Risk of cancer after low doses of ionising radiation: retrospective co
http://www.100md.com
《英国医生杂志》
1 International Agency for Research on Cancer, Lyons, France, 2 Institute of Medical Biostatistics, Epidemiology and Informatics, University of Mainz, Germany, 3 Radiation Epidemiology Branch, Division of Epidemiology and Genetics, National Cancer Institute, Bethesda, MD, USA, 4 University of Tampere, Tampere, Finland, 5 Institut Gustave-Roussy, Villejuif, France, 6 Mailman School of Public Health, Columbia University, New York, USA, 7 National Centre in HIV Epidemiology and Clinical Research, Sydney, NSW, Australia, 8 Radiation Protection Division, Health Protection Agency, Chilton, Didcot, Oxfordshire, 9 Division of Surveillance, Hazard Evaluations and Field Studies, National Institute for Occupational Safety and Health, Cincinnati, OH, USA, 10 Fukuoka Institute of Health and Environmental Sciences, Fukuoka, Japan, 11 Conseiller Médical du CEA, Paris, France, 12 Atomic Energy Commission of Canada, Deep River, ON, Canada, 13 Pacific Northwest National Laboratory, Richland, WA, USA, 14 Safety and Radiation Science, Australian Nuclear Science and Technology Organisation, NSW, Australia, 15 AECL Radiation Biology and Health Physics Branch, Chalk River Laboratories, Chalk, ON, Canada, 16 Twin Trees, Blewbury, Didcot, Oxfordshire, 17 Department of Preventive Medicine, Seoul National University College of Medicine, Seoul, Korea, 18 Epidemiological Research and Surveillance Unit in Transport, Occupation and Environment; French National Institute for Transport and Safety Research (INRETS), Arcueil, France, 19 Radiation Protection Bureau, Health Canada, Ottawa, Canada, 20 Department of Preventive Medicine, Cheju National University College of Medicine, Cheju, Korea, 21 AMYS, UNESA, Madrid, Spain, 22 Institute of Radioprotection and Nuclear Security, DESTQ/DEAS, Le Vésinet, France, 23 Nuclear Research Centre (SCK.CEN), Radiation Protection Division, Mol, Belgium, 24 Operational Radiological Protection, Nuclear Safety Council, Spain, 25 Statens Vattenfallsverk Forsmark, Osthammer, Sweden, 26 Institute for East-European Studies, Uppsala University, Sweden, 27 Department of Hygiene and Epidemiology, Faculty of Health Care and Social Work, Trnava University, Trnava, Slovak Republic, 28 Faculty of Health Sciences, American University of Beirut, Lebanon, 29 Slovenske Elektrárne, Bratislava, Slovak Republic, 30 Doseco Ltd, Jyvaskyla, Finland, 31 "Frederic Joliot-Curie" National Research Institute for Radiobiology and Radiohygiene of the "Fodor József" National Centre for Public Health, Budapest, Hungary, 32 Lithuanian Cancer Registry, Vilnius University Oncology Institute, Vilnius, Lithuania, 33 Vasternorrland County Council, Department of Research and Development, Sundsvall, Sweden, 34 WHO European Centre for Environment and Health, Rome, Italy, 35 Radiation Protection Centre, Vilnius, Lithuania, 36 Serono International SA, Geneva, Switzerland, 37 Physics and Biology Section, Radiation Protection Division, Federal Office of Public Health, Bern, Switzerland, 38 School of Clinical Medical Sciences, University of Newcastle upon Tyne, Newcastle upon Tyne, 39 Department of Epidemiology, School of Public Health, Chapel Hill, NC, USA, 40 Department of Preventive Medicine and Public Health, School of Medicine, University Autonoma de Madrid, Spain, 41 Laboratory of Epidemiology, Institute for Radiological Protection and Nuclear Safety, Fontenay-aux-Roses, France, 42 Medical Inspectorate of Factories, Geneva, Switzerland, 43 Institute of Public Health, Semmelweis University, Budapest, Hungary
Correspondence to: E Cardis cardis@iarc.fr
Objectives To provide direct estimates of risk of cancer after protracted low doses of ionising radiation and to strengthen the scientific basis of radiation protection standards for environmental, occupational, and medical diagnostic exposures.
Design Multinational retrospective cohort study of cancer mortality.
Setting Cohorts of workers in the nuclear industry in 15 countries.
Participants 407 391 workers individually monitored for external radiation with a total follow-up of 5.2 million person years.
Main outcome measurements Estimates of excess relative risks per sievert (Sv) of radiation dose for mortality from cancers other than leukaemia and from leukaemia excluding chronic lymphocytic leukaemia, the main causes of death considered by radiation protection authorities.
Results The excess relative risk for cancers other than leukaemia was 0.97 per Sv, 95% confidence interval 0.14 to 1.97. Analyses of causes of death related or unrelated to smoking indicate that, although confounding by smoking may be present, it is unlikely to explain all of this increased risk. The excess relative risk for leukaemia excluding chronic lymphocytic leukaemia was 1.93 per Sv (< 0 to 8.47). On the basis of these estimates, 1-2% of deaths from cancer among workers in this cohort may be attributable to radiation.
Conclusions These estimates, from the largest study of nuclear workers ever conducted, are higher than, but statistically compatible with, the risk estimates used for current radiation protection standards. The results suggest that there is a small excess risk of cancer, even at the low doses and dose rates typically received by nuclear workers in this study.
Ionising radiation is one of the most studied and ubiquitous carcinogens in our environment. The main basis for radiation protection recommendations is the study of survivors of the Japanese atomic bomb (A bomb), a population exposed primarily at high dose rates.1-3 The primary public health concern, however, is the protection of people from relatively low dose, protracted or fractionated exposures such as those received by the public in the general environment, by patients through repeated diagnostic procedures,4 and by radiation workers.
The effects of low dose chronic exposure to external radiation have been directly estimated in several cohorts of workers in the nuclear industry,3 but the sample size has limited the precision of these estimates. Analyses of combined cohorts have improved precision.5-7 Estimates from these analyses, however, are compatible with a range of possibilities, from a reduction of risk at low doses to risks higher than those underlying current radiation protection recommendations.
The 15 country study, an international collaborative study of cancer risk among radiation workers in the nuclear industry, was carried out to further improve the precision of direct estimates of risk after protracted low dose exposures and to strengthen the scientific basis of radiation protection.1 We present risk estimates for mortality from all cancers, excluding leukaemia, and from leukaemia excluding chronic lymphocytic leukaemia and compare them with estimates derived from data on survivors of the A bomb. We have used the term nuclear industry to refer to facilities engaged in production of nuclear power, manufacture of nuclear weapons, enrichment and processing of nuclear fuel, production of radioisotopes, or reactor or weapons research. Uranium mining is not included.
Methods
This multinational retrospective cohort study used a common protocol in 15 countries and collected information on nearly 600 000 workers. Study cohorts were defined from employment or dosimetric records of participating facilities or, where available, from centralised national dose registries. The a priori eligibility criteria for inclusion of cohorts8 were essentially complete and non-selective follow-up for mortality; availability of individual annual recorded estimates of dose for all monitored workers; and availability of information on historical monitoring policies and practices. We included all workers who had been monitored for external photon (x and ) radiation exposure through the use of personal dosimeters. Details of country specific methods are described elsewhere.9
Ascertainment of vital status and cause of death
We established vital status through linkage with national or regional death registries. In a few countries where this was not possible, appropriate records of local authorities were used. Completeness of follow-up ranged from 87% to nearly 100%. Vital statistics registries provided cause of death, which was known for over 90% of workers who died.
Adequacy of dosimetric records
We reconstructed each worker's dosimetric history using recorded doses from individual facilities or national dose registries. A study of errors in recorded doses evaluated the comparability of dose estimates across facilities and time and identified and quantified sources of bias and uncertainties.9 Doses from higher energy photons (100-3000 keV), which constituted most of the dose in most cohorts, were judged to have been measured in a comparable way over time and across facilities.9 10 The adequacy of practices and technology to measure and record dose from other radiation types (neutrons, internal exposures), however, varied substantially, particularly in earlier years. We therefore excluded workers with potential for substantial doses ( 10% of their whole body dose) from these radiation types.
Main study population
The main study population was defined as workers who had been employed in one or more facilities for at least one year (113 711 workers excluded), who had been monitored for external radiation exposure (38 521 workers excluded), and whose doses resulted predominantly from higher energy photon radiation (39 730 workers with internal contamination and 19 041 with neutron exposures excluded).
Dosimetric errors and derivation of organ doses
The major sources of errors in higher energy photon doses were dosimetry technology, exposure conditions, and calibration practices. Errors from these sources were quantified and bias factors specific to the doses to each organ of interest calculated for each model of dosimeter used and by type of facility (nuclear power plants and mixed activities facilities).11 12 Organ doses were derived by dividing recorded doses by the appropriate organ dose bias factor. We used doses to the colon and active bone marrow for analyses of mortality from all cancers excluding leukaemia and from leukaemia, respectively. All doses are expressed as dose equivalents in sieverts (Sv).
Statistical methods
Analyses were based on a linear relative risk Poisson regression model, in which the relative risk is of the form 1+Z, where Z is the cumulative dose equivalent in Sv and is the excess relative risk per Sv; 95% likelihood based confidence intervals were calculated. We used 11 a priori categories of dose (< 5, 5- < 10, 10- < 20, 20- < 50, 50- < 100, 100- < 150, 150- < 200, 200- < 300, 300- < 400, 400- < 500, 500 mSv). Analyses used only underlying cause of death. Estimates of excess relative risk were stratified for sex, age, and calendar period (both in five year categories), facility, duration of employment (< 10 years, 10 years), and socioeconomic status. In the analyses of all cancers we excluded cohorts for which socioeconomic information was unavailable or incomplete (Japan, Idaho National Engineering Laboratory (INEL), Ontario Hydro Canada), but we included them in analyses of leukaemia as the potential for confounding by socioeconomic status was thought to be less for leukaemia. To allow for a latent period between exposure and death, doses were lagged by two years for leukaemia and 10 years for other cancers, as in other studies of nuclear workers5-7 and assessments of radiation risk.1 2 Sensitivity analyses were conducted with a range of different lags. Attributable risks were estimated by multiplying the excess relative risks by the average dose in the cohort.
We have focused on the main causes of death for which radiation protection committees have provided risk estimates: all cancers excluding leukaemia and leukaemia excluding chronic lymphocytic leukaemia. Chronic lymphocytic leukaemia is excluded because it is thought to be less readily inducible by ionising radiation than other leukaemias.3 We have also presented risk estimates for solid cancers (that is, excluding lymphatic and haematopoietic malignancies) to compare with recent data for A bomb survivors13 and for all cancers excluding leukaemia, lung, and pleural cancers (which have the greatest potential for confounding by smoking, internally incorporated radionuclides, and other occupational carcinogens). We investigated confounding by smoking by separately analysing solid cancers related or unrelated to smoking and two groupings of smoking related outcomes other than cancer (all non-malignant respiratory diseases and chronic obstructive bronchitis and emphysema).
Analysis of data from survivors of A bomb
We analysed mortality data from the A bomb survivors for solid cancer to 199713 and leukaemia to 199014 using similar methods to provide risk estimates for comparison. Analyses were stratified for attained age, city, and calendar time and restricted to men aged 20-60 at exposure, the group most comparable to the workers.
Results
Overall, 598 068 workers were employed in at least one of 154 facilities. Most facilities were involved in nuclear power production; the rest specialised in different activities, including research, waste management, and production of fuel, isotopes, and weapons. The main study population comprised 407 391 workers (table 1). A total of 24 158 (5.9%) people were known to have died during the study period: 6519 from cancers other than leukaemia and 196 from leukaemia excluding chronic lymphocytic leukaemia. The total duration of follow-up was 5 192 710 person years and the total collective recorded dose was 7892 Sv. Most workers in the study were men (90%), and men received 98% of the collective dose. The overall average cumulative recorded dose was 19.4 mSv. The distribution of recorded doses was skewed (fig 1). Ninety per cent of workers received cumulative doses < 50 mSv and less than 0.1% received cumulative doses > 500 mSv.
Table 1 Cohorts included in the 15 country study
Fig 1 Distribution of cumulative radiation doses among workers included in the analyses
For all cancers excluding leukaemia, the excess relative risk was 0.97 per Sv and was significantly different from zero (95% confidence interval 0.14 to 1.97) (table 2). This estimate corresponds to a relative risk of 1.10 for a radiation dose of 100 mSv. For solid cancers, the excess relative risk was 0.87 (0.03 to 1.88), higher than but statistically compatible with the estimate for A bomb survivors (0.32 per Sv). The excess relative risk for leukaemia excluding chronic lymphocytic leukaemia was 1.93 per Sv (< 0 to 8.47), which gives a relative risk of 1.19 for a dose of 100 mSv. This estimate is between the linear and linear quadratic extrapolations from data on A bomb survivors (table 2).
Table 2 Estimates of excess relative risk per Sv (95% confidence interval) for all cancers excluding leukaemia, solid cancers, and leukaemia excluding chronic lymphocytic leukaemia, for nuclear workers and survivors of A bomb in Japan *
Table 3 assesses the possible confounding effect of smoking. Excess relative risks ranged between 0.59 per Sv (-0.29 to 1.70) for all cancers excluding leukaemia and lung and pleural cancer, and 0.91 per Sv (-0.11 to 2.21) for smoking related cancers.
Table 3 Estimates of excess relative risk per Sv (95% confidence intervals) for specific causes of death
The increased risk for smoking related cancers was mainly due to an increased risk of lung cancer (1.86 per Sv, 0.26 to 4.01). Other smoking related cancers showed little evidence of an increased risk (0.21 per Sv, < 0 to 2.01). Risk estimates for mortality from non-malignant respiratory diseases and from chronic obstructive bronchitis and emphysema were raised but not significantly different from zero (excess relative risk per Sv 1.16, -0.53 to 3.84, and 2.12, -0.57 to 7.46, respectively).
Discussion
BEIR V. Committee on the Biological Effects of Ionizing Radiation. The effects on populations of exposure to low levels of ionizing radiation. Washington, DC: National Academy of Sciences, 1990.
International Commission on Radiological Protection ICRP. Recommendations of the international commission on radiological protection. Oxford: Pergamon Press, 1991. (ICRP publication 60.)
United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR). Sources and effects of ionizing radiation. New York: United Nations, 2000.
Berrington de González A, Darby S. Risk of cancer from diagnostic X-rays: estimates for the UK and 14 other countries. Lancet 2004;363: 345-51.
IARC Study Group on Cancer Risk among Nuclear Industry Workers. Direct estimates of cancer mortality due to low doses of ionising radiation: an international study. Lancet 1994;344: 1039-43.
Cardis E, Gilbert ES, Carpenter L, Howe G, Kato I, Armstrong BK, et al. Effects of low doses and low dose rates of external ionizing radiation: cancer mortality among nuclear industry workers in three countries. Radiat Res 1995;142: 117-32.
Muirhead CR, Goodill AA, Haylock RGE, Vokes J, Little MP, Jackson DA, et al. Occupational radiation exposure and mortality: second analysis of the national registry for radiation workers. J Radiol Prot 1999;19: 3-26.
Cardis E, Estève J. International collaborative study of cancer risk among nuclear industry workers. I—Report of the feasibility study. II—Protocol report 92/001. Lyons: International Agency for Research on Cancer, 1992.
Cardis E, Martuzzi M, Amoros E. International collaborative study of cancer risk among nuclear industry workers. III—Procedures document, rev 1. Lyons: International Agency for Research on Cancer, 1997. (97/002.)
Fix JJ, Salmon L, Cowper G, Cardis E. A retrospective evaluation of the dosimetry employed in an international combined epidemiological study. Radiat Protect Dosimetry 1997;74: 39-53.
Thierry-Chef I, Pernicka F, Marshall M, Cardis E, Andreo P. Study of a selection of 10 historical types of dosemeter: variation of the response to Hp(10) with photon energy and geometry of exposure. Radiat Prot Dosimetry 2002;102: 101-13.
Thierry-Chef I, Cardis E, Ciampi A, Delacroix D, Marshall M, Amoros E, et al. A method to assess predominant energies of exposure in a nuclear research centre—Saclay (France). Radiat Prot Dosimetry 2001;94: 215-25.
Preston DL, Shimizu Y, Pierce DA, Suyama A, Mabuchi K. Studies of mortality of atomic bomb survivors. Report 13: Solid cancer and noncancer disease mortality: 1950-1997. Radiat Res 2003;160: 381-407.
Pierce DA, Shimizu Y, Preston DL, Vaeth M, Mabuchi K. Studies of the mortality of atomic bomb survivors. Report 12, Part I. Cancer: 1950-1990. Radiat Res 1996;146: 1-27.
International Agency for Research on Cancer. IARC monographs on the evaluation of carcinogenic risks to humans. Vol 83. Tobacco smoke and involuntary smoking. Lyons: IARC, 2004.
Lee DJ, LeBlanc W, Fleming LE, Gomez-Marin O, Pitman T. Trends in US smoking rates in occupational groups: the national health interview survey 1987-1994. J Occup Environ Med 2004;46: 538-48.
Gribbin MA, Weeks JL, Howe GR. Cancer mortality (1956-1985) amongst male employees of Atomic Energy of Canada Limited with respect to occupational exposure to external low linear energy transfer ionizing radiation. Radiat Res 1993;133: 375-80.
Murata M, Miyake T, Inoue Y, Ohshima S, Kudo S, Yoshimura T, et al. Life-style and other characteristics of radiation workers at nuclear facilities in Japan: base-line data of a questionnaire survey. J Epidemiol 2002;12: 310-9.
Peterson G, Gilbert ES, Buchanan JA, Stevens RG. A case-cohort study of lung cancer ionizing radiation, and tobacco smoking among males at the Hanford sites. Health Phys 1990;58: 3-11.
Carpenter L, Fraser P, Booth M, Higgins C, Beral V. Smoking habits and radiation exposure. J Radiol Protection 1989;9: 286-7.
Auvinen A, Pukkala E, Hyv?nen H, Hakama M, Ryt?maa T. Cancer incidence among Finnish nuclear reactor workers. J Occup Environ Med 2002;44: 634-8.(E Cardis, head of radiation group1, M Vr)
Correspondence to: E Cardis cardis@iarc.fr
Objectives To provide direct estimates of risk of cancer after protracted low doses of ionising radiation and to strengthen the scientific basis of radiation protection standards for environmental, occupational, and medical diagnostic exposures.
Design Multinational retrospective cohort study of cancer mortality.
Setting Cohorts of workers in the nuclear industry in 15 countries.
Participants 407 391 workers individually monitored for external radiation with a total follow-up of 5.2 million person years.
Main outcome measurements Estimates of excess relative risks per sievert (Sv) of radiation dose for mortality from cancers other than leukaemia and from leukaemia excluding chronic lymphocytic leukaemia, the main causes of death considered by radiation protection authorities.
Results The excess relative risk for cancers other than leukaemia was 0.97 per Sv, 95% confidence interval 0.14 to 1.97. Analyses of causes of death related or unrelated to smoking indicate that, although confounding by smoking may be present, it is unlikely to explain all of this increased risk. The excess relative risk for leukaemia excluding chronic lymphocytic leukaemia was 1.93 per Sv (< 0 to 8.47). On the basis of these estimates, 1-2% of deaths from cancer among workers in this cohort may be attributable to radiation.
Conclusions These estimates, from the largest study of nuclear workers ever conducted, are higher than, but statistically compatible with, the risk estimates used for current radiation protection standards. The results suggest that there is a small excess risk of cancer, even at the low doses and dose rates typically received by nuclear workers in this study.
Ionising radiation is one of the most studied and ubiquitous carcinogens in our environment. The main basis for radiation protection recommendations is the study of survivors of the Japanese atomic bomb (A bomb), a population exposed primarily at high dose rates.1-3 The primary public health concern, however, is the protection of people from relatively low dose, protracted or fractionated exposures such as those received by the public in the general environment, by patients through repeated diagnostic procedures,4 and by radiation workers.
The effects of low dose chronic exposure to external radiation have been directly estimated in several cohorts of workers in the nuclear industry,3 but the sample size has limited the precision of these estimates. Analyses of combined cohorts have improved precision.5-7 Estimates from these analyses, however, are compatible with a range of possibilities, from a reduction of risk at low doses to risks higher than those underlying current radiation protection recommendations.
The 15 country study, an international collaborative study of cancer risk among radiation workers in the nuclear industry, was carried out to further improve the precision of direct estimates of risk after protracted low dose exposures and to strengthen the scientific basis of radiation protection.1 We present risk estimates for mortality from all cancers, excluding leukaemia, and from leukaemia excluding chronic lymphocytic leukaemia and compare them with estimates derived from data on survivors of the A bomb. We have used the term nuclear industry to refer to facilities engaged in production of nuclear power, manufacture of nuclear weapons, enrichment and processing of nuclear fuel, production of radioisotopes, or reactor or weapons research. Uranium mining is not included.
Methods
This multinational retrospective cohort study used a common protocol in 15 countries and collected information on nearly 600 000 workers. Study cohorts were defined from employment or dosimetric records of participating facilities or, where available, from centralised national dose registries. The a priori eligibility criteria for inclusion of cohorts8 were essentially complete and non-selective follow-up for mortality; availability of individual annual recorded estimates of dose for all monitored workers; and availability of information on historical monitoring policies and practices. We included all workers who had been monitored for external photon (x and ) radiation exposure through the use of personal dosimeters. Details of country specific methods are described elsewhere.9
Ascertainment of vital status and cause of death
We established vital status through linkage with national or regional death registries. In a few countries where this was not possible, appropriate records of local authorities were used. Completeness of follow-up ranged from 87% to nearly 100%. Vital statistics registries provided cause of death, which was known for over 90% of workers who died.
Adequacy of dosimetric records
We reconstructed each worker's dosimetric history using recorded doses from individual facilities or national dose registries. A study of errors in recorded doses evaluated the comparability of dose estimates across facilities and time and identified and quantified sources of bias and uncertainties.9 Doses from higher energy photons (100-3000 keV), which constituted most of the dose in most cohorts, were judged to have been measured in a comparable way over time and across facilities.9 10 The adequacy of practices and technology to measure and record dose from other radiation types (neutrons, internal exposures), however, varied substantially, particularly in earlier years. We therefore excluded workers with potential for substantial doses ( 10% of their whole body dose) from these radiation types.
Main study population
The main study population was defined as workers who had been employed in one or more facilities for at least one year (113 711 workers excluded), who had been monitored for external radiation exposure (38 521 workers excluded), and whose doses resulted predominantly from higher energy photon radiation (39 730 workers with internal contamination and 19 041 with neutron exposures excluded).
Dosimetric errors and derivation of organ doses
The major sources of errors in higher energy photon doses were dosimetry technology, exposure conditions, and calibration practices. Errors from these sources were quantified and bias factors specific to the doses to each organ of interest calculated for each model of dosimeter used and by type of facility (nuclear power plants and mixed activities facilities).11 12 Organ doses were derived by dividing recorded doses by the appropriate organ dose bias factor. We used doses to the colon and active bone marrow for analyses of mortality from all cancers excluding leukaemia and from leukaemia, respectively. All doses are expressed as dose equivalents in sieverts (Sv).
Statistical methods
Analyses were based on a linear relative risk Poisson regression model, in which the relative risk is of the form 1+Z, where Z is the cumulative dose equivalent in Sv and is the excess relative risk per Sv; 95% likelihood based confidence intervals were calculated. We used 11 a priori categories of dose (< 5, 5- < 10, 10- < 20, 20- < 50, 50- < 100, 100- < 150, 150- < 200, 200- < 300, 300- < 400, 400- < 500, 500 mSv). Analyses used only underlying cause of death. Estimates of excess relative risk were stratified for sex, age, and calendar period (both in five year categories), facility, duration of employment (< 10 years, 10 years), and socioeconomic status. In the analyses of all cancers we excluded cohorts for which socioeconomic information was unavailable or incomplete (Japan, Idaho National Engineering Laboratory (INEL), Ontario Hydro Canada), but we included them in analyses of leukaemia as the potential for confounding by socioeconomic status was thought to be less for leukaemia. To allow for a latent period between exposure and death, doses were lagged by two years for leukaemia and 10 years for other cancers, as in other studies of nuclear workers5-7 and assessments of radiation risk.1 2 Sensitivity analyses were conducted with a range of different lags. Attributable risks were estimated by multiplying the excess relative risks by the average dose in the cohort.
We have focused on the main causes of death for which radiation protection committees have provided risk estimates: all cancers excluding leukaemia and leukaemia excluding chronic lymphocytic leukaemia. Chronic lymphocytic leukaemia is excluded because it is thought to be less readily inducible by ionising radiation than other leukaemias.3 We have also presented risk estimates for solid cancers (that is, excluding lymphatic and haematopoietic malignancies) to compare with recent data for A bomb survivors13 and for all cancers excluding leukaemia, lung, and pleural cancers (which have the greatest potential for confounding by smoking, internally incorporated radionuclides, and other occupational carcinogens). We investigated confounding by smoking by separately analysing solid cancers related or unrelated to smoking and two groupings of smoking related outcomes other than cancer (all non-malignant respiratory diseases and chronic obstructive bronchitis and emphysema).
Analysis of data from survivors of A bomb
We analysed mortality data from the A bomb survivors for solid cancer to 199713 and leukaemia to 199014 using similar methods to provide risk estimates for comparison. Analyses were stratified for attained age, city, and calendar time and restricted to men aged 20-60 at exposure, the group most comparable to the workers.
Results
Overall, 598 068 workers were employed in at least one of 154 facilities. Most facilities were involved in nuclear power production; the rest specialised in different activities, including research, waste management, and production of fuel, isotopes, and weapons. The main study population comprised 407 391 workers (table 1). A total of 24 158 (5.9%) people were known to have died during the study period: 6519 from cancers other than leukaemia and 196 from leukaemia excluding chronic lymphocytic leukaemia. The total duration of follow-up was 5 192 710 person years and the total collective recorded dose was 7892 Sv. Most workers in the study were men (90%), and men received 98% of the collective dose. The overall average cumulative recorded dose was 19.4 mSv. The distribution of recorded doses was skewed (fig 1). Ninety per cent of workers received cumulative doses < 50 mSv and less than 0.1% received cumulative doses > 500 mSv.
Table 1 Cohorts included in the 15 country study
Fig 1 Distribution of cumulative radiation doses among workers included in the analyses
For all cancers excluding leukaemia, the excess relative risk was 0.97 per Sv and was significantly different from zero (95% confidence interval 0.14 to 1.97) (table 2). This estimate corresponds to a relative risk of 1.10 for a radiation dose of 100 mSv. For solid cancers, the excess relative risk was 0.87 (0.03 to 1.88), higher than but statistically compatible with the estimate for A bomb survivors (0.32 per Sv). The excess relative risk for leukaemia excluding chronic lymphocytic leukaemia was 1.93 per Sv (< 0 to 8.47), which gives a relative risk of 1.19 for a dose of 100 mSv. This estimate is between the linear and linear quadratic extrapolations from data on A bomb survivors (table 2).
Table 2 Estimates of excess relative risk per Sv (95% confidence interval) for all cancers excluding leukaemia, solid cancers, and leukaemia excluding chronic lymphocytic leukaemia, for nuclear workers and survivors of A bomb in Japan *
Table 3 assesses the possible confounding effect of smoking. Excess relative risks ranged between 0.59 per Sv (-0.29 to 1.70) for all cancers excluding leukaemia and lung and pleural cancer, and 0.91 per Sv (-0.11 to 2.21) for smoking related cancers.
Table 3 Estimates of excess relative risk per Sv (95% confidence intervals) for specific causes of death
The increased risk for smoking related cancers was mainly due to an increased risk of lung cancer (1.86 per Sv, 0.26 to 4.01). Other smoking related cancers showed little evidence of an increased risk (0.21 per Sv, < 0 to 2.01). Risk estimates for mortality from non-malignant respiratory diseases and from chronic obstructive bronchitis and emphysema were raised but not significantly different from zero (excess relative risk per Sv 1.16, -0.53 to 3.84, and 2.12, -0.57 to 7.46, respectively).
Discussion
BEIR V. Committee on the Biological Effects of Ionizing Radiation. The effects on populations of exposure to low levels of ionizing radiation. Washington, DC: National Academy of Sciences, 1990.
International Commission on Radiological Protection ICRP. Recommendations of the international commission on radiological protection. Oxford: Pergamon Press, 1991. (ICRP publication 60.)
United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR). Sources and effects of ionizing radiation. New York: United Nations, 2000.
Berrington de González A, Darby S. Risk of cancer from diagnostic X-rays: estimates for the UK and 14 other countries. Lancet 2004;363: 345-51.
IARC Study Group on Cancer Risk among Nuclear Industry Workers. Direct estimates of cancer mortality due to low doses of ionising radiation: an international study. Lancet 1994;344: 1039-43.
Cardis E, Gilbert ES, Carpenter L, Howe G, Kato I, Armstrong BK, et al. Effects of low doses and low dose rates of external ionizing radiation: cancer mortality among nuclear industry workers in three countries. Radiat Res 1995;142: 117-32.
Muirhead CR, Goodill AA, Haylock RGE, Vokes J, Little MP, Jackson DA, et al. Occupational radiation exposure and mortality: second analysis of the national registry for radiation workers. J Radiol Prot 1999;19: 3-26.
Cardis E, Estève J. International collaborative study of cancer risk among nuclear industry workers. I—Report of the feasibility study. II—Protocol report 92/001. Lyons: International Agency for Research on Cancer, 1992.
Cardis E, Martuzzi M, Amoros E. International collaborative study of cancer risk among nuclear industry workers. III—Procedures document, rev 1. Lyons: International Agency for Research on Cancer, 1997. (97/002.)
Fix JJ, Salmon L, Cowper G, Cardis E. A retrospective evaluation of the dosimetry employed in an international combined epidemiological study. Radiat Protect Dosimetry 1997;74: 39-53.
Thierry-Chef I, Pernicka F, Marshall M, Cardis E, Andreo P. Study of a selection of 10 historical types of dosemeter: variation of the response to Hp(10) with photon energy and geometry of exposure. Radiat Prot Dosimetry 2002;102: 101-13.
Thierry-Chef I, Cardis E, Ciampi A, Delacroix D, Marshall M, Amoros E, et al. A method to assess predominant energies of exposure in a nuclear research centre—Saclay (France). Radiat Prot Dosimetry 2001;94: 215-25.
Preston DL, Shimizu Y, Pierce DA, Suyama A, Mabuchi K. Studies of mortality of atomic bomb survivors. Report 13: Solid cancer and noncancer disease mortality: 1950-1997. Radiat Res 2003;160: 381-407.
Pierce DA, Shimizu Y, Preston DL, Vaeth M, Mabuchi K. Studies of the mortality of atomic bomb survivors. Report 12, Part I. Cancer: 1950-1990. Radiat Res 1996;146: 1-27.
International Agency for Research on Cancer. IARC monographs on the evaluation of carcinogenic risks to humans. Vol 83. Tobacco smoke and involuntary smoking. Lyons: IARC, 2004.
Lee DJ, LeBlanc W, Fleming LE, Gomez-Marin O, Pitman T. Trends in US smoking rates in occupational groups: the national health interview survey 1987-1994. J Occup Environ Med 2004;46: 538-48.
Gribbin MA, Weeks JL, Howe GR. Cancer mortality (1956-1985) amongst male employees of Atomic Energy of Canada Limited with respect to occupational exposure to external low linear energy transfer ionizing radiation. Radiat Res 1993;133: 375-80.
Murata M, Miyake T, Inoue Y, Ohshima S, Kudo S, Yoshimura T, et al. Life-style and other characteristics of radiation workers at nuclear facilities in Japan: base-line data of a questionnaire survey. J Epidemiol 2002;12: 310-9.
Peterson G, Gilbert ES, Buchanan JA, Stevens RG. A case-cohort study of lung cancer ionizing radiation, and tobacco smoking among males at the Hanford sites. Health Phys 1990;58: 3-11.
Carpenter L, Fraser P, Booth M, Higgins C, Beral V. Smoking habits and radiation exposure. J Radiol Protection 1989;9: 286-7.
Auvinen A, Pukkala E, Hyv?nen H, Hakama M, Ryt?maa T. Cancer incidence among Finnish nuclear reactor workers. J Occup Environ Med 2002;44: 634-8.(E Cardis, head of radiation group1, M Vr)