Originals
Work-related Risk Factors for Workplace Violence among Korean Employees
2014 Volume 56 Issue 1 Pages 12-20
Details
2014 Volume 56 Issue 1 Pages 12-20
Objectives: The aim of this study was to identify work-related risk factors for workplace violence in a representative sample of Korean employees. Methods: We analyzed the associations between work-related factors and workplace violence in 29,171 employees using data from the 2011 Korean Working Conditions Survey. The survey included questions about verbal abuse, unwanted sexual attention, threats and behavior that humiliated the victim, physical violence, bullying/harassment and sexual harassment, and a respondent who answered yes to any of these 6 items was considered a victim of workplace violence. Results: The prevalences of verbal abuse, unwanted sexual attention and threats/ behavior that humiliated victims in the month preceding the study were 4.8, 1.0 and 1.5%, respectively. The prevalences of physical violence, bullying/harassment and sexual harassment in the year preceding the study were 0.7, 0.3 and 0.4%, respectively. Service workers had higher prevalences of overall workplace violence. Non-regular workers (OR=2.38, 95% CI=2.01–2.84), working more than 60 hours per week as opposed to 40–48 hours per week (OR=1.83, 95% CI=1.45–2.31) and night shift work (OR=1.88, 95% CI=1.54–2.30) were significant risk factors associated with workplace violence. Conclusions: Long working hours, job insecurity and night shift work were associated with a significant increase in workplace violence among Korean employees.
(J Occup Health 2014; 56: 12–20)
Workplace violence has been on the rise since the early 1990s, and now many societies recognize it as a major public health issue1). The term “workplace violence” has been used to refer to a wide spectrum of conditions, ranging from severe bodily injury to mental stress, which has led to confusion and the inability to compare results obtained from different studies. At present, the most widely accepted concept of workplace violence is based on the definition of the World Health Organization (WHO). The WHO considers both physical and psychological violence to fall under the heading of workplace violence. Physical violence describes the use of physical force against another person or group that results in physical, sexual, or psychological harm. Psychological violence describes the intentional use of power against another person or group that can result in harm to physical, mental, spiritual, moral, or social development2). Workplace violence includes physically injuring or assaulting a person, leading to actual harm; verbal abuse, including swearing, insults, or condescending language; aggressive body language indicating intimidation, contempt, or disdain; harassment, including mobbing, bullying, and racial and sexual harassment; and the expression of intent to cause harm, including threatening behavior as well as verbal and written threats3). Recently, psychological violence has attracted growing concern as a result of studies that indicate psychological violence and harassment, rather than physical violence, represent more common threats to most workers4).
It has been reported that workplace violence is associated with a number of adverse health conditions, including cardiovascular disease5), posttraumatic stress disorder6), depression and anxiety7–9), absence from work due to illness10, 11) and impaired well-being12, Workplace violence also affects an employee's satisfaction with and intention to leave their job13, 14).
Outside of Europe, there have been few studies of the prevalence of workplace violence in samples representative of the working population. According to a study based on the 2005 European Working Conditions Survey (EWCS) by the European Foundation for the Improvement of Living and Working Conditions (Eurofound), which was administered in 31 European countries, the prevalences of exposure to physical violence were 7% and 8% in men and women, respectively. In addition, the prevalences of sexual harassment were 1% and 3%, while the prevalences of bullying were 4% and 6% in men and women, respectively15. The prevalence of exposure to verbal abuse was 10.8% according to the 2010 EWCS, which was administered in 27 countries16.
In order to prevent workplace violence, it is important to define its risk factors. Risk factors are typically categorized into personal characteristics of the victims or perpetrators, such as gender, age, alcohol use and aggressive history17, 18), and work-related factors including job category1) and organizational attributes19, 20). A number of studies have investigated factors related to workplace violence, but the majority of these have been performed in a specific workplace, most often hospitals21–23). As a result, interest in workplace violence has focused on the healthcare sector, and national guidelines and strategies have been developed only for health-care workers2). Some studies dealing with workplace violence have been done in Asian countries. However, as in other regions, these studies have focused on health-care workers24–26). The aim of this study was to identify work-related risk factors for workplace violence in Korean employees using a nationally representative data set including all working groups.
Data were derived from the third Korean Working Conditions Survey (KWCS), which was conducted in 2011 by the Korea Occupational Safety and Health Agency (KOSHA). The KWCS is a national survey collecting information on social/occupational health indicators present in the working environment. The methodology and survey questionnaire used by the KWCS were very similar to those used in the European Working Conditions Survey (EWCS). In both studies, a representative sample of the economically active population aged 15 to 64 years, i.e., persons who were either employees or self-employed at the time of the interview, was selected. The study design for both surveys employed multistage random sampling based on the Population and Housing Census27). The primary sampling units were the population and housing enumeration district, and these were selected by systematic sampling in proportion to the number of households within the enumeration district. The secondary sampling units were household and household member. Ten target households were selected in the enumeration district by systematic sampling method from the household lists and district maps. Professional surveyors interviewed 50,000 workers in 50,000 households (1 per household). The response rate was 36%, and the contact rate was 57% in the 2011 KWCS. The validity and reliability of the 2010 KWCS, the study design of which was same as that of the 2011 KWCS, were evaluated, and the quality of the survey was assured28).
SubjectsThe 2011 KWCS collected data from 50,032 subjects representative of the working population. In this study, we included only 25,195 respondents who were employed and entered their present employment more than one year previously (sum of weights=29,171). Our study design was approved by the Institutional Review Board of the Seoul St. Mary's Hospital (approval ID: KC12EISI0413).
MeasurementWorkplace violence in this study was assessed using six questions. The subjects were asked to answer yes or no regarding whether they had experienced 1) verbal abuse, 2) unwanted sexual attention, or 3) threats and behavior intended to humiliate them over the last month, as well as 4) physical violence, 5) bullying/harassment, or 6) sexual harassment over the past 12 months. Respondents were considered targets of workplace violence if they answered yes to any of these six items.
The work-related factors measured included the length of employment contract (<1 month, 1 month to 1 year or >1 year), occupation (managers, professionals, clerks, sales workers, agricultural workers, craft and related trades, plant and machine operators and elementary occupations), the number of hours worked per week (<40, 40–48, 49–60 or >60), company size (1–4, 5–50 or ≥50 workers) and whether or not the job included night shift work. The potential confounders considered included gender, age, educational level and income.
Statistical analysisThe appropriate sampling weights from the KWCS were applied to the results of this study in order to make the findings nationally representative. The associations between workplace violence and gender, age group, education, type of employment contract, occupation, number of hours worked per week, size of company, and presence of night shift work were analyzed using the χ2-test. A series of univariate and multiple logistic regression analyses were then conducted to examine the associations between individual work-related factors and workplace violence. Multiple logistic regression analyses were adjusted for gender, age group, education level, and income. The significance level for all statistical analyses was p<0.05 using a two-tailed test. SAS version 9.2 (SAS Institute, Cary, NC, USA) was used to analyze the survey data.
The characteristics of the study participants and the prevalence of workplace violence are shown in Table 1, with all numbers reflecting weighted frequencies. There were 17,802 male workers and 11,369 female workers. More than half of the respondents were in their thirties or forties. Approximately 90% of the participants had obtained a high school education or higher. Most of the subjects were regular workers (78.6%). The most common job types were clerks (25.4%), professionals and related workers (20.9%) and elementary occupations (12.7%).
The prevalence of verbal abuse, unwanted sexual attention, and threats/behavior intended to humiliate the victim during the last month were 4.8%, 1.0% and 1.5%, respectively. The prevalence of physical violence, bullying/harassment, and sexual harassment over the last year were 0.7%, 0.3% and 0.4%, respectively. The overall prevalence of workplace violence was 5.8%. There were significant differences between men and women in the prevalence of verbal abuse (men>women), unwanted sexual attention and sexual harassment (women>men). Workers under 30 years of age experienced the highest prevalence of workplace violence overall (6.3%). Regarding the highest prevalence rates for each type of workplace violence, that for verbal abuse was in the 50s, that for unwanted sexual attention was in the 30s, those of threats and humiliating behavior, physical violence and bullying/ harassment were over 60 and that of sexual harassment was under 30. The prevalence of workplace violence by work-related characteristics is shown in Table 2.
The associations between work-related factors and workplace violence obtained from multiple logistic regression analyses are shown in Table 3. After controlling for age, sex, education and income, non-regular worker status (OR=2.38, 95% CI=2.01−2.84), working more than 60 hours (OR=1.83, 95% CI=1.45−2.31, reference =40−48 hours/week), and night shift work (OR=1.88, 95% CI=1.54−2.30) were identified as significant risk factors for workplace violence.
In this study, the prevalence of workplace violence in Korean workers was lower than that of European workers based on the EWCS. For example, the prevalence of verbal abuse in the month preceding the survey was 4.9% in our study, which was about half the prevalence reported in the EWCS 2010, which drew on respondents from 27 European countries (EU27). Similarly, the prevalence of having experienced threats/behavior intended to humiliate the victim was 1.4% in Korea, which was very low compared with the prevalence of 5.0% in the EU27. Finally, the prevalence of physical violence (0.7%) in our study was less than half that found in the EU27 (1.9%), and the prevalence of bullying/harassment (0.4%) was also much lower than that reported in the EU27 (4.1%)16). Generally, it is difficult to compare statistics on exposure to workplace violence and harassment across different countries. This is due to the different definitions and classifications used to delimit the concepts being studied, different methodologies for collecting and processing information, different time frames analyzed and cultural differences in the experience of violence and harassment. In this case, however, the methodology and questionnaire used in the KWCS were the same as those used in the EWCS; thus, the results of the two surveys are quite comparable. For this reason, the difference in the prevalence of workplace violence between Korea and the EU27 is probably not due to a lack of clarity in the definition of workplace violence, variation in the time frames studied or a difference in measurement methodology. It may be due to differences in organizational culture and structure29), as perceptions of whether workplace violence is socially acceptable are culturally influenced. In addition, important terms such as “verbal abuse” and “threats” could be interpreted differently in different societies. One problem encountered when studying workplace violence is underreporting. Some researchers have posited that underreporting of workplace violence is due to an employee's belief that a violent attack on them is their own fault or occurred because they were not competent30). We speculate that both the perception and the report of being a victim of workplace violence is lower in Korea than in western nations because Korean culture traditionally values the group rather than the individual and attaches importance to patience and concession. In addition, the concept of workplace violence is quite new in Korean society, and there are no regulations or policy interventions specific to violence in the workplace yet. This may result in a reduced sensitivity to the perception of workplace violence in Korea.
| Total** | Verbal abuse | Unwanted sexual attention | Threats and humiliating behavior | Physical violence | Bullying / harassment | Sexual harassment | Workplace violence+ | ||
|---|---|---|---|---|---|---|---|---|---|
| Sex | Male | 17,802 | 908 (5.1) * | 86 (0.5) * | 284 (1.6) | 136 (0.8) | 55 (0.3) | 30 (0.2) * | 1,037 (5.8) |
| Female | 11,369 | 496 (4.4) * | 197 (1.7) * | 142 (1.2) | 72 (0.6) | 39 (0.3) | 89 (0.8) * | 644 (5.7) | |
| Age | <30 | 5,110 | 270 (5.3) * | 59 (1.2) * | 83 (1.6) | 24 (0.5) * | 11 (0.2) | 28 (0.6) * | 337 (6.6) * |
| 30–39 | 8,864 | 412 (4.7) * | 119 (1.3) * | 145 (1.6) | 72 (0.8) * | 32 (0.4) | 44 (0.5) * | 519 (5.9) * | |
| 40–49 | 8,335 | 355 (4.3) * | 68 (0.8) * | 92 (1.1) | 45 (0.5) * | 23 (0.3) | 34 (0.4) * | 427 (5.1) * | |
| 50–59 | 5,098 | 293 (5.7) * | 31 (0.6) * | 76 (1.5) | 50 (1.0) * | 18 (0.4) | 9 (0.2) * | 316 (6.2) * | |
| >60 | 1,764 | 74 (4.2) * | 7 (0.4) * | 29 (1.7) | 18 (1.0) * | 10 (0.6) | 4 (0.3) * | 82 (4.6) * | |
| Education | <Middle school | 2,774 | 147 (5.3) * | 15 (0.6) * | 35 (1.3) | 27 (1.0) | 9 (0.3) | 3 (0.1) * | 161 (5.8) * |
| High school | 10,218 | 636 (6.2) * | 123 (1.2) * | 171 (1.7) | 86 (0.8) | 35 (0.3) | 54 (0.5) * | 738 (7.2) * | |
| >College | 16,179 | 622 (3.8) * | 145 (0.9) * | 220 (1.4) | 94 (0.6) | 49 (0.3) | 63 (0.4) * | 781 (4.8) * | |
| Income‡ (US Dollars / month) | <900 | 6,793 | 371 (5.5) * | 78 (1.2) * | 93 (1.4) * | 33 (0.5) * | 21 (0.3) * | 39 (0.6) * | 439 (6.5) * |
| 900–1,300 | 11,242 | 618 (5.5) * | 145 (1.3) * | 194 (1.7) * | 123 (1.1) * | 57 (0.5) * | 61 (0.5) * | 765 (6.8) * | |
| >1,300 | 10,611 | 382 (3.6) * | 55 (0.5) * | 119 (1.1) * | 41 (0.4) * | 12 (0.1) * | 16 (0.2) * | 440 (4.1) * | |
| Total | 29,171 | 1,404 (4.8) | 283 (1.0) | 425 (1.5) | 208 (0.7) | 93 (0.3) | 120 (0.4) | 1,419 (5.8) | |
| Total** | Verbal abuse | Unwanted sexual attention | Threats and humiliating behavior | Physical violence | Bullying / harassment | Sexual harassment | Workplace violence+ | ||
|---|---|---|---|---|---|---|---|---|---|
| Employment contract | Regular | 22,937 | 910 (4.0) * | 193 (0.8) * | 284 (1.2) * | 169 (0.7) | 70 (0.3) | 69 (0.3) * | 1,104(4.8)* |
| Non-regular | 6,235 | 494 (7.9) * | 90 (1.4) * | 141 (2.3) * | 39 (0.6) | 23 (0.4) | 51 (0.8) * | 577 (9.3) * | |
| Job | Managers | 605 | 30 (5.0) | 4 (0.6) | 15 (2.5) | 2 (0.3) | 0 (0.0) | 1 (0.2) | 37 (6.1) * |
| Professionals and related | 6,091 | 157 (2.6) | 49 (0.8) | 62(1.0) | 24 (0.4) | 3 (0.0) | 12 (0.2) | 210(3.5)* | |
| Clerks | 7,417 | 237 (3.2) | 68 (0.9) | 78 (1.0) | 39 (0.5) | 26 (0.4) | 38 (0.5) | 307 (4.1) * | |
| Service | 2,164 | 175 (8.1) | 57 (2.6) | 66 (3.1) | 28 (1.3) | 16 (0.7) | 26 (1.2) | 216 (10.0) * | |
| Sales workers | 2,753 | 154 (5.6) | 50 (1.8) | 51 (1.8) | 16 (0.6) | 12 (0.4) | 21 (0.8) | 192 (7.0) * | |
| Agriculture, forestry and fisheries | 96 | 1 (0.7) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 1 (0.7) * | |
| Craft and related trades | 2,928 | 159 (5.4) | 7 (0.2) | 32(1.1) | 21 (0.7) | 6 (0.2) | 7 (0.2) | 175 (6.0) * | |
| Plant, machine operators and assemblers | 3,336 | 269 (8.1) | 26 (0.8) | 71 (2.1) | 59 (1.8) | 17 (0.5) | 12 (0.3) | 292 (8.7) * | |
| Elementary occupations | 3,700 | 218 (5.9) | 24 (0.6) | 47 (1.3) | 19 (0.5) | 14 (0.4) | 3 (0.1) | 247 (6.7) * | |
| Working time (hr / week) | <40 | 2,016 | 75 (3.7) * | 12 (0.6) * | 18 (0.9) * | 10 (0.5) * | 5 (0.2) | 6 (0.3) | 84 (4.2) * |
| 40–18 | 16,141 | 575 (3.6) * | 135 (0.8) * | 169 (1.0) * | 84 (0.5) * | 40 (0.2) | 61 (0.4) | 699 (4.3) * | |
| 49–60 | 8,410 | 535 (6.4) * | 102 (1.2) * | 157 (1.9) * | 81 (1.0) * | 36 (0.4) | 45 (0.5) | 644 (7.7) * | |
| >61 | 2,605 | 220 (8.4) * | 34 (1.3) * | 81 (3.1) * | 32 (1.2) * | 13 (0.5) | 7 (0.3) | 254 (9.8) * | |
| Company size(number of workers) | 1–4 | 5,964 | 256 (4.3) | 53 (0.9) | 52 (0.9) * | 32 (0.5) * | 18 (0.3) | 24 (0.4) | 300 (5.0) |
| 5–19 | 14,795 | 720 (4.9) | 161 (1.1) | 195 (1.3) * | 138 (0.9) * | 47 (0.3) | 57 (0.4) | 872 (5.9) | |
| 50–299 | 5,238 | 277 (5.3) | 51 (1.0) | 112(2.1) * | 28 (0.5) * | 18 (0.4) | 26 (0.5) | 341 (6.5) | |
| >300 | 2,338 | 106 (4.5) | 12 (0.5) | 54 (2.3)* | 8 (0.3) * | 8 (0.3) | 6 (0.3) | 117 (5.0) | |
| Shift work | No | 26,360 | 1,121 (4.3) | 249 (0.9) * | 332(1.3) * | 154 (0.6) * | 72 (0.3) * | 104 (0.4) | 1,353 (5.1) * |
| Yes | 2,812 | 283 (10.1) | 35 (1.2) * | 93 (3.3) * | 54 (1.9) * | 21 (0.7) * | 16 (0.6) | 328 (11.7) * | |
| Total | 29,171 | 1,404 (4.8) | 283 (1.0) | 425 (1.5) | 208 (0.7) | 93 (0.3) | 120 (0.4) | 1,419 (5.8) | |
| Adjusted OR* | 95% CI | |
|---|---|---|
| Employment contract | ||
| Regular | 1.00 | |
| Non-regular | 2.38 | 2.01–2.84 |
| Job | ||
| Managers | 2.04 | 1.13–3.67 |
| Professionals | 1.00 | |
| Clerks | 1.34 | 1.04–1.74 |
| Service | 2.73 | 2.02–3.69 |
| Sales workers | 1.75 | 1.32–2.34 |
| Agricultural, forestry and fisheries | 0.17 | 0.02-1.24 |
| Craft and related trades | 1.69 | 1.23–2.33 |
| Plant, machine operators and assemblers | 2.08 | 1.56–2.77 |
| Elementary occupations | 1.48 | 1.04–2.10 |
| Night shift work | ||
| No | 1.00 | |
| Yes | 1.88 | 1.54–2.30 |
| Hours worked per week | ||
| <40 | 0.54 | 0.38–0.77 |
| 40–48 | 1.00 | |
| 49-60 | 1.70 | 1.45–1.99 |
| ≥61 | 1.83 | 1.45–2.31 |
| Company size (No. of workers) | ||
| 1–4 | 1.00 | |
| 5–49 | 1.49 | 1.22–1.81 |
| 50–299 | 1.81 | 1.41–2.33 |
| ≥300 | 1.40 | 0.97–2.00 |
Our findings showed that service workers had a higher prevalence of workplace violence compared with other occupations (OR [Odds ratio]=2.73; 95% CI [Confidence interval]=2.02−3.69, reference group=professionals). Service workers include police workers, firefighters, professional drivers and workers in restaurants. These workers usually experience interpersonal job demands and perform emotional labor. In these stressful work environments, experience of workplace violence could be prevalent.
In our study, individuals who worked more than 60 hours per week were more likely to be victims of workplace violence than those who worked 40–48 hours per week (OR=1.83, 95% CI=1.45−2.31). We chose 40–48 hours per week as the reference category because most employees were working 40–48 hours per week. Article 53 of the Korean Labor Standards Act sets the upper limit on work hours at roughly 52 per week31). Because this limit does not include overtime work performed on holidays, work hours in Korea may be substantially longer. Kimet al. reported that the average Korean employee spends 48.57 hours a week in the workplace, that 20.72% of the employed work more than 60 hours a week and that working more than 60 hours a week is associated with increased health problems32). A study on healthcare staff in Japan also showed that the risks of both physical aggression and verbal abuse were significantly higher in workers with long working hours25). Furthermore, in a study of medical professionals from 12 Chinese hospitals, long working hours (10 hr/day), and working the night shift were found to be risk factors for workplace violence, including verbal abuse33). These findings could be explained by invoking either a direct or underlying effect. If we regard workplace violence as a workplace hazard, then the more hours worked (i.e., the greater the exposure), the greater the probability of being victimized. In support of this view, one study found a positive correlation between the time two individuals spent together and whether or not workplace violence occurred between them34). Alternatively, an underlying effect mediated by fatigue and frustration could be respon- sible35). Jobs that require people to work longer hours may exhaust and frustrate workers, increasing the likelihood of workplace violence. In support of this hypothesis, research suggests that long working hours result in sleep problems, fatigue, depression and an increase in subjective health complaints by disturbing physiological recovery mechanisms36).
Our findings showed that individuals scheduled to work night shifts were more likely to experience workplace violence than those who worked during the day (OR=1.88, 95% CI=1.54−2.30). Night shift work is a well-known risk factor for external violence, i.e., violence perpetrated by persons from outside the organization1). This effect is not limited to external violence, however. A study of nursing staff in European countries showed that the prevalence of harassment by a superior or colleagues was higher in a night shift group than in a day shift group21). Similarly, a study performed in Taiwan found that night shift work is related to an increased risk of workplace violence, especially sexual harassment, for nurses37). Another study on Danish workers in the eldercare sector demonstrated that evening or night workers were 1.3–1.8 times more likely to experience bullying, threats, and violence38). It is possible that the high proportion of aggressive behavior reported in the health-care sector is due to night shift work. The fact that both night shift work and long working hours are associated with workplace violence may provide further support for an underlying mechanism of fatigue and frustration. Previous studies reporting the influence of night shift work on workplace violence have been limited to health-care settings, but our study found that this effect is not restricted to hospital workers.
In our study, non-regular workers were more likely to experience workplace violence than regular workers (OR=2.38, 95% CI=2.01−2.84). There are few reports examining the effect of job stability on workplace violence. Discussions about the possible link between layoffs or downsizing and workplace violence were frequent during the 1990s, however, and these factors are related to job security. A number of studies have found an association between downsizing/restructuring and bullying or occupational violence at the hands of supervisors, clients, or other workers39, 40). Similarly, in studies of workplace violence in hospital workers, insecurity and reduced staffing levels were significant risk factors for being bullied by supervisors or being abused or assaulted by patients or their fami- lies41). Although layoffs are associated with anger, depression and aggression among both victims and survivors, some researchers suggest that the manner in which layoffs are conducted is a better predictor than whether or not layoffs occur42). If layoffs are not carried out appropriately, workers may feel the process is unjust and become angry. Clearly, more evidence regarding the role of job insecurity in workplace violence needs to be collected.
Regarding company size, companies with 5–299 workers had a higher prevalence of workplace violence compared with companies with 1–4 workers (OR=1.49 and 95% CI=1.22−1.81 for companies with 5–49 workers, OR=1.81, 95% CI=1.41−2.33 for 50–299 workers' company). But the OR for companies with more than 300 workers was lower than that companies with 5–299 workers and marginally significant (OR=1.40, 95% CI=0.97−2.00). Overall, workers in medium- and large-sized companies had experienced workplace violence more than those in small-sized companies, though we could not find a constant tendency for workplace violence to increase as the size of the companies increased. In a nationwide study in Hong Kong, large private organizations (100 or more workers) tended to have more coworker violence than medium-sized (50–100 workers) and small organizations (less than 50 workers)43). If more coworkers, supervisors, and customers mean more chances to become involved in workplace aggression, then large organizations may host an increased prevalence of violence simply as a function of their size.
Our study had several limitations. One major limitation is that the study was cross-sectional, and therefore it could not prove a cause and effect relationship. Nevertheless, we postulate that the snapshot we obtained of the work environment using our survey was likely to capture long-term work-related factors as well as events experienced in the year preceding the study. Generally, the working environment of a company does not change suddenly. So, we thought it is reasonable to interpret that work-related factors such as working hours affect workplace violence rather than workplace violence affects work-related factors. In addition, we reanalyzed the data after excluding workers recently employed (<1 year), because newly employed workers (less than 1 year) did not have the same period to experience violence in their current companies or their violence experiences might have occurred in their former companies. The second limitation is the possibility that prevalence of workplace violence might be under- or overestimated due to measurement by self-report. It is possible that experience of minor violence was not reported. In this study, however, we presume that under- or overestimation may not affect the direction of the associations between workplace violence and work-related factors because it is likely to contribute to the association in a non-differential manner. The third limitation is that we did not use the same time frame for defining workplace violence. The interview questionnaire utilized different time frames (1 month or 1 year) according to the usual frequencies of the different types of workplace violence, because verbal abuse, unwanted sexual attention and threats and behavior intended to humiliate the respondent were considered frequent incidents in the workplace compared with other incidents (physical violence, bullying/harassment and sexual harassment). So, we had no choice but to use an operational definition (yes to any of 6 items) when defining workplace violence for evaluating association between workplace violence and occupational factors.
Our report is the largest study to date of risk factors related to workplace violence in Asia. Whereas the majority of studies on workplace violence have been limited to hospital workers, ours covers all occupations, making the results representative of Korea's working population nationwide. In summary, our data support a substantial effect of work-related factors, such as long working hours and night shift work, on workplace violence. It will be important to consider these factors when crafting interventions to curb workplace violence, and elucidating these risk factors may facilitate the identification of high-risk groups that could benefit from prevention programs.
Acknowledgment: This study was supported by the Korean Occupational Safety and Health Agency (KOSHA). The authors gratefully acknowledge support from KOSHA.
Conflict of Interest: The authors declare that they have no conflict of interest.
