Abstract
BACKGROUND:
There is a lack of studies that investigated the effect of a wide range of work environmental factors on stress and depression in Japan.
OBJECTIVES:
To examine the association of work environment factors with stress and depression among workers in Japan.
METHODS:
We conducted questionnaire surveys of workers that mainly engage in desk work in Japan. Stress was assessed through the Perceived Stress Scale (PSS), depression through the Patient Health Questionnaire-9 (PHQ-9), and work environment through physical and psychological workplace environment questionnaires. Workers were divided into low and high stress groups based on PSS score (median split), and divided into non-depressed and depressed groups based on their PHQ-9 score (< 5, and ≥5); these groups were then compared with their working environment. In addition, a multiple regression analysis was performed.
RESULTS:
Responses were obtained from 210 subjects. Multiple regression analysis showed that “Ability to work at one’s own pace” and “Ability to apply personal viewpoint to work,” etc., had effect on stress, while “Workplace harassment” and “Support from colleagues,” etc., had effect on depression.
CONCLUSIONS:
The results suggest that stress and depression in Japanese workers are related to factors such as job demands, control of work, workplace harassment, and psychological safety.
Introduction
In 2017, mental illness was the leading cause of disability, and major depressive disorder was the leading mental disorder in terms of years lived with disability in the world [1]. In Japan, the estimated lifetime prevalence of depression is 6.2% (estimates for 2002–2006), making it the most common psychiatric disorder in the country [2]. The cost of disease is also enormous, estimated to exceed 3.0 trillion JPY (approximately 2.8 billion USD) per year [3]. Depression has a large impact in the field of occupational health because it is particularly prevalent in the working age population (20–65 years old), and more than half of the social loss due to depression is caused by absence or absenteeism [2].
Not only due to the direct psychiatric symptoms but also physical comorbidity may lead to the productivity loss. Depression has been associated with cardiovascular morbidity and mortality [4], and with gastrointestinal symptoms [5].
To date, various studies have reported the relationship between work environment factors and stress and/or depression. For example, it has been shown that workplace violence (bullying, racial harassment, intimidation, physical violence) [6], work demand, control, and support from others [7] are associated with stress. Such stresses may increase the risk of depression [8]. Regarding the relationship between work environment factors and depression, physical and psychological demands, abusive supervision [9], lack of participation in decision-making and problem solving, lack of support and feedback [10], high demand with low control of work, and high effort and low reward [4], have all been identified as risk factors for depression. Theorell et al., in their review and meta-analysis of 59 studies that investigated depression and work environment from European and English-speaking countries, reported that moderately strong evidence was found for job strain (high psychological demands and low decision latitude), low decision latitude and bullying having significant impact on development of depressive symptoms. Limited evidence (grade two) was shown for psychological demands, effort reward imbalance, low support, unfavorable social climate, lack of work justice, conflicts, limited skill discretion, job insecurity and long working hours [11]. Thus, a number of studies have shown the association of work environment factors with stress and/or depression, however, of note, most of them are from Northern American and European countries.
On the other hand, a limited number of studies have been reported from Asian countries. Chang et al., a study targeted Taiwanese healthcare worker reported that work anxiety and workplace violence was associated with depression [12]. Yamasaki, targeting manufacturing company employees, reported that long working hours was associated with depression among employees with high strain jobs [13]. Tsuno and Kawakami, also targeting manufacturing company, reported that 1.8% of Japanese employees experienced physical assaults during working time and it had impact on depression especially among employee with high socioeconomic status [14]. Yokouchi and Hashimoto reported in their survey focusing on fairness in workplace that both low deviation and high deviation of interactional fairness perception were significantly and positively associated with serious psychological distress [15].
Thus, there are some studies focusing on specific population and/or topic but there is a lack of studies in recent years that investigated the impact of wide range of environmental factors including job demands, control of work, workplace harassment, support others, and workplace atmosphere on stress and depression in Japan. It is reported that working hours and non-scheduled working hours are longer in Japanese workplaces than European and American countries [16, 17]. Moreover, there is a growing consensus in the literature that in European and American cultural contexts there is a strong belief in the independence and autonomy of the self, whereas in East Asian cultural contexts there is a contrasting view of the self as interdependent [18]. Therefore, some different patterns regarding the relationship between work environment factors and stress/depression may be seen in Japan from European and American countries, such as stronger impact of working hours, relationship with co-workers or supervisors. Therefore, this study aimed to collect the workers’ stress and depression and to examine how a wide range of work environments relate to stress and depression workers in Japan.
Methods
This study is part of a research project supported by the Japan Agency for Medical Research and Development (AMED) (grant number: 18le0110008h0001; “Unobtrusive Sensing Technology for Quantifying Stress and Wellbeing to Promote a Healthy Workplace”).
The main purpose of this study was to examine the relationship between stress, well-being and biological signals such as heart rate variability, voice, and electrodermal activity from various perspectives. In this paper, however, we focus on participants’ responses to the questionnaires that were also given as part of the study. The concept, methodology, and overall goals of the study are described elsewhere [19].
Study population
The inclusion criteria for participants in this study were adult workers (≥20 years old) working at companies that mainly engage in desk work. Those currently undergoing treatment for mental disorders such as depression were excluded from the study. This study was conducted from November 5, 2018 to January 24, 2020.
Data collection
Data collection was done as follows:
Demographic characteristics
Study participants were asked to provide demographic characteristics including age, gender, type of job, type of employment, years of employment, household income, etc. as well as lifestyle information, such as commute time, commute method, sleep hours on weekdays, sleep hours on holidays, smoking habits (smoking or non-smoking), and drinking habits (none/socially or habitually).
Self-report questionnaires on work environment stress, depressive symptoms, and sleep quality
Researchers sent participants an e-mail with a URL link to the research website that provided self-report questionnaires about stress and depressive symptoms, and participants completed the questionnaires online. Participants were asked to answer to the questionnaires about work environment on baseline, about stress on 4-weeks follow up, about depression on 2- and 4-weeks follow up, and about sleep quality on daily bases.
Work environment were assessed using The Brief Job Stress Questionnaire (revised edition) (BJSQ) [20]. The BJSQ is an 80-item measure of one’s psychosocial work environment and employee and organizational outcomes, and it was used to assess subjects’ work environments. Answers were given using a 4-point Likert scale. The scores were calculated so that a higher score indicates a more favorable environment. In addition, for this study, the research team independently added 27 questions regarding work environment to the existing list of 80 items, for a total of 107 questions. Since December 2015, companies with 50 or more workers have been required to conduct stress checks once a year, and a shortened version of this questionnaire (57 items) [21] has been widely used in Japan [22].
Stress was assessed using the Japanese version of the Perceived Stress Scale (PSS) [23]. The PSS is a measure to evaluate comprehensive stress developed by Sheldon Cohen [24]. Responses were made using a 5-point Likert scale (“Not at all” being 0 points, “It always happened” being 4 points). The total score range is 0–40, with a higher score indicating higher stress. The study used a Japanese version of the PSS (α= 0.71).
Depression was assessed using the Japanese version of the Patient Health Questionnaire-9 (PHQ-9) [25]. The PHQ-9 is a commonly used verification screening tool for depression developed by Robert L. Spitzer [26]. It is a self-report questionnaire consisting of nine questions about depression symptoms, such as anhedonia, depressed mood, and suicidal tendencies. Respondents answer based on how frequently they have experienced each of the nine symptoms over the past two weeks: “Not at all,” “Several days,” “More than half the days,” and “Nearly every day.” The score range is 0–27 points: 0–4 = asymptomatic, 5–9 = mild symptoms, 10–14 = moderate symptoms, 15–19 = moderate to severe symptoms, and ≥20 = severe symptoms [25]. A score of 10 or more is considered to be the threshold for the possible presence of a major depressive disorder [27]. The study used a Japanese version of the PHQ-9 (α= 0.88). [28]
For daily sleep quality, the Japanese version of the Athens Insomnia Scale (AIS-J) was used to gauge the quality of sleep based on a 4-point scale (1: Very dissatisfied, 2: Fairly dissatisfied, 3: Slightly dissatisfied, 4: Satisfied) [29].
Statistical analysis
In order to examine if work environment factors are different between the high vs. low stress groups, or depressed vs. non-depressed groups, we have split the whole population into two groups. For stress, we used a median of PSS scores and placed participants with higher scores in high stress group (PSS ≥19.0) and participants with lower scores in low stress group (PSS < 19.0). For depression, based on the cut-off values, participants scoring 0–4 points were assigned to the “non-depressed” group, and those with ≥5 points to the “depressed” group [25]. In addition, as a post hoc analysis, based on the threshold for the likelihood of the presence of major depressive symptoms [27], participants were also divided into a 0–9 point group termed the “non-depressed or mildly depressed” group and a≥10 point group termed the “moderately or severely depressed” group.
T-test and Chi-squared test analyses were used to compare subject characteristics such as age and job type with stress and depression. Participants were divided into two groups (agree and disagree) for work environment according to their scores on the BJSQ 4-point scale, and then a Chi-squared test was performed on the high stress, low stress, and various depressive symptom groups.
There are some items asked about emotions and the environment unrelated to the workplace in the BJSQ, and these items were excluded from the current analysis. For example, we omitted “I’m full of energy,” “I’m tired,” and satisfaction about family.
In order to examine the relationship between work environment and stress or depressive symptoms, a multiple regression analysis was conducted with stress and depressive scores as dependent variables and work environment as an independent variable, and the work environment factors that correlated with stress and depressive symptoms were confirmed. For independent variables, items with a correlation coefficient P value of < 0.2 or less were selected.
In addition, as a post hoc analysis, we categorized the items into 11 sub-domains (Job demands, Control of work, Work atmosphere, Organizational factors, Physical environment, Workplace harassment, Relationship with superiors, Relationship with colleagues, Relationship with others, Aptitude, Psychological safety) so that one can understand the impact of conceptualized work environment factors on stress and depression. The sub-domains scores were calculated, and T-test was conducted to examine whether there were significant differences in the sub-domains scores between the high and low stress, and the depressed group and non-depressed group.
All analyses were performed using the statistical package SPSS version 25.0 for Windows.
Results
A total of 249 subjects participated in this study, but 39 were excluded due to missing data. Thus, the final sample comprised 210 participants.
Target subject characteristics
The mean age (Mean±SD) was 39.1±8.53, and male 112/210 (53.3%), administrative work 93/210 (45.4%), managerial work 61/210 (29.8%), mean length of service±SD = 9.84±8.34. The mean PSS score was 19.13±5.78, and PHQ-9 score was 5.63±4.05. Those with no depressive symptoms accounted for 47.1% of participants, mild depression was 37.6%, moderate depression was 10.5%, moderate to severe depression was 4.8%, and severe depression was 0.0% (Table 1).
Subjects’ demographic characteristics, and relationship between stress, depression, and demographic characteristics
Subjects’ demographic characteristics, and relationship between stress, depression, and demographic characteristics
Numbers in bold indicate that the differences did not disappear after Bonferroni correction.
When comparing Demographic Characteristics among the high and low stress groups, the low stress group reported better quality of sleep than the high stress group (Table 1). The significant relationship between depressive symptoms and age, job type, and household income was seen. In addition, the depressed group reported poorer sleep quality than the non-depressed group (Table 1).
Relationship between stress, depression, and work environment
Significant differences were found between the high and low stress groups regarding the items listed below with low stress groups indicated better work environment: “Ability to work at one’s own pace,” “There is a difference of opinion in my department,” “The atmosphere in my workplace is friendly,” “In our workplace we understand and acknowledge each other,” “Emotional burden,” “A workplace that respects each individual’s values,” and “Sitting close to someone one doesn’t like” (Table 2).
Relationship between stress, depression, and work environment (Chi-squared test)
Relationship between stress, depression, and work environment (Chi-squared test)
Numbers in bold indicate that the differences did not disappear after Bonferroni correction.
On the other hand, no statistical difference in work environment between the non-depressed group and the depressed group (Table 2). However, when comparing the non-depressed or mildly depressed group with the moderately or severely depressed group, the moderately or severely depressed group had a worse work environment than the non-depressed or mildly depressed group, and these results were related to the following items: “Decide the order in which work is done,” “Ability to apply personal viewpoint to work,” “Support from colleagues,” and “I am being bullied at work (including sexual and power harassment)” (Table 2).
The results of the multiple regression analysis are shown in Table 3. It was shown that the following work environment factors had an effect on stress: “Ability to work at one’s own pace,” “Someone else (people other than yourself) is being bullied (including sexual and power harassment),” “Inability to get work done in time,” “Emotional burden,” “Ability to apply personal viewpoint to work,” and “Risk losing one’s job.”
Relationship between stress and work environment (multiple regression analysis)
Stepwise method. B = Beta, SE = Standard error, R2 = coefficient of determination. The predictor variables included: Inability to get work done on time, Ability to work at one’s own pace, Decide the order in which work is done, Ability to apply personal viewpoint to work, Rarely use one’s skills or knowledge at work, There is a difference of opinion in my department, My department doesn’t get along with other departments, The atmosphere in my workplace is friendly, My work environment is not good, The job is right for me, It’s a rewarding job, Support from superior, Support from colleagues, Ability to rely on one’s colleagues, My superior will listen to me, My colleagues will listen to me, Emotional burden, Multiple people demand that we contradict each other, Know what one’s duties and responsibilities are, Opportunities to develop your strengths at work, Getting a salary and bonus for my work, Getting the credit my superior deserve, Risk losing one’s job, My superior treats me with honesty, My workplace provides chances to make up for failures, Information from management is reliable, Employees’ opinions are heard when there are changes in the workplace and at work, A workplace that respects each individual’s values, The results of personnel evaluations are fully explained, Education is provided to motivate and help with careers, I am being bullied at work (including sexual and power harassment), In our workplace we understand and acknowledge each other, Have to learn new things every time, Have no one to help or share work when one is busy, The workplace doesn’t protect me in a work emergency, Cramped for the size of the room, Feel free to have conversations about things other than work-related matters during the workday, Someone else (people other than yourself) is being bullied (including sexual and power harassment), I am in a position or job that is commensurate with my abilities and experience, It’s a workplace where you are often reprimanded (not just by yourself), Sitting close to someone one doesn’t like, Not free to leave one’s desk during business hours, I can’t take a vacation when I want to, No comfortable resting place, No natural light in the room, You can express your feelings honestly without being afraid of others’ reactions or feeling ashamed, You have freedom to make decisions regarding your work. For independent variables, 47 items were input with a correlation coefficient P value of < 0.2 or less.
It was shown that the following factors had an effect on depression: “Decide the order in which work is done,” “I am being bullied at work (including sexual and power harassment),” “Support from colleagues,” “The workplace doesn’t protect me in a work emergency,” “Ability to apply personal viewpoint to work,” and “Emotional burden” (Table 4).
Relationship between depression and work environment (multiple regression analysis)
Stepwise method. B = Beta, SE = Standard error, R2 = coefficient of determination. The predictor variables included: Have to do a lot of work, Inability to get work done on time, Have to work hard, It’s a very physical job, Ability to work at one’s own pace, Decide the order in which work is done, Ability to apply personal viewpoint to work, Rarely use one’s skills or knowledge at work, There is a difference of opinion in my department, My department doesn’t get along with other departments, The atmosphere in my workplace is friendly, My work environment is not good, The job is right for me, It’s a rewarding job, Support from superior, Support from colleagues, Ability to rely on one’s colleagues, My superior will listen to me, My colleagues will listen to me, Emotional burden, Multiple people demand that we contradict each other, Know what one’s duties and responsibilities are, Opportunities to develop your strengths at work, Getting a salary and bonus for my work, Getting the credit my superior deserve, Risk losing one’s job, My superior treats me with honesty, If you work hard and do your job you will be praised, My workplace provides chances to make up for failures, Information from management is reliable, Employees’ opinions are heard when there are changes in the workplace and at work, A workplace that respects each individual’s values, The results of personnel evaluations are fully explained, People in various positions (regular, non-regular, part-time, etc.) are respected as members of the workplace, Education is provided to motivate and help with careers, I am being bullied at work (including sexual and power harassment), In our workplace we understand and acknowledge each other, Have to do several things at once, Have no one to help or share work when one is busy, Have no substitute when one is absent, The workplace doesn’t protect me in a work emergency, Cramped for the size of the room, Feel free to have conversations about things other than work-related matters during the workday, Do not exchange greetings at work, Someone else (people other than yourself) is being bullied (including sexual and power harassment), I am in a position or job that is commensurate with my abilities and experience, It’s a workplace where you are often reprimanded (not just by yourself), My superior or colleague is constantly checking one’s work, Sitting close to someone one doesn’t like, Not being able to take free breaks during work hours, I can’t take a vacation when I want to, You need to stay/go back to work even if your job is done, No comfortable resting place, No natural light in the room, You can express your feelings honestly without being afraid of others’ reactions or feeling ashamed, You have freedom to make decisions regarding your work. For independent variables, 56 items were input with a correlation coefficient P value of < 0.2 or less.
When we compared the sub-domains of work environment between the high vs. low stress groups, the low stress groups reported better “Job demands,” “Workplace atmosphere,” “Organizational factors,” “Physical environment,” and “Workplace harassment,” compared to the high stress groups. When we compared the sub-domains of work environment between the depressed vs. non-depressed groups, the non-depressed group had better “Aptitude,” and “Psychological safety,” compared to the depressed group (Appendix).
This study examined the relationship between physical and psychological environmental factors experienced in the workplace and stress and depression among 210 Japanese workers that mainly engage in desk work. The results showed that stress and depression were related to work environment factors such as “Job demands,” “Control of work,” “Workplace harassment,” “Psychological safety,” “Support from colleagues,” and “Fear of losing one’s job.” We recognize that this is the first study in Japan to examine the association of stress and depression with a wide range of work environment factors that workers may experience.
First, based on multiple regression analysis, among the factors related to job demands and work control, the study found that stress was influenced by factors such as “Inability to get work done on time” and “Ability to work at one’s own pace.” As discussed before, it is reported that working hours and non-scheduled working hours are longer in Japanese workplace, but reflecting recent government demand to reduce overtime, “Inability to get work done on time,” “Ability to work at one’s own pace” are emerging reasons for stress among workers. In addition, the study found that depression was influenced by “Decide the order in which work is done.” Interestingly, it was also suggested that “Emotional burden” related to job demands have influenced both stress and depression. Previous studies have also shown that job demands and control are associated with stress and depression [9, 31]. In addition, Kim et al. found a threefold higher risk of presenteeism among non-depressive workers with higher job demands than among those with lower demands [32]. Furthermore, with regard to emotional burden, previous studies have similarly shown an association between stress and emotional exhaustion (due to emotional regulation) [33], an association between depression and emotional burden [34], and an association between emotional demands and prolonged absenteeism, suggesting that emotional burden may be harmful to workers’ health [35].
Second, regarding the relationship with and support from others, the present study found that “Support from colleagues” was associated with depression, similar to the results of previous studies [12, 14]. However, stress and “Support from colleagues,” and stress and depression and factors related to relationship with one’s superiors were unaffected. Previous studies have suggested an association between stress and “Support from colleagues,” [36] and between depression and “Support from superiors.” [12] There are also studies showing that the relationship with superiors has no effect on stress, while support from colleagues has an effect [30]. These different results may be due to differences in the occupations studied, as previous studies focused on nurses and teachers, while this study mainly included those engaged in desk work. In the future, further examination of these aspects, such as how different occupations are affected, is necessary.
Additionally, it was shown that workplace harassment factors had an impact; for example, “Someone else at work is being bullied” affected stress and “I am being bullied at work (including sexual and power harassment)” affected depression. Previous studies have shown associations between workplace harassment and symptoms of stress and depression [9, 37], and similar results were found in this study.
Edmondson et al. were the first to report that psychological safety, is important in workplace with regards to performance and creativity [38, 40]. She argued that psychological safety is better treated as a team-level climate and defined it as the “shared belief held by members of a team that the team is safe for interpersonal risk taking.” [38] In this study, among factors related to psychological safety, depressive symptoms were associated with “Whether the workplace will protect you in case of an emergency.” Notably, stress and depression were commonly associated with factors such as “Ability to apply personal viewpoint to work.” It has been shown in previous studies that the ability to apply personal viewpoint to work is related to depression [13], and the results of the present study similarly indicated that factors related to psychological safety may have an impact on workers’ stress and depression. According to Kitayama et al. Asians justify their choice only when they believe that their choices are known to others and, thus, they try to defray worry of their disapproval [41]. The reason for “Ability to apply personal viewpoint to work” being related with stress/depression in our study may be reflecting such Asian characteristics.
It was also shown that stress and the factor “Fear of losing one’s job” were connected. Previous studies have also suggested an association between job insecurity and stress [42], and this study showed similar results.
On the other hand, other studies have shown an association between depressive symptoms and rewards [4, 35], but this study found no association. In 2019, income distribution surveys in Japan found that annually, 12.8% of the population makes 2 million to < 3 million yen, 23.3% makes 3 million to < 5 million yen, 16.8% makes 5 million to < 7 million yen, 15.1% makes 7 million to < 10 million yen, and 12.1% makes ≥10 million yen [43]. The participants in this study belong to relatively high-income groups, and it is possible that depressive symptoms in such groups may not be associated with remuneration.
The improvements in the environmental factors may reduce stress and depression among workers in Japan. For example, among job demand, “Emotional burden” had effect on stress and depression, and “Inability to get work done on time” had effect on stress. Therefore, some measures to decrease the stress such as having meeting on regular bases to share emotional burden and adjust workload may be useful. As an aside, this study showed that daily sleep quality was commonly associated with stress and depressive symptoms among employees. Previous studies have shown that sleep is associated with stress and depressive symptoms [44, 45]. This study also showed the association between sleep quality and stress and depressive symptoms.
Surprisingly, as many as 54.8% of all subjects in this study had mild depression based on PHQ-9 scores, and the average stress score was 19.13±5.78. In a separate cross-sectional study of workers aged 18 years and older, the distribution of PHQ-9 scores was 73% for non-depressed and 27% for mild symptoms or more [46]. The high prevalence of depression in this study could be due to differences in the study populations. The subjects of this study were mostly full-time workers, while previous studies included part-time workers, housewives, retirees, and students. Moreover, stress scores averaged 20.31±5.42 in a study of university students in Japan [23] and 17.2±5.7 in a study of 33 workers aged 20–50 [47]. In a Chinese study of 9,507 persons aged 18 years or older (average age 47.5±14.1), those aged 18–44 years scored 19.6±4.7 points, those aged 45–59 years scored 19.2±5.0 points, and those aged 60–94 years scored 18.6±5.1 points [48]. Therefore, the subjects in this study tended to have a comparatively higher rate of depression, while their stress scores were similar to those found in previous studies. The fact that the stress scores in this study are similar to those of previous studies, but the rate of depression is higher, seems to be a contradiction, and we do not have a clear answer for why this may be the case.
Limitations
Since the subjects in this study were mainly deskwork employees, the generalizability of the results may not be possible for other work sectors (i.e., non-desk workers). Moreover, we did not adapt random sampling method, the sample should not be regarded as the representative of desk worders in Japan. At the same time, the data were collected from the participants with similar work sectors, one should take the possibility of a common variance bias into consideration when interpreting the results. The number of the moderately or severely depressed group was small, there might be a type II error. In addition, we did not ask about factors outside the work environment, such as a participant’s personality, which might have had some impact on the outcomes. Since the questionnaires used were self-report, subjective bias may have occurred in subjects’ answers. Lastly, it is possible that some of the subjects did not answer the BJSQ honestly, as it asked sensitive questions about their work environment. However, since the results of this study’s questionnaires are not made available to the subjects’ workplaces, and subjects were advised of that fact, we consider untruthful answers to be unlikely.
Conclusion
This study examined how psychological and social environmental factors experienced in the workplace may affect stress and depression among Japanese workers. The results showed that stress and depression were related to environmental factors such as “Job demands,” “Control of work,” “Workplace harassment,” “Psychological safety,” “Support from colleagues,” and “Fear of losing one’s job.” However, this study is not a causal study. In the future, it is necessary to investigate the relationship between work environment and stress and depression by conducting intervention research to reduce workers’ stress and depression.
Footnotes
Acknowledgments
This study was funded by the Japan Agency for Medical Research and Development (AMED) (grant number: 18le0110008h0001; “Unobtrusive Sensing Technology for Quantifying Stress and Wellbeing to Promote a Healthy Workplace”). The authors are grateful to the participants of this study, to the companies that willingly agreed to provide sites for this research: Aflac Life Insurance Japan Ltd.; AI messenger, Inc.; CA ADvance, Inc.; Halmek Holdings, Inc.; INTAGE Healthcare Inc.; Nomura Securities Co., Ltd.; SmartShopping, Inc.; SoldOut, Inc.; The Dai-ichi Life Insurance Company, Ltd.; WEDDING PARK CO., Ltd.; and ZENSHO HOLDINGS CO., LTD. The authors also thank Mr. Kuo-Ching Liang, Mr. Ryo Takemura, Mr. Asuka Koshi, Ms. Yoko Usami, Ms. Yuki Ishikawa, Ms. Kumiko Hiza, and Ms. Hiromi Mikami for their support during this study.
Conflict of interest
None of the authors have any conflicts of interest to declare.
Appendix A
Relationship between stress, depression, and sub-domains of work environment (T-test). Numbers in bold indicate that the differences did not disappear after Bonferroni correction.
Item
Stress (PSS)
Depression (PHQ-9)
Low group (PSS < 19.0)
High group (PSS ≥19.0)
p-value
Non-depressed group (PHQ-9 score = 0–4)
Depressedgroup (PHQ-9 score ≥5)
p-value
n = 102
n = 108
n = 99
n = 111
mean
SD
mean
SD
mean
SD
mean
SD
Job demands
2.39
0.33
2.21
0.34
2.34
0.35
2.26
0.34
0.100
Control of work
3.32
0.46
3.20
0.52
0.082
3.31
0.45
3.21
0.53
0.133
Workplace atmosphere
3.38
0.37
3.14
0.51
3.31
0.40
3.21
0.50
0.122
Organizational factors
2.83
0.46
2.61
0.50
2.82
0.45
2.63
0.52
0.005
Physical environment
3.08
0.62
2.75
0.69
3.00
0.67
2.82
0.67
0.056
Workplace harassment
3.74
0.44
3.42
0.71
3.66
0.52
3.50
0.68
0.059
Relationship with superiors
2.89
0.63
2.67
0.65
0.012
2.87
0.62
2.70
0.66
0.063
Relationship with colleagues
2.92
0.65
2.67
0.63
0.005
2.88
0.67
2.71
0.63
0.064
Relationship with others
2.91
0.40
2.74
0.49
0.010
2.89
0.42
2.76
0.48
0.052
Aptitude
3.15
0.45
2.92
0.53
3.14
0.49
2.94
0.51
Psychological safety
2.81
0.30
2.70
0.34
0.012
2.82
0.33
2.69
0.31
Appendix B
Relationship between stress and sub-domains of work environment (multiple regression analysis) Stepwise method. B = Beta, SE = Standard error, R2 = coefficient of determination. The predictor variables included: Job demands, Control of work, Work atmosphere, Organizational factors, Physical environment, Workplace harassment, Relationship with superiors, Relationship with colleagues, Relationship with others, Aptitude, Psychological safety. For independent variables, 11 sub-domains were input with a correlation coefficient P value of < 0.2 or less.
Independent variables
Stress (PSS)
Unstandardized coefficients
Standardized coefficients
t-value
P-value
B
SE
β
Job demands
–4.12
1.06
–0.25
–3.88
< 0.001
Aptitude
–2.40
0.75
–0.21
–3.19
0.002
Workplace harassment
–1.42
0.63
–0.15
–2.25
0.026
R2
0.22
Appendix C
Relationship between depression and sub-domains of work environment (multiple regression analysis) Stepwise method. B = Beta, SE = Standard error, R2 = coefficient of determination. The predictor variables included: Job demands, Control of work, Work atmosphere, Organizational factors, Physical environment, Workplace harassment, Relationship with superiors Relationship with colleagues, Relationship with others, Aptitude, Psychological safety. For independent variables, 11 sub-domains were input with a correlation coefficient P value of < 0.2 or less.
Independent variables
Depression (PHQ-9)
Unstandardized coefficients
Standardized coefficients
t-value
P-value
B
SE
β
Organizational factors
–1.43
0.57
–0.17
–2.49
0.013
Work harassment
–1.29
0.44
–0.20
–2.95
0.004
Control of work
–1.52
0.57
–0.18
–2.66
0.008
R2
0.16