The association between Korean employed workers’ on-call work and health problems, injuries

Background On-call work is a form of work that requires the person to work at any time during the on-call period. Thus, on-call work is often regarded as one of the most severe stress factors. This study investigates the associations between on-call work and health problems, injuries. Methods This study was based on the 3rd Korean Working Conditions Survey. Total of 29,246 employed workers who had been working for at least 1 year were included. Logistic regression analysis was performed to investigate the association between on-call work and health problems, injuries. Results The odds ratios for on-call workers in terms of physical health problems, psychological health problems, and injuries were 1.33 (95% confidence interval [CI] 1.22-1.44), 1.31 (95% CI 1.08-1.60), and 2.76 (95% CI 2.26-3.37), respectively. Analysis of the detailed symptoms revealed odds ratios in on-call workers of 2.06 for hearing problems (95% CI 1.63-2.62); 1.71 for skin problems (95% CI 1.38-2.12); 1.22 for back pain (95% CI 1.08-1.38); 1.23 for muscular pains in upper limbs (95% CI 1.12-1.34); 1.27 for muscular pains in lower limbs (95% CI 1.15-1.40); 1.46 for headache, eye fatigue (95% CI 1.32-1.60); 1.37 for abdominal pain (95% CI 1.02-1.85); 1.43 for depression or anxiety disorders (95% CI 1.07-1.93); 1.36 for fatigue (95% CI 1.24-1.49); and 1.41 for insomnia and general sleep difficulties (95% CI 1.13-1.76). Conclusions The present study found that on-call work results in an increased risk of health problems and injuries. This study is the result of analyses of broad range of the job spectrum in Korean employed workers; thus, future studies are necessary to determine the effects of on-call work in various job groups.


Background
On-call work, by definition, is a form of work that requires the person to work at any time during the on-call period. Therefore, a person often has to work during both daytime and nighttime hours, leading to stressful situations for the worker [1]. Thus, on-call work is often regarded as one of the most severe stress factors [2][3][4]. This was reflected in a European working conditions survey (EWCS) performed to establish the foundation for creating a healthy work environment; on-call work assessment was first introduced in the 5th European EWCS performed in 2010. Similarly, in Korea, assessment of on-call work was first included the 2nd Korean Working Conditions Survey in 2010, which defined on-call work as "immediately providing work or service if contacted or called". The present study also used this definition.
Previous studies have reported the various effects of oncall work on workers. The health-related effects of on-call work include stress-related symptoms such as exhaustion, irritation, sleep disorders, memory disorders, and headache [4]. A previous study on nurses reported that on-call work was associated with musculoskeletal symptoms including pain in back and muscles of the upper limbs [5]. The neuropsychological effects of on-call work include depressed mood [6], sleep deprivation and insomnia [7], sleepiness during the daytime [7,8], reduced cognitive function [9,10], and difficulty in concentrating [10]. Furthermore, a study on residents showed that long on-call work periods were associated with occupational injuries and malpractice [11].
On-call work may also affect socio-psychological stress. A previous study of Korean physicians showed an association between the frequency of on-call duty during the week and work-related stress [12]. Another study showed the relationship between on-call work and job dissatisfaction [2,13] and also between on-call work and turnover intention [14], suggesting that job dissatisfaction from on-call work can affect turnover intention. In the US, the annual loss of operating budget due to job change is an estimated 5% [15]. Therefore, increased turnover of employees from on-call work may also become a social cost problem. Moreover, irregular working hours due to on-call work is associated with work-life imbalance [16], and a study of on-call workers and their spouses demonstrated the negative effects of on-call work on family life including constraints in their lives, forced sacrifice by spouses, and communication issues among family members [17]. These results clearly showed that on-call work affects not only work life but also life outside of work. In addition, the negative effects of on-call work on other family members were clearly demonstrated. Thus, on-call work not only has negative effects both physically and psychologically but may also result in socio-psychological stress, eventually affecting the quality of life of workers and their family members in addition to the social costs.
However, there are very few studies about on-call work in Korea. Moreover, previous studies generally focused on specific occupations (i.e., nurses and physicians) and no previous study on the effect of on-call work on health and other aspects has included a broad range of the job spectrum. This study aimed to investigate the possible associations between on-call work and its possible negative effects on health and other injuries by utilizing data from the 3rd Korean Working Conditions Survey performed in 2011 regarding on-call work-related personal and occupational characteristics and work environment in a large sample of workers representative of Korean workers nationwide.

Study population
This study was based on the 3rd Korean Working Conditions Survey performed by Korean Occupational Safety and Health Research Institute in 2011. The survey was conducted on a representative sample of workers (≥15 years). A worker was defined as "a person who during the reference week did any work for pay or profit". A total of 50,032 subjects were included in the study, and in-person interviews were performed by professional interviewers from professional survey services after obtaining informed consent. This study focused on employed workers from the entire of 50,032 subjects. An employed worker was defined as "a person who signed either expressed or implied employment contract with individual, family, or business and is receiving salary or daily wage or spot goods in return for their labor". Furthermore, since we aimed to investigate the effect of on-call work on the workers, only the employed workers who had been working for at least 1 year were included. Eventually, a total of 29,246 subjects were included in this study. Although the age limit for the Working Conditions Survey was 15 years, since the subjects included in this study had been working for at least 1 year, the subjects in this study were at least 16 years of age.

On-call work
The subject was defined as working on-call if they responded "Yes" to the question "Do you work on-call (immediately providing work or service if contacted or called)?"

Health problems and injuries
For health problems and injuries, the subject was defined as having health problems and injuries if they answered "Yes" to the question "Over the last 12 months, did you suffer from any of the following health problems?". The questionnaire contained sub-categories for questions regarding health problems and injuries including hearing problems, skin problems, back pain, muscular pains in upper limbs, muscular pains in lower limbs, headaches or eyestrain, abdominal pain, respiratory difficulties, cardiovascular diseases, injuries, depression or anxiety, fatigue, insomnia and general sleep difficulties. Any report of hearing problems, skin problems, backache, muscular pains in upper limbs, muscular pains in lower limbs, headaches or eyestrain, abdominal pain, respiratory difficulties, cardiovascular diseases, fatigue was defined as a physical health problem. Similarly, if the subject had depression or anxiety, insomnia, general sleep difficulties, the subject was defined as having a psychological health problem.

General and occupational characteristics
The general characteristics of the subjects investigated in this study included gender, age, education level, income level, alcohol consumption, smoking, obesity, and hypertension. For obesity and hypertension, the subject was defined as having a chronic condition if they answered "Yes" to the question "Were you ever diagnosed with chronic obesity or hypertension by your physician?". However, for obesity, if the subject answered "Not obese" to the additional question "What is your current status in terms of obesity?", the subject was not considered obese.
The occupational characteristics included job type, employment type, working hours, and shift work. Two methods were used to categorize the job types. First, the subjects were asked, "What is the job type that best describes your current occupation?". Based on their responses, professional workers and senior managers were defined as "Professional", general office workers as "Office", sales workers and service providers as "Service", and skilled/semi-skilled/unskilled occupations and agricultural/ forestry/fishery workers as "Physical". Second, based on the middle categorization of the Korean Employment Classification of Occupation (KECO), the subjects' occupations were categorized into 24 job groups. Employment type was defined as regular or temporary, and working hours were categorized as ≤ 40 h, 41 -60 h, or > 60 h per week.

Work environment and work-related stress
Work environment was assessed in terms of exposures to physical, chemical, and ergonomic factors. If the subject answered "All of the time", "Almost all of the time", "Around 3/4 of the time", "Around half of the time" or "Around 1/4 of the time" to the question "Are you exposed at work to ...?", the subject was considered to be exposed to the factor. However, if the subject answered "Almost never" or "Never" to the same question, the subject was not considered to be exposed to the factor. The physical factors included the vibrations, noise, high temperatures, and low temperatures. The chemical factors included breathing in smoke, fumes, powder or dust, vapors such as solvents and thinners, handling or being in skin contact with chemical products or substances, and tobacco smoke from other people. Lastly, the ergonomic factors included tiring or painful positions, lifting or moving people, carrying or moving heavy loads, standing, and repetitive hand or arm movements.
For work-related stress, if the subject answered "Always", "Most of the time", or "Sometimes" to the question "For each of the following statements, please select the response which best describes your work situation -You experience stress in your work.", the subject was defined as having work-related stress. If the answer was "Rarely" or "Never", the subject was defined as being free of work-related stress.

Data analysis
This study utilized IBM SPSS Statistics for Windows version 19.0 (IBM Corp: Armonk, NY, USA) for analysis of the data after applying weighting adjustments. Personal and occupational characteristics, work environment, and work-related stress were analysed using frequency analysis. Chi-square analysis was used to determine the associations between on-call work and characteristics of the subject and on-call work and health problems and injuries. In order to determine the risk of health problems and injuries due to on-call work, bivariate logistic regression analysis was performed to calculate the odds ratios. Model I was the analysis after adjusting for personal characteristics, while Model II was additionally adjusted for occupational characteristics including job type, employment type, working hours, and shift work. Model III further adjusted for physical, chemical, and ergonomic factors of the work environment, while Model IV had additional adjustment for work-related stress in addition to all controls from Model III.

Results
General and occupational characteristics of the study participants A total of 29,246 subjects were included in this study. Frequency analysis and descriptive statistics were used to analyze the personal and occupational characteristics, work environment, and work-related stress. The personal characteristics of the subjects included the following: 60.9% of the subjects were male; 59.1% were between 30 and 49 years of age; 55.7% had education levels higher than college graduation; 53.5% were non-smokers; 51.5% drank alcohol once a week or less; 98.2% were not obese; and 95.3% were not diagnosed with hypertension. A majority of the subjects were "physical" workers (33.6%), and only 8.7% were "professional" workers. Moreover, a majority of the subjects were regular employees, and 51.5% worked between 41 and 60 h per week, while 39.6% worked 40 h or less per week. Most of the subjects were not working on a rotating basis (i.e., no shift work). More than half of the subjects responded that they did not have exposure to physical, chemical, or ergonomic factors in their work environment, but 72.1% reported that they suffered from work-related stress.
Difference of general occupational characteristics according to on-call work 2722 subjects (9.3%) were on-call workers; the remaining 26,524 (90.7%) were not on-call workers. Chi-square analysis was used to analyze the differences in each factor according to the on-call work status. The results revealed significant differences between on-call workers and nonon-call workers in all factors except for hypertension and working hours. Assessment of the general characteristics revealed a higher rate of on-call workers in the following groups: male, ≥50 years, high school graduates, monthly income ≥2 million won, smokers, and obese subjects. Among job types, there were more on-call workers in the "physical" group, while the "professional" group had fewer on-call workers. The rate of on-call workers was also higher for temporary and shift workers. Assessment of work environment revealed a higher rate of on-call workers exposed to physical, chemical, or ergonomic factors, as well as work-related stress ( Table 1).
The subjects were divided into 24 job classifications based on KECO categorizations and were analyzed for oncall work. The highest rate of on-call workers (31.7%) were those in the law, police, firefighting, and prison-related job group. The job classifications that had higher rates of on-call workers than healthcare and medical-related jobs (12.8%) include soldiers (30.2%), driver and transportationrelated jobs (14.6%), information and communicationsrelated jobs (13.4%), and construction-related jobs (12.9%). Moreover, although the proportions of on-call workers were low for management, accounting, and office-related (8.7%) and sales (8.1%) jobs, the absolute number of oncall workers in these job classifications were quite high, at 21.2%(578 subjects) and 7.8%(214 subjects) of the entire on-call workers, respectively ( Table 2).

On-call work and its association with health problems, injuries
Chi-square analysis was performed to determine the relationship between on-call work and health problems and injuries. The results indicated that all health problems except for cardiovascular disease had statistically significant higher prevalence among on-call workers (Table 3).
Bivariate logistic regression analysis was performed to determine the effect of on-call work on the risk of health problems and injuries. The risks were higher for on-call workers than those for non-on-call workers for the majority of health problems and injuries. This outcome did not change even after adjusting for personal and occupational characteristics, work environment, and work-related stress. The odds ratios for on-call workers in terms of physical health problems, psychological health problems, and injuries were 1

Discussion
This study utilized data from the 3rd Korean Working Conditions Survey on employed workers and analysed the relationships between their characteristics and health problems and injuries as well as the impact of on-call work on health problems and injuries. Analysis of the subjects' characteristics according to the on-call work showed differences in both personal characteristicsgender, age, education, income, smoking, alcohol consumption, and obesityand occupational characteristicsjob type, employment type, and shift work. Especially for the employment type, temporary workers were more often on-call workers. Unstable employment status in these temporary workers likely forces them to cope with stressful work environments. Moreover, previous studies reported that on-call workers have extended working hours (both daytime and nighttime) [1] and are exposed to both shift and night-time work [14], which may explain why on-call workers feel more fatigued in general. In fact, the analysis findings of the current study revealed a higher rate of on-call workers for shift work. However, there was no statistically significant difference in the rate of on-call workers based on working hours. On-call workers had increased exposure to all analysed physical, chemical, and ergonomic factors, suggesting that these workers tend to work in worse work environments. The rate of on-call workers was also higher in the group that experienced work-related stress, further supporting the previous finding that on-call work is a key factor of work-related stress [2][3][4]. The rates of on-call workers were higher in law, police, firefighting, and prison-related jobs; soldiers; driving and transportation-related jobs; information and communications-related jobs; and construction-related jobs, compared to the rate in healthcare and medicalrelated jobs, which was traditionally thought as a job category with a high rate of on-call workers. This result indicates that in addition to traditional studies of on-call workers that focus on healthcare and medical-related jobs, additional studies are necessary in order to analyse other job groups.
Analysis of the effects of on-call work on health problems and injuries showed higher odds ratios for the majority of health-related issues and injuries in on-call workers compared to those of non-on-call workers, with exceptions of respiratory difficulties, cardiovascular diseases. These results did not change even after adjusting for personal and occupational characteristics, work environment, and work-related stress. To our knowledge, no previous study has investigated the relationship between on-call work and hearing problems, skin problems, or abdominal pain. A potential explanation that could explain the relevance of on-call work with these problems is circadian disruption. In a study on sudden sensorineural hearing loss patients, 61.8% of the subjects reported insomnia before suffering from hearing loss, and the circadian clock gene was reduced compared to the control group [18]. It is thought that circadian disruption can be caused by irregular sleeping patterns that may occur during on-call work, which may cause hearing problems. Previous study have shown that exposure to light at night causes circadian disruption and decreased melatonin synthesis [19]. Melatonin may protect against psoriasis because it regulates the inflammatory response and antioxidant activity [20]. Previous study on shift workers have also shown that circadian disruption and decreased melatonin are associated with psoriasis [21]. In addition, in a mouse-based experiment, it was possible to regulate the circadian clock to control psoriasis-like skin inflammation by controlling the IL-23 expression in T cells [22], suggesting that circadian disruption may be related to inflammatory skin problems. Our study shows that the risk of skin problems is higher for on-call workers even after adjusting for the shift work. This may due to on-call workers also have disrupted circadian rhythm, which may caused by their irregular working hours, and light exposure at night. Previous studies have demonstrated the relationship between on-call work and indigestion [4] and possible induction of gastrointestinal tract diseases due to an imbalanced lifestyle from irregular eating habits and lack of sleep [23]. Thus, on-call workers who have irregular eating habitssince they are expected to work when neededwere expected to have various gastrointestinal tract-related issues and diseases. However, similar to hearing problems, additional studies are necessary in order to identify the potential mediating factors of skin problems and abdominal pain. However this is only one potential explanation that could explain the relevance of on-call work with hearing problems, skin problems and abdominal pain, so further studies are required.
A previous study reported that on-call work is associated with musculoskeletal pain (i.e., back and shoulder pain) [5], a result in agreement with our findings that on-call work is associated with back and upper/lower limb muscle pain. Furthermore, one possible explanation for these musculoskeletal symptoms was a lack of a rest and recovery period [24]. On-call workers have a shorter time for rest and recovery due to irregular working hours, which could lead to musculoskeletal symptoms in these workers.
A previous study of Finnish anesthesiologists reported that on-call work is associated with exhaustion, frustration, sleep disorders, memory disorders, and headache. Greater work-related burdens from on-call work resulted in an increased severity of these symptoms, and these symptoms disappeared during vacation [4]. These results indicate a strong relationship between on-call work and symptoms such as headache or fatigue. From our study of Korean workers, on-call work was associated with headache, eye fatigue, and general fatigue.
A previous study discussed depressed mood in on-call workers [6], and another suggested the possible relationship between on-call work and negative emotions [25]. Furthermore, on-call work was associated with daytime sleepiness, insomnia, and sleep deprivation [7,8], and possibly dysthymic disorder from continued sleep deprivation [26]. These results suggest that on-call work may be a cause of sleep deprivation and sleep disorders as well as for negative emotions in on-call workers who are constantly sleep deprived due to the nature of their work. Moreover, previous studies showed the association between on-call work and work dissatisfaction [2,13] and the result of a study on the relationship between work dissatisfaction and general mental health (i.e., depression and anxiety) [27] indicated that work dissatisfaction from on-call work might cause depression or anxiety. These results were further A previous study suggested the association between oncall work and work-related injuries [11]; the present study also found this association in Korean workers. On-call work is associated with reduced cognitive function and concentration [9,10] and studies have shown that sleep deprivation can cause reduced cognitive function [28,29]. Therefore, reduced cognitive function and concentration from on-call work can increase the risk of injuries.
This study has many strengths. To our knowledge, it is the first Korean study to determine the association between on-call work and health problems and injuries. Furthermore, unlike traditional studies that focused on a specific job group, this study included a wide range of job groups in order to investigate the relationship between oncall work and health problems, injuries. In the present study, higher rates of on-call workers were found in other job groups compared to that of the healthcare-related job group, which was the target job group in previous studies. This finding suggests the need for future studies focusing on other job groups. Second, this study utilized data from the Korean Working Conditions Survey and the 29,246 subjects included in this study were representative of Korean employed workers nationwide. In addition, the utilization of trained investigators to perform the survey minimized arbitrary interpretation of the survey responses.
This study had several limitations. First, the crosssectional study design based on data from the Korean Working Conditions Survey made it difficult to identify the clear causal relationship between on-call work and health problems and injuries. Future studies are needed to identify the clear causal relationships. Second, the only measurement tools for variables were the responses to the Korean Working Conditions Survey. More specifically, for the work environment, exposures were determined by dichotomous measures of the survey responses rather than actual assessments of work the environments. Furthermore, instead of objective measurements such as physician interviews, blood test, or medical imaging, self-reporting was used to assess health problems. And there may be a misclassification of the job types, because there were only four job types, so jobs of different character may belong to same category. Lastly, the assessed medical history of the subjects included only the presence of obesity and hypertension.

Conclusion
This is one of the first studies identified a significant relationship between on-call work in Korean workers and health problems and injuries. Moreover, this study included a wide range of job groups in order to investigate the relationship between on-call work and health problems, injuries. Future studies are required to identify clear causal relationships; similarly, additional discussions are needed in order to reduce the adverse effects of on-call work.
Abbreviations EWCS: European working conditions survey; KECO: Korean employment classification of occupation