Objectives Suicide is a leading cause of death in the working-age population. We investigated suicide mortality among diagnostic medical radiation workers in relation to their demographic and occupational factors in South Korea.
Methods The study population consisted of all diagnostic medical radiation workers enrolled in the National Dosimetry Registry from 1996 to 2011. The registry data were linked with mortality data through the end of 2017. We calculated age-standardised suicide rates, standardised mortality ratios (SMRs) and rate ratios (RRs) by demographic and occupational factors.
Results A total of 207 suicides were identified among 94 367 medical radiation workers, exhibiting a suicide rate of 14.0 per 100 000 person-years. Compared with the general population, suicide rates were lower for both male and female workers (SMR 0.49, 95% confidence interval (CI) 0.42 to 0.57; SMR 0.54, 95% CI 0.41 to 0.70, respectively). Similarly, decreased SMRs were observed across job titles and other work-related factors. However, a shorter duration of employment was positively associated with RRs for suicide; risks were 2.74 (95% CI 1.56 to 4.81) and 4.66 (95% CI 1.53 to 14.20) times higher in male and female workers with less than 1 year of employment, respectively, than in those with at least 10 years of employment.
Conclusions Diagnostic medical radiation workers in South Korea showed lower suicide rates than in the general population. However, a shorter duration of employment was associated with higher risk of suicide. Suicide prevention efforts could target workers engaged in short-term employment.
- health care workers
- mortality studies
Statistics from Altmetric.com
What is already known about this subject?
Prior studies in a few countries have shown higher suicide mortality rates in medical workers than in the general population. However, these findings were inconsistent.
Few studies have investigated the association between suicide and work-related factors in specific occupations.
What are the new findings?
Diagnostic medical radiation workers in South Korea had lower suicide rates than in the general population; the results were consistent across a range of demographic and job-related factors.
However, a shorter duration of employment was associated with higher risk of suicide among diagnostic medical radiation workers with the highest risk observed among workers having less than 1 year of employment.
How might this impact on policy or clinical practice in the foreseeable future?
The findings suggest that short-term workers are a vulnerable group that suicide prevention efforts should target.
Strategies such as providing adequate support for short-term workers and improving working conditions may contribute to suicide prevention in the workplace.
Suicide is a leading cause of death in the working-age population globally.1 Suicide rates have been reported to vary across occupations.2–4 Some studies have identified medical workers (eg, doctors, nurses and dentists) as a high-risk group for suicide5–8; however, the results have not been consistent.9–14 Medical workers are subject to various occupational hazards such as work overload, stress, workplace violence, shift work, exposure to neurotoxic agents and increased knowledge of lethal suicide methods, which may increase suicide risk.15 16 Nevertheless, a few previous studies have investigated job-related risk factors of suicide in medical workers.
Among the Organisation of Economic Cooperation and Development member countries, South Korea had the highest suicide rate in 2017.17 Furthermore, suicide is one of the leading causes of death in South Korea.18 A recent study based on a cohort of more than 90 000 diagnostic medical radiation workers in South Korea found suicide to be the third leading cause of death in this group.19 Given that the number of medical radiation workers has increased rapidly in South Korea20 and suicide is a major cause of death in these workers,19 there is a need to investigate the factors related to suicide in this occupational group.
The purpose of this study, therefore, was to extend the investigation of suicide mortality rates among South Korean diagnostic medical radiation workers compared with those in the general population with the latest updated mortality data and to examine whether suicide was associated with demographic and occupational factors by internal comparison. Considering the rapid expansion of the medical workforce, the findings can inform strategies to prevent suicide among medical workers.
The study population encompassed all 94 395 diagnostic medical radiation workers enrolled in the National Dosimetry Registry (NDR) in South Korea from 1996 to 2011. The NDR has been the government-operated centralised dosimetry data registry for all diagnostic radiation workers since 1996.20 Information contained in the NDR includes name, sex, personal identification number, job title, quarterly dose data, and the beginning and end dates of the measurement period. Based on the NDR job classification codes, workers were grouped into seven categories: radiologists, non-radiologist physicians, dentists, dental hygienists, radiologic technologists, nurses and others. After excluding workers with invalid data (n=28), 94 367 workers were included in the analysis. Workers who died at over 90 years of age (n=27) were followed up only to 89 years of age due to the potential reduced reliability of cause of death after this age.
Ascertainment of vital status
To ascertain the cause of death among the study population, their personal identification numbers were sent to Statistics Korea, which then linked these numbers to the mortality data with coverage through the end of 2017. This linkage is highly specific due to the uniqueness of individual personal identification numbers in South Korea; Statistics Korea maintains an almost 100% level of completeness and combines the data from the death certificates issued by physicians and 22 other sources of administrative data such as autopsy records, emergency records, criminal and traffic accident investigation data to complete a final dataset.18 The underlying causes of death were coded according to the International Classification of Diseases and Related Health Problems, 10th Revision (ICD-10). Suicides were identified using ICD-10 codes X60–X84.
Ascertainment of occupational factors
The workers were grouped into seven job categories based on the NDR information. Medical facility locations were classified as either metropolis (ie, Seoul and six other metropolitan cities) or province according to governmental administrative divisions. Duration of employment was based on the period between the beginning and end dates of badge measurement for workers enrolled after 1996. For workers enrolled before or in 1996, the start year of radiologic practice for each sex and occupational group was imputed using the mode of the age of job start from our previous survey among diagnostic medical radiation workers.21 The cumulative individual badge doses based on dose at a tissue depth of 10 mm from the dosimeter were calculated by combining the quarterly badge readings for the workers enrolled in the NDR. For workers who started before or in 1996 (n=15 171; 16% of the total population), historical badge doses were reconstructed using a model in which annual doses were determined as a log-linear function of time and age.22
Each worker’s follow-up period started in 1996 or the start year of work based on the NDR, whichever occurred later. The end of the follow-up period was the date of death, or age 89, or the end date of the study (31 December 2017), whichever occurred earliest. The DATAB module in Epicure (Risk Sciences International, V.2.0, Ottawa, Canada) was used to create a person-year table stratified by sex, attained age (<25, 5 years intervals from age 25–89 years), calendar year (1996–2000, 2001–2005, 2006–2010, 2011–2017), year of birth (<1960, 1960–1969, 1970–1979, ≥1980), year of entry (1996–2000, 2001–2005, 2006–2011), job title (the seven categories given above), type of medical facility (hospital, clinic, others), area of medical facility (metropolis, province), duration of employment in years (<1, 1–4, 5–9, ≥10) and cumulative radiation badge doses (<0.5, 0.5–0.9, 1.0–2.4, 2.5–9.9, ≥10.0 mSv).
Crude suicide rates per 100 000 person-years, overall and by sex and job title, were calculated by dividing the number of suicide deaths by the person-years in the corresponding groups. Directly age-standardised suicide rates were also calculated by sex for each year using WHO’s World Standard Population.23 Standardised mortality ratios (SMRs) for suicide were calculated as the ratio of observed to expected deaths under the assumption that the observed suicide follows a Poisson distribution. The expected number of suicide cases for each cell was computed as the product of the number of person-years and the age-specific, sex-specific and calendar year-specific South Korean suicide rates (http://kosis.kr/eng/).
Rate ratios (RRs) for potential risk factors of suicide were calculated using Poisson regression models. The 95% confidence intervals (CIs) of RRs were obtained using the maximum likelihood method with the AMFIT module of Epicure. We modelled age and calendar year as continuous variables, and adding a quadratic term of age to the model did not improve the model fit. Akaike information criterion (AIC) was used to identify the best fit model. The final model included age based on the AIC, and additionally calendar year to control for any time trends. Analyses were stratified by sex given the marked difference in suicide rates between males and females.24 Sensitivity analyses were conducted by including only workers who started practice after 1996. The variations in suicide rates and associated occupational characteristics between physicians (ie, radiologists, non-radiologist physicians and dentists) and non-physicians (ie, radiologic technologists, nurses, dental hygienists and others) were also examined because different socioeconomic positions might be associated with suicide risk.
A total of 207 suicides (156 males, 51 females) were identified among all diagnostic medical radiation workers in South Korea during the study period (table 1). The age-standardised and sex-standardised suicide rate for all workers was 14.0 per 100 000 person-years; suicide rates were approximately 2.5 times higher in male workers (20.1 per 100 000 person-years) than in female workers (7.8 per 100 000 person-years). Moreover, suicide rates varied according to job title; relatively high rates were observed in radiologic technologists, dentists and radiologists, in contrast to relatively low rates in dental hygienists and nurses. Trends in suicide rates in diagnostic medical radiation workers were generally similar to those of the general population during the study period, with a peak in 2008–2011, although the suicide rate in the study population was lower than the general population in both male and female workers (figure 1).
SMRs for suicide were reduced in both male (SMR 0.49, 95% CI 0.42 to 0.57) and female workers (SMR 0.54, 95% CI 0.41 to 0.70) compared with the general population (table 2). Reduced SMRs of suicide were consistently observed among diagnostic medical radiation workers across different strata of age at the baseline, calendar year, year of birth, year of job entry, job title, type and area of medical facility, duration of employment and cumulative occupational radiation doses. SMRs for suicide were close to null among male and female workers employed less than 1 year but significantly decreased with increased duration of employment.
Among the diagnostic medical radiation workers, RRs of suicide did not differ across strata of individual variables, with the exception of employment duration (table 3). RRs of suicide were higher with shorter durations of employment; suicide rates were 2.74 (95% CI 1.56 to 4.81), 1.89 (95% CI 1.22 to 2.92) and 1.82 (95% CI 1.20 to 2.76) times higher among male workers employed less than 1 year, 1–4 years and 5–9 years, respectively, compared with those employed 10 or more years. The pattern was similar for female workers; corresponding RRs of suicide in female workers were 4.66 (95% CI 1.53 to 14.20), 2.31 (95% CI 0.81 to 6.62) and 1.33 (95% CI 0.42 to 4.22), respectively.
The sensitivity analysis restricted to workers employed after 1996 (n=79 196) showed similar results (online supplementary table 1). Stratified analyses according to job title (physicians vs non-physicians) showed similar findings as well; the association between shorter employment duration and higher RRs of suicide were similarly observed in both physicians and non-physicians (online supplementary table 2).
Our findings indicate lower suicide rates in South Korean diagnostic medical radiation workers than in the general population. Reduced suicide rates were observed in both male and female workers and were consistently found across job titles and other work-related factors. However, employment duration was inversely associated with suicide rates; the highest RR of suicide was found in workers employed for less than 1 year. Findings were consistent in sensitivity analyses restricted to workers employed after 1996 and in stratified analyses by job title (physicians vs non-physicians). These findings suggest that short-term workers are a vulnerable group who should be targeted for suicide prevention efforts. Although the suicide rates among diagnostic medical radiation workers were lower than in the general population, strategies such as providing adequate support for short-term workers and improving working conditions may contribute to suicide prevention in the workplace.
The increased suicide risk in diagnostic medical radiation workers employed for less than 1 year was consistent with findings from other occupational cohorts that showed elevated risks for all external causes of death such as injury and suicide among short-term blue-collar workers.25–27 Possible explanations including health-related selection processes such as pre-employment hospitalisation and unhealthy lifestyles such as alcohol abuse among short-term workers28 could also be applicable to this study population. Although a few studies have examined the overall mortality in short-term blue-collar workers, few have focused on suicide mortality among white-collar workers in previous cohort studies. Furthermore, new employees may face difficulties in adjusting to work conditions, and poor mental health may contribute to an increased suicide risk. Therefore, job adjustment and support programmes for new workers, lifestyle and mental health screenings, and appropriate referrals for support or treatment may be effective in preventing suicide in short-term workers. Workplace suicide prevention activities have proved effective in reducing suicide, although the findings were based on a small number of studies.29 To better understand specific suicide risk factors (both pre-employment and post-employment factors such as vulnerability and job stress) that are potentially modifiable in short-term workers and inform prevention strategies, further investigations are needed.
Higher suicide rates in male than female diagnostic medical radiation workers across all job titles in this population were consistent with the hypothesis that males exhibit generally higher risk of suicide due to their vulnerability to socioeconomic changes and risk-taking behaviours.30 31 However, female physicians showed higher RRs than male physicians, although the absolute rates were smaller than in male physicians in other countries.5 8 Possible gender effects modifying the association between occupation and suicide risk32 merit further investigations.
The reduced suicide risk in our diagnostic medical radiation workers compared with the general population could be due to the healthy worker effect33; namely, healthy individuals are more likely to seek, gain, retain employment than less healthy individuals are. Furthermore, the relative socioeconomic advantage of medical radiation workers could mitigate the adverse effect of other suicide risk factors. Our findings of SMRs (ie, close to null at the beginning of work and significant reduction with increased work duration) may imply that the healthy survivor effect would have a greater impact than the healthy selection effect on suicide risk among South Korean diagnostic medical radiation workers. Although the SMR findings have the limitation of indicating the specific risk factors, they would be informative in determining overall suicide risks and helpful in assessing the nature of suicide risks among diagnostic medical radiation workers.
Low SMRs for suicide among male workers in this study are comparable to those of doctors in the USA,10 England and Wales,9 and Taiwan12; however, the findings of low SMRs in female workers differed from those in the USA,10 England and Wales,9 and Norway,11 which showed higher SMRs among female workers than the general population. Potential heterogeneity among studies, however, prevents a direct comparison of SMRs due to the difference of study periods and sociocultural backgrounds in each country. A recent review reported that the risk of suicide in physicians is not homogeneous between countries, as working conditions vary between countries and across times.8
The strength of this study lies in its relatively large population size. Moreover, the inclusion of all diagnostic medical radiation workers in South Korea and data recorded in the NDR make it possible to investigate the patterns of suicide according to various demographic and work-related characteristics. There are, however, several limitations to the study. First, there is a lack of information about suicide methods and specific risk factors for suicide such as mental disorders and alcohol abuse. Second, the index of employment duration might be an imperfect indicator because some workers might have moved to or from other departments or job positions where no radiation exposure exists; thus, their duration of employment could be underestimated. However, the study population comprised specialised professionals working in specific departments; hence, such changes in job positions are expected to be infrequent. Third, suicide rates may be underestimated due to the misclassification of suicide as other causes of death such as undetermined intent, as shown in previous studies.34 35 However, the potential underestimation could be non-differential, and the SMRs and RRs would not be affected because of their relative nature.
In summary, our findings showed a decreased suicide rate in South Korean diagnostic medical radiation workers than in the general population. However, within this group, suicide risk was greater in workers with a shorter duration of employment. Our findings contribute to a better understanding of the overall suicide risk and associated characteristics in diagnostic medical radiation workers. Further investigation is needed to better understand specific risk factors for suicide in short-term workers to inform effective suicide prevention strategies.
Contributors WJL conceptualised the research and wrote the first draft of the manuscript. ESC and YJB performed the data analyses and interpreted the findings with WJL. S-SC and C-YH provided advice on the data analyses and critically revised the manuscript. All authors contributed to the draft revision and approved the final manuscript.
Funding This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (no. 2020R1A2C1008891).
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval This study was reviewed and approved by the Institutional Review Board of Korea University (KUIRB-2019-0092-03).
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement No data are available.
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.