Objectives To evaluate the risk for all-cause and cause-specific mortality in diagnostic medical radiation workers in South Korea.
Methods The study population included all diagnostic medical radiation workers enrolled in the National Dosimetry Registry (NDR) between 1996 and 2011. NDR data were linked with mortality data obtained from national registries through 2015. Standardised mortality ratios (SMRs) and relative standardised mortality ratios (rSMRs) were calculated for external comparison and for adjustment of the cohort’s overall healthiness.
Results A total of 1099 deaths (974 in men and 125 in women) were reported from among 80 837 medical radiation workers. The SMRs for all causes of death were significantly lower than expected in both men (SMR 0.45, 95% CI 0.42 to 0.48) and women (SMR 0.49, 95% CI 0.41 to 0.58). No excesses were observed for any specific cause of death. The findings were similar by job title, calendar year of entry and year of birth. However, relative to all causes of death, mortality from all cancers (rSMR 1.60, 95% CI 1.41 to 1.82), leukaemia, colon cancer, stomach cancer and diseases of the circulatory system increased significantly among male workers. The results for female workers were limited due to small number of deaths; however, the rSMR for all cancers was significantly elevated (rSMR 1.70, 95% CI 1.17 to 2.46).
Conclusions This cohort showed lower mortality among diagnostic medical radiation workers than in the general population. However, occupational factors may have been involved in the increased relative mortality for several causes of death.
- healthy worker effect
- occupational exposure
Statistics from Altmetric.com
What is already known about this subject?
Studies of medical radiation workers have reported lower mortality than that in the general population in a few countries.
There has been a rapid increase in the number of medical radiation workers and changes in their working environments.
What are the new findings?
Lower all-cause and specific-cause mortality rates have been observed among South Korean diagnostic medical radiation workers; the results were similar by job title, year of entry and year of birth.
Relative to all causes of death, however, the mortality rates for all cancers, leukaemia, cancers of the stomach and the colon and diseases of the circulatory system were significantly increased among male workers.
How might this impact on policy or clinical practice in the foreseeable future?
Although medical radiation workers have a more favourable survival than the general population, occupational factors that may relate to an increased relative mortality require further investigation.
Medical radiation workers, a rapidly increasing professional group due to the expanding use of modern medical practices, are routinely exposed to occupational radiation.1 Epidemiological studies of these workers have reported decreased rates of all-cause mortality and all cancers among radiologic technologists in the USA and Japan and Canadian medical workers compared with those among the general population, whereas there is increased mortality for all cancers among radiologists in the USA and the UK compared with those among other physicians.2
However, few recent cohort studies have assessed medical radiation workers and the healthy worker effect, which could mask risks associated with occupational exposures, has seldom been evaluated in medical workers. As part of an extended diagnostic medical radiologic technologists’ study in South Korea,3 we constructed a registry-based cohort by merging National Dose Registry (NDR) and mortality data among all diagnostic medical radiation workers.
The purpose of this study was to evaluate mortality among South Korean diagnostic medical radiation workers compared with that of the general population and to examine the healthy worker effect in this population.
The study cohort comprised all 94 396 diagnostic medical radiation workers enrolled in the NDR from 1996 to 2011. The NDR is a government-operated centralised dosimetry data registry for all diagnostic radiation workers and has been in operation since 1996 by the Korea Centers for Disease Control and Prevention. Based on NDR work history classification codes, the workers were grouped into seven categories: radiologists (n=1520), doctors (n=18 684), dentists (n=15 705), dental hygienists (n=13 488), radiologic technologists (n=26 356), nurses (n=7561) and other medical assistants (n=11 082). Medical workers involved with nuclear medicine and therapeutic departments are not included in this system. Of the total workers in the database, those had been employed for <1 year (n=13 541), had died in the year of entry (n=15) or had incomplete personal identification numbers (n=3) were excluded; thus, 80 837 workers were included in the analysis. Workers who died at age >85 years (n=53) were followed-up only until 85 years of age due to the potential reduced reliability of cause of death after this age.
To ascertain the cause of death among study participants, personal identification numbers were sent to Statistics Korea, who then linked these numbers to the mortality data. This linkage is highly specific because of the uniqueness of the individual personal identification number in South Korea; Statistics Korea has maintained an almost 100% level of completeness (http://kostat.go.kr). Mortality data were classified according to the underlying cause of death based on the International Classification of Diseases and Related Health Problems, 10th Revision (ICD-10).
Standardised mortality ratios (SMRs) were calculated as the ratio of observed to expected deaths using Epicure.4 Each worker’s follow-up period started from 1996 or from the starting year of work based on the NDR, whichever occurred later. The starting date was assumed to be 1 July of the starting year due to the lack of exact entry date information. The end of the follow-up period was the earlier of the date of death or the ending date of the study (31 December 2015). Using the DATAB module in Epicure, a person-year table was stratified by sex, attained age (<30 and 5-year intervals from age 30 to 85), year of birth (<1960, 1960–1969, 1970–1979, ≥1980), year of entry (<2001, 2001–2005, >2005) and job title (seven categories, as described above). The expected all-cause and specific-cause deaths were computed by multiplying the person-years in each sex-specific, age-specific and calendar year-specific stratum using South Korean mortality rates from 1996 to 2015. Stratified analyses were conducted by sex, job title, year of entry and birth year. For statistical reasons, only causes of death characterised by 10 or more cases for each sex were included. To adjust for the healthy worker effect, we also calculated the relative SMR (rSMR) as the ratio of the cause-specific SMR to the SMR for the other causes, omitting the cause of interest. The equation for rSMR is as follows: rSMR x =SMR x /SMRnot x.5
Of 1 003 311 person-years, 1099 deaths were reported among diagnostic medical radiation workers during the study period in South Korea (table 1). SMR analysis of all-cause mortality revealed a significant deficit for both men (SMR 0.45, 95% CI 0.42 to 0.48) and women (SMR 0.49, 95% CI 0.41 to 0.58) compared with the national rates. Most of the causes of death were statistically significantly decreased and no significant excesses were observed for any specific cause of death. Analyses of the sex-specific cancer mortality results were similar; however, the other results among women were limited because of the small numbers of deaths.
The rSMRs were significantly elevated for all cancers (rSMR 1.60, 95% CI 1.41 to 1.82), leukaemia, cancers of the stomach and the colon and diseases of the circulatory system including hypertensive heart disease and ischaemic heart disease, whereas the rSMRs for diseases of the digestive system and accidents were significantly lower than expected among male workers. For female workers, the rSMR was significantly elevated for all cancers (rSMR 1.70, 95% CI 1.17 to 2.46).
Stratified analyses for all causes of death and all cancers by job title, year of entry and birth year did not show any particular pattern (online Supplementary tables 1 and 2).
This first mortality analysis among South Korean medical radiation workers indicates a more favourable survival than that of the general population. These deficits were observed for all causes of death among men and women and were consistent according to job title, year of entry and birth year. However, after adjusting for the lower overall mortality of the cohort, male workers experienced higher mortality from all cancers, leukaemia, cancers of the stomach and the colon and diseases of the circulatory system. Our findings imply that medical radiation workers may have certain occupational hazards related to their increased relative mortality although they were healthier than the general population.
The lower overall mortality in this study compared with that of the general population, a typical phenomenon in occupational cohorts, was slightly greater than that of medical radiation workers in the USA, UK, Canada and Japan.2 Compared with other studies, the relatively young age of the workers and the short follow-up of our cohort may have contributed to this difference, although heterogeneity among studies prevents direct comparison of SMRs. The lack of particular patterns in stratified analyses from the year of entry and the birth year may also be related to those of our cohort characteristics. The pattern of a stronger healthy worker effect for non-malignant diseases than that for cancer was generally consistent with typical findings in occupational epidemiology6; however, the SMRs for specific causes of death require further follow-up. The SMRs from diseases of the digestive system and non-transport accidents showed the lowest estimates; these may be related to a relatively higher socioeconomic status in medical workers than that in the general population.
However, several specific causes of death were increased after accounting for the lower overall mortality in the cohort. The switch from deficit SMR to excess rSMR has also been observed in other occupational populations and has been interpreted as a possible result of occupational factors.5 7 In our study, the increased rSMRs in a few causes of death may be related to a variety of occupational risk factors at hospitals including physical, chemical and biological hazards.8 Because medical radiation workers in several countries have significantly elevated mortality rates for all cancers and leukaemia compared with those in other medical workers,2 occupational radiation exposure may be an important potential risk factor that should be considered related to health outcomes. Further studies with detailed information on radiation exposure are warranted. South Korean medical radiation workers reportedly receive a generally low-dose of occupational radiation; however, several hundred workers have received more than 5 mSv per quarter every year.9 The increased rSMRs of stomach and colon cancer may relate to lifestyle factors; however, this requires further investigation along with more detailed information on diet.
This study has strength in its relatively large size, involving all working diagnostic medical radiation workers from 1996 to 2011 in South Korea and reflecting current working conditions. The study weaknesses include the limited information on work history and lifestyle factors due to the registry linkage design, a short follow-up and the small number of deaths, particularly among women. In addition, our calculation of rSMR is not a direct method to account for the lower overall mortality.
In summary, we observed that South Korean medical radiation workers have decreased all-cause mortality compared with that of the general population; however, their increased relative mortality for a few causes of death may suggest occupational hazards. An extension of this study linking cancer incidence and developing occupational exposure indices is under way. Further investigations are warranted to clarify the findings in medical radiation workers.
The authors thank Dr Dale L Preston for helpful advice on the use of Epicure for SMR estimation.
Contributors WJL developed the idea for the research and wrote the first draft of the manuscript. SK and YJB performed the data analyses. ESC and KML provided advice on analysis of data and revised the manuscript. WJL incorporated the comments and wrote the final revision. All authors contributed to the draft revision and approved the final manuscript.
Funding This work was supported by a grant from the Korea Health Technology R&D Project, through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (grant number: HI16C1186).
Competing interests None declared.
Patient consent Not required.
Ethics approval Institutional Review Board of Korea University (1040548-KU-IRB-16-203-A-1).
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.