, workers had lower rates of mortality and cancer incidence of hematologic cancers compared to the general Canadian male population. No statistically significant relationship was observed between RDP exposures or γ-ray doses, or a combination of both, and mortality or cancer incidence of hematologic cancers. However, there was suggestive, though statistically non-significant, evidence of an increased risk of CLL and HL incidence, and NHL mortality with increasing γ-ray dose. Further follow-up of hematologic cancer incidence may shed more light on the risks from lowdose γ-radiation. Pooling of the data from this and other uranium worker cohorts will improve the statistical power of future analyses.
Sources of funding The updated Eldorado study was supported by the Canadian Nuclear Safety Commission; the Government of Saskatchewan; Cameco Corporation; and AREVA Resources Canada Inc. The study was conducted to address the recommendation of the 1993 Joint Federal-Provincial Panel on Uranium Mining Developments in Northern Saskatchewan. All data checking, analysis, interpretation and report writing was done independently of the mining companies, which had no part in any aspect of the work of the scientific research team of the present study. Additional funding for this study was provided by the U.S. National Cancer Institute (NCI grant CA132918 to L.B.Z.).
L.B. Zablotska et al. / Environmental Research 130 (2014) 43–50
Ethics in publishing The Eldorado study was conducted in accordance with accepted ethical practices and was approved by Health Canada0 s Research Ethics Board and Institutional Review Board Services.
Acknowledgments We wish to thank the members of the Saskatchewan Uranium Miners Cohort (SUMC) Study Working Group and Steering Committee (the Canadian and Saskatchewan governments, and the mining industry0 s management and workers health and safety representatives) for their technical and practical advice, for keeping the respective organizations and stakeholders informed, and for their in-kind support. We thank Health Canada for the use of the National Dose Registry and Statistics Canada for the linkage of the cohort nominal roll file with the tax summary, mortality and cancer incidence data. The views and conclusions expressed are those of the scientific research team and not of Statistics Canada. We gratefully acknowledge the contribution of the provincial and territorial Vital Statistics Registrars and Cancer Registries. We thank Ms. Karen Bartley for review of the early draft of the manuscript. References Azizova, T.V., Muirhead, C.R., Druzhinina, M.B., et al., 2010. Cardiovascular diseases in the cohort of workers first employed at Mayak PA in 1948–1958. Radiat. Res. 174, 155–168. Blair, A., Stewart, P., Lubin, J.H., Forastiere, F., 2007. Methodological issues regarding confounding and exposure misclassification in epidemiological studies of occupational exposures. Am. J. Ind. Med. 50, 199–207. Boice Jr, J.D., 1992. Radiation and non-Hodgkin0 s lymphoma. Cancer Res. 52, 5489s–5491s. Boice Jr, J.D., Bigbee, W.L., Mumma, M.T., Heath Jr, C.W., Blot, W.J., 2009. Cancer incidence in municipalities near two former nuclear materials processing facilities in Pennsylvania—an update. Health Phys. 96, 118–127. Boice Jr, J.D., Cohen, S.S., Mumma, M.T., Chadda, B., Blot, W.J., 2007. Mortality among residents of Uravan, Colorado who lived near a uranium mill, 1936–84. J. Radiol. Prot. 27, 299–319. Boice Jr, J.D., Mumma, M.T., Blot, W.J., 2010. Cancer incidence and mortality in populations living near uranium milling and mining operations in grants, New Mexico, 1950–2004. Radiat. Res. 174, 624–636. Breslow, N.E., Day, N.E., 1987. Statistical Methods in Cancer Research. Volume 2— The Design and Analysis of Cohort Studies. International Agency for Research on Cancer, Lyon Canadian Cancer Society. Canadian Cancer Statistics 2013. Toronto, ON, 2013. Cardis, E., Vrijheid, M., Blettner, M., et al., 2007. The 15-country collaborative study of cancer risk among radiation workers in the nuclear industry: estimates of radiation-related cancer risks. Radiat. Res 167, 396–416. Darby, S.C., Whitley, E., Howe, G.R., et al., 1995. Radon and cancers other than lung cancer in underground miners: a collaborative analysis of 11 studies. J. Natl. Cancer Inst. 87, 378–384. Fritz, A., Jack, A.S., Parkin, D.M., et al., 2000. International Classification of Diseases for Oncology (ICD-O). World Health Organization, Geneva Gibson, T.M., Smedby, K.E., Skibola, C.F., et al., 2013. Smoking, variation in N-acetyltransferase 1 (NAT1) and 2 (NAT2), and risk of non-Hodgkin lymphoma: a pooled analysis within the InterLymph consortium. Cancer Causes Control 24, 125–134. Goldberg, M.S., Carpenter, M., Theriault, G., Fair, M., 1993. The accuracy of ascertaining vital status in a historical cohort study of synthetic textiles workers using computerized record linkage to the Canadian Mortality Data Base. Can. J. Public Health 84, 201–204. Guseva Canu, I., Jacob, S., Cardis, E., et al., 2011. Uranium carcinogenicity in humans might depend on the physical and chemical nature of uranium and its isotopic composition: results from pilot epidemiological study of French nuclear workers. Cancer Causes Control. 22, 1563–1573. Harley, N.H., Robbins, E.S., 2009. Radon and leukemia in the Danish study: another source of dose. Health Phys. 97, 343–347. Harris, N.L., Jaffe, E.S., Diebold, J., et al., 1999. World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: report of the Clinical Advisory Committee meeting-Airlie House, Virginia, November 1997. J. Clin. Oncol 17, 3835–3849. Howe, G.R., Chiarelli, A.M., Lindsay, J.P., 1988. Components and modifiers of the healthy worker effect: evidence from three occupational cohorts and implications for industrial compensation. Am. J. Epidemiol. 128, 1364–1375.
49
Howe, G.R., Nair, R.C., Newcombe, H.B., Miller, A.B., Abbatt, J.D., 1986. Lung cancer mortality (1950–80) in relation to radon daughter exposure in a cohort of workers at the Eldorado Beaverlodge uranium mine. J. Natl. Cancer Inst. 77, 357–362. Howe, G.R., Nair, R.C., Newcombe, H.B., Miller, A.B., Burch, J.D., Abbatt, J.D., 1987. Lung cancer mortality (1950–80) in relation to radon daughter exposure in a cohort of workers at the Eldorado Port Radium uranium mine: possible modification of risk by exposure rate. J. Natl. Cancer Inst. 79, 1255–1260. Hsu, W.L., Preston, D.L., Soda, M., et al., 2013. The incidence of leukemia, lymphoma and multiple myeloma among atomic bomb survivors: 1950–2001. Radiat. Res. 179, 361–382. IARC (International Agency for Research on Cancer), 1980. Some Metals and Metallic Compounds., IARC Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Humans. World Health Organization, International Agency for Research on Cancer, Lyon, France, pp. 1–438 Kesminiene, A., Evrard, A.S., Ivanov, V.K., et al., 2008. Risk of hematological malignancies among Chernobyl liquidators. Radiat. Res. 170, 721–735. L’Abbe, K.A., Howe, G.R., Burch, J.D., et al., 1991. Radon exposure, cigarette smoking, and other mining experience in the Beaverlodge uranium miners cohort. Health Phys. 60, 489–495. Lane, R.S., Frost, S.E., Howe, G.R., Zablotska, L.B., 2010. Mortality (1950–1999) and cancer incidence (1969–1999) in the cohort of Eldorado uranium workers. Radiat. Res. 174, 773–785. Laurier, D., Valenty, M., Tirmarche, M., 2001. Radon exposure and the risk of leukemia: a review of epidemiological studies. Health Phys. 81, 272–288. Leuraud, K., Billon, S., Bergot, D., et al., 2007. Lung cancer risk associated to exposure to radon and smoking in a case-control study of French uranium miners. Health Phys. 92, 371–378. Leuraud, K., Schnelzer, M., Tomasek, L., et al., 2011. Radon, smoking and lung cancer risk: results of a joint analysis of three European case-control studies among uranium miners. Radiat. Res 176, 375–387. Marsh, J.W., Blanchardon, E., Gregoratto, D., et al., 2012. Dosimetric calculations for uranium miners for epidemiological studies. Radiat. Prot. Dosim. 149, 371–383. Mohner, M., Gellissen, J., Marsh, J.W., Gregoratto, D., 2010. Occupational and diagnostic exposure to ionizing radiation and leukemia risk among German uranium miners. Health Phys. 99, 314–321. Mohner, M., Lindtner, M., Otten, H., Gille, H.G., 2006. Leukemia and exposure to ionizing radiation among German uranium miners. Am. J. Ind. Med. 49, 238–248. Muirhead, C.R., O’Hagan, J.A., Haylock, R.G., et al., 2009. Mortality and cancer incidence following occupational radiation exposure: third analysis of the National Registry for Radiation Workers. Br. J. Cancer 100, 206–212. Musselman, J.R., Blair, C.K., Cerhan, J.R., Nguyen, P., Hirsch, B., Ross, J.A., 2013. Risk of adult acute and chronic myeloid leukemia with cigarette smoking and cessation. Cancer Epidemiol 37, 410–416. NRC (National Research Council), 1999. Committee on Health Risks of Exposure to Radon. Health Effects of Exposure to Radon. Biological Effects of Exposure to Ionizing Radiation: BEIR VI. NRC, National Academy Press, Washington, DC. NRC (National Research Council), 2006. Committee to Assess Health Risksfrom Exposure to Low Levels of Ionizing Radiation. HealthRisks From Exposure to Low Levels of Ionizing Radiation. BEIR VII Phase 2. NRC, National Academies Press, Washington, DC. Polednak, A.P., Frome, E.L., 1981. Mortality among men employed between 1943 and 1947 at a uranium- processing plant. J. Occup. Med. 23, 169–178. Preston, D.L., Kusumi, S., Tomonaga, M., et al., 1994. Cancer incidence in atomic bomb survivors. Part III. Leukemia, lymphoma and multiple myeloma, 1950– 1987. Radiat. Res. 137, S68–S97. Preston, D.L., Lubin, J.H., Pierce, D.A., McConney, M.E., 1993. EPICURE User0 s Guide. Hirosoft International Corporation, Seattle, WA Preston, R.J., Boice Jr, J.D., Brill, A.B., et al., 2013. Uncertainties in estimating health risks associated with exposure to ionising radiation. J. Radiol. Prot. 33, 573–588. Psaltopoulou, T., Sergentanis, T.N., Kanellias, N., Kanavidis, P., Terpos, E., Dimopoulos, M.A., 2013. Tobacco smoking and risk of multiple myeloma: a meta-analysis of 40 observational studies. Int. J. Cancer 132, 2413–2431. Rericha, V., Kulich, M., Rericha, R., Shore, D.L., Sandler, D.P., 2006. Incidence of leukemia, lymphoma, and multiple myeloma in Czech uranium miners: a casecohort study. Environ. Health Perspect. 114, 818–822. Richardson, D.B., Sugiyama, H., Wing, S., et al., 2009. Positive associations between ionizing radiation and lymphoma mortality among men. Am. J. Epidemiol. 169, 969–976. Richardson, D.B., Wing, S., Schroeder, J., Schmitz-Feuerhake, I., Hoffmann, W., 2005. Ionizing radiation and chronic lymphocytic leukemia. Environ. Health Perspect. 113, 1–5. Romanenko, A., Finch, S., Hatch, M., et al., 2008. The Ukrainian-American study of leukemia and related disorders among Chornobyl cleanup workers from Ukraine: III. Radiation risks. Radiat. Res. 170, 711–720. Ron, E., 1998. Ionizing radiation and cancer risk: evidence from epidemiology. Radiat. Res. 150, S30–S41. Schnelzer, M., Hammer, G.P., Kreuzer, M., Tschense, A., Grosche, B., 2010. Accounting for smoking in the radon-related lung cancer risk among German uranium miners: results of a nested case-control study. Health Phys 98, 20–28. Shimizu, Y., Kodama, K., Nishi, N., et al., 2010. Radiation exposure and circulatory disease risk: Hiroshima and Nagasaki atomic bomb survivor data, 1950–2003. BMJ, 340; p. b5349.
50
L.B. Zablotska et al. / Environmental Research 130 (2014) 43–50
Taeger, D., Johnen, G., Wiethege, T., et al., 2009. Major histopathological patterns of lung cancer related to arsenic exposure in German uranium miners. Int. Arch. Occup. Environ. Health 82, 867–875. Tamura, K., Sawada, H., Izumi, Y., et al., 2001. Chronic lymphocytic leukemia (CLL) is rare, but the proportion of T-CLL is high in Japan. Eur. J. Haematol. 67, 152–157. Tomasek, L., 2011. Interaction of radon and smoking among Czech uranium miners. Radiat. Prot. Dosimetry 145, 238–242. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation). Sources and Effects of Ionizing Radiation. UNSCEAR 2006 Report to the General
Assembly with Scientific Annexes. Volume I. Annex A: Epidemiological Studies of Radiation and Cancer. UNSCEAR, New York, 2008. Vrijheid, M., Cardis, E., Ashmore, P., et al., 2008. Ionizing radiation and risk of chronic lymphocytic leukemia in the 15-country study of nuclear industry workers. Radiat. Res 170, 661–665. WHO (World Health Organization). 1998. International Classification of Diseases, Ninth Revision (ICD-9), vol. 1. WHO, Geneva. Zablotska, L.B., Bazyka, D., Lubin, J.H., et al., 2013. Radiation and the risk of chronic lymphocytic and other leukemias among chornobyl cleanup workers. Environ. Health Perspect. 121, 59–65.
