ワーク・ファミリー・バランスに関する上司行動と労働機能障害の関係
2026 年 48 巻 2 号 p. 113-124
詳細
2026 年 48 巻 2 号 p. 113-124
The aim of this study was to examine the association between Family Supportive Supervisor Behaviors (FSSB), defined as supervisors’ behaviors that assist subordinates in balancing work and family life, and work functioning impairment. A cross-sectional online survey was conducted among 17,893 employees from six Japanese companies, and data from 12,027 respondents were included in the analysis. FSSB was assessed using the Japanese version of Hammer’s 14-item scale, and work functioning impairment was measured with the Work Functioning Impairment Scale (WFun). Multivariate regression analyses were performed to evaluate the association between FSSB and WFun scores, adjusting for age, gender, marital status, presence of children under age 18, presence of intra-family caregivers, and companies. After testing for interaction effects between FSSB and gender, stratified analyses were conducted. Additionally, we assessed the relationship between each of the four FSSB subdimensions and work functioning impairment. The results indicated a significant negative association between FSSB and WFun scores. The interaction between FSSB and gender was statistically significant. Stratified analyses revealed that higher FSSB scores were associated with lower WFun scores in both males and females, with a stronger association among males. Among the four subdimensions, emotional support, instrumental support, and creative work-family management were significantly associated with lower work functioning impairment, while role modeling behaviors was not. These findings suggest that strengthening FSSB may help reduce work functioning impairment in workplaces with increasingly diverse workforces.
More females are joining the labor market worldwide, in both developed and emerging economies, and this trend is expected to continue growing [1–3]. As more individuals take on both work and family roles, the likelihood of work-family conflict (WFC)—a conflict that happens when work and family roles interfere with each other—has increased [4]. In the past, couples often divided these roles, but as social norms change, more people handle both [5]. This situation has led to a higher risk of WFC, which affects employees’ mental health [6, 7] and work performance [8].
Previous studies have found that WFC leads to psychological distress [6], stress reactions [7], and lower self-evaluations of work performance [8]. These negative effects could harm individual health and reduce workplace productivity [9]. Companies are expected to create work environments that enable employees to balance work and family life according to their needs. When a company actively invests in improving the quality of its working environment, it can enhance employees’ mental and physical health, increase job satisfaction, and improve organizational performance [10–12]. To achieve this, many companies are implementing flexible working arrangements and paid leave policies [13,14]. However, having these policies does not always mean that employees will use them, but employees are more likely to use them if supervisors understand these systems implemented by the company [15].
Supervisors play a key role in helping subordinates balance work and family life [16–19]. When supervisors actively support work-family policies and encourage subordinates to use them, workers are more likely to benefit from these policies, which may help reduce the negative effects of WFC. Family Supportive Supervisor Behaviors (FSSB) refers to specific actions supervisors take to help subordinates manage their work and family roles [20]. It comprises four subdimensions: emotional support, instrumental support, role modeling behaviors, and creative work-family management [20].
Previous research on FSSB has demonstrated that it reduces WFC [17, 21], enhances the harmony between work and family roles [17–19], decreases psychological distress [22], improves psychological health [21], and positively impacts job performance [18, 21, 23]. However, research on FSSB and job performance has focused on only certain industries, metropolitan hospitals and grocery stores, which makes it difficult to say whether the findings apply to other occupations [18, 21, 23]. Previous studies examining the relationship between FSSB and subordinate performance have shown that performance improved following FSSB training [23], but, among the four subdimensions, only creative work-family management demonstrated an indirect effect on performance, suggesting that this dimension may be particularly effective [23]. From an occupational health perspective, workplace interventions are often designed to target specific behaviors or resources. Therefore, analyzing FSSB at the subdimension level is useful for identifying which types of supervisor behaviors are most relevant to enhancing employees’ health and job performance, and for developing more focused and practical interventions.
Job performance is the measurable activities and behaviors that contribute to the achievement of organizational goals [24]. Since job performance is an outcome-based measure that depends on the industry and evaluation criteria, a different approach is needed to understand the effects of FSSB in a variety of workplaces. To address this, we focus on work functioning impairment, which measures employees’ input-side ability to do their work and is less affected by industry-specific evaluation systems [25].
Female employees often bear greater household roles than male employees [26–28]. As a result, they may require more support from supervisors and the workplace to continue working while balancing work and family demands. Supervisor support and workload adjustments reduce WFC, which in turn may improve mental health and job satisfaction [29–31], ultimately helping subordinates maintain effective work functioning [32]. Moreover, female employees are generally more likely to experience WFC than their male counterparts [33]. Therefore, the association between FSSB and work functioning impairment may differ by gender, and it is expected to be particularly strong among female employees.
This study examines the association between FSSB and work functioning impairment among Japanese office workers. We propose the following three hypotheses: Higher FSSB scores will be associated with lower work functioning impairment; gender will moderate the association between FSSB and work functioning impairment, such that the negative association will be stronger for females than for males; and among the four subdimensions of FSSB, certain dimensions will be more strongly associated with lower work functioning impairment. Despite growing interest in FSSB, no studies have examined its association with work functioning impairment. Furthermore, potential gender differences in this association have not been explored. This study aims to provide insights into the role of FSSB in supporting subordinates’ work functioning across diverse occupational settings.
Study Design and Participants
We conducted a cross-sectional study among employees from six large Japanese companies (five manufacturing and one service) that are part of a study group focused on improving employee health in Japan. The survey was administered from July to December, 2020, utilizing an online questionnaire. The survey was administered to all 17,893 employees of the six companies. The 12,027 employees who agreed to participate in the study and responded to the questionnaire were included in the study (response rate 67.2%). All the questions in this online survey were mandatory, so there were no missing data.
A priori sample size calculation was conducted using G*Power (version 3.1.9.7). Assuming a medium effect size (f² = 0.15), α = 0.05, power = 0.80, and 8 predictors (FSSB plus 7 covariates), the required sample size was estimated to be 109 participants. The final sample in this study included approximately 12,027 participants, substantially exceeding the required size and ensuring sufficient statistical power. As the participating companies were predetermined, it was not feasible to select a sample of a specific size from within them. The study protocol was approved by the Ethics Committee of Medical Research at the University of Occupational and Environmental Health, Japan (H26-026). We obtained consent from all the participants regarding the research. We adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.
Family Supportive Supervisor Behaviors
Family Supportive Supervisor Behaviors (FSSB) was measured using the Japanese version (FSSB-J) of Hammer’s 14-item scale [34]. A sample item from the FSSB-J scale is, “My supervisor is willing to listen to my problems in juggling work and non-work life”. Items are rated on a scale from 1 (strongly disagree) to 5 (strongly agree). The total score of the FSSB ranges from 14 to 70. Higher scores indicate a better FSSB. The internal consistency of the FSSB-J scale in this study was high (Cronbach’s α = 0.97). Furthermore, we calculated the mean score for each of the four FSSB subdimensions (emotional support: 4 items, instrumental support: 3 items, role modeling behaviors: 3 items, and creative work-family management: 4 items) and assessed their internal consistency using Cronbach’s α. The Cronbach’s α coefficients for the four subdimensions were 0.95 for emotional support, 0.90 for instrumental support, 0.93 for role modeling behaviors, and 0.91 for creative work-family management, indicating good internal consistency across all subdimensions.
Work Functioning Impairment
Work functioning impairment was assessed using the Work Functioning Impairment Scale (WFun), comprising seven questions [25]. A sample item from the WFun scale reads, “I could not behave socially.” Items are rated on a scale from 1 (not at all) to 5 (almost every day). The total score of the WFun ranges from 7 to 35. Higher scores indicate a greater degree of work functioning impairment. The scale in this sample demonstrated good internal consistency (Cronbach’s α = 0.92).
Covariate variables
We selected age, gender, marital status, presence of children under age 18, presence of intra-family caregivers, and companies as covariates, based on previous studies [35, 36]. Age and gender differences in Japan significantly affect the time spent on housework [35]. Younger and female caregivers tend to take on greater responsibilities within the family [36]. Covariates were treated as continuous variables for age, and categorical variables for the rest: gender (male or female), marital status (married, unmarried, or divorced/widowed), presence of children under 18 (present or absent), presence of intra-family caregivers (present or absent), and companies (A, B, C, D, E, or F)
Statistical analysis
Multivariate regression analysis was performed to evaluate the association between FSSB level and work functioning impairment. The FSSB score was used as the explanatory variable and the WFun score as the outcome variable, with both treated as continuous variables. We calculated the unstandardized partial regression coefficient (B), and 95% confidence interval (CI) for the WFun score in an age-gender adjusted model, and then in a multivariate-adjusted model that also included marital status, presence of children under 18, presence of intra-family caregivers, and companies.
To examine the presence of an interaction between FSSB and gender, we included an interaction term for FSSB and gender in the multivariable-adjusted model and tested its statistical significance. When the interaction was statistically significant, stratified analyses by gender were conducted to evaluate the association between FSSB and work functioning impairment within each group.
As a supplementary analysis, all four subdimension scores were simultaneously entered into the regression model to examine their independent associations with work functioning impairment. Because the subdimensions contained different numbers of items and therefore had different score ranges, standardized regression coefficients (β) were used to remove the effect of scale differences and to enable direct comparison of the relative contribution of each subdimension. We assessed multicollinearity by calculating the variance inflation factor (VIF) for each subdimension.
All analyses were performed using Stata Statistical Software (Release 18; StateCorp LLC, College Station, TX, USA).
Table 1 presents the basic attributes of the participants analyzed. Of the 12,027 individuals, 25.7% were female. The average age was 43.2 years old (SD = 10.4). Regarding marital status, 72.0% were married, and 25.0% were unmarried. Participants with children under the age of 18 represented 46.7% of the sample. Participants with an intra-family caregiver represented 6.5% of the sample. The average the WFun score was 13.9 (SD = 6.0).
| N (%) | Mean (SD) | ||
|---|---|---|---|
| Total | 12,027 | ||
| Age | 43.2 (10.4) | ||
| Gender (female) | 3,089 (25.7%) | ||
| Marital status | |||
| Married | 8,657 (72.0%) | ||
| Unmarried | 3,010 (25.0%) | ||
| Divorced/widowed | 360 (3.0%) | ||
| Children under age 18 | 5,613 (46.7%) | ||
| Intra-family caregivers | 787 (6.5%) | ||
| Company | |||
| A (Manufacturing) | 2,327 (19.3%) | ||
| B (Manufacturing) | 4,103 (34.1%) | ||
| C (Manufacturing) | 1,623 (13.5%) | ||
| D (Services) | 448 (3.7%) | ||
| E (Manufacturing) | 501 (4.2%) | ||
| F (Manufacturing) | 3,025 (25.2%) | ||
| WFun score | 13.9 (6.0) | ||
SD: standard deviation, WFun score: work functioning impairment score.
Table 2 shows the results of a multivariate regression analysis of the association between the FSSB score and the WFun score. In an age- gender adjusted model, the FSSB score was significantly negatively associated with the WFun score (B = –0.11, P < 0.001). The same result was observed in the multivariate-adjusted model (B = –0.10, P < 0.001).
| Age, gender-adjusted | Multivariate-adjusted | Multivariate-adjusted | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B | 95% CI | P value | B | 95% CI | P value | B | 95% CI | P value | ||||
| FSSB score | -0.11 | -0.12 | -0.10 | <0.001 | -0.10 | -0.11 | -0.09 | <0.001 | -0.11 | -0.12 | -0.10 | <0.001 |
|
Gender, female (refer to male) |
-0.01 | -0.26 | 0.24 | 0.929 | -1.25 | -2.25 | -0.25 | 0.014 | ||||
| FSSB score x gender | 0.03 | 0.01 | 0.05 | 0.012 | ||||||||
B: unstandardized partial regression coefficient, CI: confidence interval, FSSB score: family supportive supervisor behaviors score.
Multivariate model was adjusted for age, gender, marital status, children under age 18, intra-family caregivers, and companies.
In analysis to confirm an interaction effect, the interaction term between FSSB score and gender was found to be statistically significantly related to WFun score (FSSB score x gender: P=0.012). The results of the analysis by gender are shown in Table 3. A significant negative association between FSSB and WFun scores was observed in both males and females (male, P < 0.001; female, P < 0.001). In males, there was a large decrease in WFun score when FSSB increased (male: B = –0.11, 95% CI = [–0.12, –0.10]; female: B = –0.08, 95% CI = [–0.10, –0.07]).
| Multivariate-adjusted | |||||
|---|---|---|---|---|---|
| B | 95% CI | P value | |||
| Among male workers | |||||
| FSSB score | -0.11 | -0.12 | -0.10 | <0.001 | |
| Among female workers | |||||
| FSSB score | -0.08 | -0.10 | -0.07 | <0.001 | |
B: unstandardized partial regression coefficient, CI: confidence interval, FSSB score: family supportive supervisor behaviors score.
Multivariate model was adjusted for age, marital status, children under age 18, intra-family caregivers, and companies.
Table 4 shows the results of the multiple regression analysis examining the associations between the four FSSB subdimensions and work functioning impairment. All VIF values were below 5.0, suggesting no serious multicollinearity among the subdimensions, although some degree of correlation may exist. Emotional support, instrumental support, and creative work-family management were significantly associated with lower work functioning impairment (emotional support: β = –0.07, B = –0.43, 95% CI = [–0.63, –0.24], P < 0.001; instrumental support: β = –0.07, B = –0.47, 95% CI = [–0.68, –0.26], P < 0.001; creative work-family management: β = –0.07, B = –0.47, 95% CI = [–0.68, –0.25], P < 0.001), while role modeling behaviors was not statistically significant (β = –0.01, B = –0.06, 95% CI = [–0.25, 0.12], P = 0.509).
| FSSB Subdimension | Age, gender-adjusted | Multivariate-adjusted | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| β | B | 95% CI | P value | VIF | β | B | 95% CI | P value | VIF | |||
| Emotional support | -0.08 | -0.46 | -0.66 | -0.27 | <0.001 | 3.40 | -0.07 | -0.43 | -0.63 | -0.24 | <0.001 | 3.47 |
| Instrumental support | -0.08 | -0.52 | -0.73 | -0.31 | <0.001 | 3.50 | -0.07 | -0.47 | -0.68 | -0.26 | <0.001 | 3.52 |
| Role modeling behaviors | -0.01 | -0.04 | -0.22 | 0.15 | 0.682 | 2.72 | -0.01 | -0.06 | -0.25 | 0.12 | 0.509 | 2.76 |
| Creative work-family management | -0.07 | -0.49 | -0.70 | -0.27 | <0.001 | 3.68 | -0.07 | -0.47 | -0.68 | -0.25 | <0.001 | 3.68 |
β: standardized regression coefficient, B: unstandardized regression coefficient, CI: confidence interval, VIF: variance inflation factor. Multivariate model was adjusted for age, gender, marital status, children under age 18, intra-family caregivers, and companies.
This study found that higher FSSB scores were associated with lower work functioning impairment scores. This association remained significant even after adjusting for gender, age, marital status, presence of children under 18, presence of intra-family caregivers, and companies. The interaction between FSSB and gender was also statistically significant. Stratified analyses showed that higher FSSB scores were associated with lower WFun scores for both males and females, although the association was stronger in males than in females. Among the four subdimensions of FSSB, emotional support, instrumental support, and creative work-family management were significantly associated with lower work functioning impairment, while role modeling behaviors were not. To the best of our knowledge, this is the first study to examine the association between FSSB and work functioning impairment. These findings strongly support the importance of supervisor support in reducing work functioning impairment.
Based on previous research, two possible mechanisms may explain the gradual decrease in work functioning impairment as the FSSB scores increase. The first mechanism is that perception of FSSB helps individuals better balance their work and family roles, thereby reducing work functioning impairment. Work styles, including time management, are important for maintaining this balance [37]. A study of teleworkers found that greater perception of FSSB made it easier to manage work-family boundaries and was linked to higher life satisfaction [38]. Several other studies have shown that higher perception of FSSB is associated with lower WFC and a greater sense of fulfillment in both work and family roles [17, 18, 21, 38, 39]. A link between time management and work functioning impairment has also been reported. For example, a study of nurses in advanced medical institutions in Brazil found that poor time management was related to higher work functioning impairment [40]. Similarly, a study of 21,500 service industry workers in Japan showed that shorter working hours and flexible work arrangements were associated with reduced work functioning impairment [41]. These findings suggest that higher perceptions of FSSB enable subordinates to better manage their work and family roles, potentially leading to reduced work functioning impairment.
The second mechanism is that perceptions of FSSB positively influence subordinates’ mental health, leading to reduced work functioning impairment. Prior studies on FSSB have shown that higher FSSB scores are associated with lower psychological distress [22] and improved mental health among subordinates [21, 42, 43]. Social support may help reduce work functioning impairment [44]. This is supported by a cross-sectional study of employees from 17 companies, which found that supervisor support indirectly affected work functioning impairment by influencing psychological and physical stress responses [45]. These previous studies suggest that higher perceptions of FSSB may positively impact subordinates’ mental health, potentially preventing work functioning impairment. Therefore, an increase in perceptions of FSSB may improve the management of mental health and work-family roles, potentially reducing work functioning impairment. Future research should further investigate these mechanisms.
There was a significant interaction between FSSB and gender in this study. Contrary to expectations, however, high FSSB was more strongly associated with reduced work functioning impairment in males than in females. Family-supportive organizational policies, practices and culture can influence both the actual provision of FSSB and its perception [20]. Perceptions of FSSB are subjectively evaluated by subordinates. In societies characterized by greater gender inequality, subordinates may perceive such support more positively, which can heighten their sense of gratitude and enhance their motivation at work [46]. In contrast, in societies where gender equality is well established, FSSB may be regarded as a natural aspect of balancing work and family life. Although subordinates who recognize FSSB may still feel motivated to work, the magnitude of this effect is likely to be smaller than in societies with greater gender inequality [46].
Japan, the setting of this study, was ranked 118th out of 146 countries in the World Economic Forum’s 2024 Global Gender Gap Report, placing it last among the Group of Seven nations [47]. The Japanese government has been implementing various policies to foster a society in which all citizens can work comfortably, and, overall, society is gradually progressing toward greater gender equality [48]. Understanding and support for working females, both at home and in the workplace, have expanded, and females themselves are increasingly able to choose the type and extent of support they receive [49]. In contrast, although working males are increasingly taking on more family roles, their work roles have not diminished [27]. Moreover, despite society’s overall progress toward enabling all individuals to balance work and family life, understanding and support for males in the workplace remain insufficient. In 2020, when this survey was conducted, the rate of males taking childcare leave was only 12.7% [50]. Males who are hesitant to seek job adjustments for family reasons may come to have perceptions of FSSB once they become aware of it, which could, in turn, lead to meaningful reductions in work functioning impairment.
The analysis of the four FSSB subdimensions showed that emotional support, instrumental support, and creative work-family management were significantly associated with work functioning impairment. Emotional support and instrumental support are shown to reduce WFC through direct assistance and work adjustments, thereby alleviating subordinates’ psychological burden [51]. Creative work-family management refers to supervisors’ behaviors that involve redesigning work and optimizing team functions to enable subordinates to balance their work and family roles [17]. Supervisors who take actions to enable their subordinates to work effectively both at work and in their personal lives help reduce WFC [52, 53].
Role modeling behaviors showed no significant association with work functioning impairment in this study. Role modeling behaviors refer to supervisors showing, through their own actions, how to manage work and family responsibilities [17]. Previous longitudinal research has also reported that role modeling behaviors have a weaker correlation with subordinate job performance, as evaluated by supervisors, compared with other subdimensions of FSSB [23]. One possible explanation is that even if supervisors act as role models for work-family balance, their ways of balancing work and family may not necessarily match the ways subordinates wish to achieve such balance [54]. Nevertheless, when supervisors engage in role modeling behaviors based on moral principles and values, these behaviors may increase subordinates’ trust in them [55]. A meta-analysis examining the mechanisms through which FSSB influences outcomes identified the development of trust between supervisors and subordinates as a key pathway by which FSSB enhances work outcomes such as performance [56]. Future research should therefore examine the extent to which role modeling behaviors contribute to the development of supervisor-subordinate trust, which may help clarify why no association with work functioning impairment was observed in this study.
There are some limitations in this study. First, because FSSB reflects supervisors’ supportive behaviors and is likely influenced by smaller units such as departments or teams, a multilevel analysis that accounts for such nested structures would be preferable. However, this study did not collect data at these more detailed hierarchical levels. To address this limitation, we included the six companies as categorical variables in the analytical model and adjusted for covariates to reduce confounding due to inter-company differences. This approach allowed us to examine the association between FSSB and work functioning impairment while minimizing the influence of company-level variation. In addition, since the ICC was relatively small (0.03), indicating limited similarity in data across companies, we used regression analysis rather than multilevel analysis. Future research should collect data at the department and team levels to enable a more precise evaluation of these nested effects.
Second, the generalizability of the findings may be limited because this study was conducted in large Japanese companies during the early stages of the COVID-19 pandemic. Cultural, economic, and political contexts influence awareness of gender equality [57]. During the pandemic, stay-at-home measures and increased time spent at home may have reinforced gender imbalances in the division of household chores compared with pre-pandemic conditions [58]. This created a unique context in which subordinates’ perceptions of FSSB may have differed from those under normal circumstances. In addition, variations in job type and employment status (e.g., part-time employment) can also influence subordinates’ perceptions of FSSB [22, 23, 59]. Further research should therefore be conducted with a wider range of participants across different organizational settings and time periods.
Third, this study was a cross-sectional observational study, and therefore it was not possible to determine a causal relationship between FSSB perception and work functioning impairment. As with other observational studies, the possibility of unmeasured confounding factors cannot be completely ruled out. Although we adjusted for important background characteristics related to work and family life, there may be other factors that influenced the results. Future studies should employ longitudinal designs with a wider range of variables to strengthen the findings.
Fourth, this study had a response rate of 67.2%, indicating that 32.8% of employees did not participate. This raises the possibility of non-response bias, which could have resulted in either an overestimation or underestimation of the observed association between FSSB and work functioning impairment. Since we were unable to obtain attribute information on non-respondents from the participating companies, it remains unclear what kind of employees chose not to respond. However, the gender distribution in our sample (25.7% female) is generally consistent with the proportion of female employees in Japanese manufacturing industries (29.3%) [60], suggesting that the overall representativeness may not have been substantially affected. Nonetheless, the potential influence of non-response should be acknowledged when interpreting the findings.
Fifth, the participants in this study greatly exceeded the calculated sample size. A large sample size is advantageous because it increases statistical power and enhances the generalizability of the findings, but, although a statistically significant association was observed between FSSB and work functioning impairment, the effect size was small, suggesting that the practical significance of this association is limited. Therefore, in practice, it is important to implement FSSB as a long-term and continuous measure, to target groups that are more likely to benefit (e.g., men), and to seek synergistic effects by combining it with other related initiatives.
Sixth, a self-administered questionnaire might introduce self-selection and recall biases. However, since the FSSB is the subordinates’ perception of the supportive supervisors’ behaviors and work functioning impairment is a subjective perception of one’s own input-side ability to do their work, it is an appropriate measure based on self-evaluation.
This study analyzed data from 12,027 employees of six Japanese private companies using a self-administered questionnaire. The findings indicated that higher FSSB was associated with lower work functioning impairment, with the association stronger among males than among females. Furthermore, among the four subdimensions of FSSB, emotional support, instrumental support, and creative work-family management showed particularly strong associations with reduced work functioning impairment. These findings suggest that enhancing supervisors’ understanding and practice of FSSB may be an effective strategy for reducing work functioning impairment in organizations with diverse workforces, including both male and female employees. Overall, the findings of this study provide valuable insights for companies seeking to develop strategies for creating a more supportive and employee-friendly workplace. Future research should further explore the underlying mechanisms of these associations and examine their applicability in different organizational and cultural contexts.
We thank the companies and health insurance associations affiliated with the Collabo-Health Study Group for their cooperation in this study, and EWEL, Inc. for its secretariat function.
This study was supported and funded by a research grant from the Collabo-Health study group (no grant number). The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication.
Dr. Nagata T reports personal fees from BackTech Inc., EWEL Inc. and Sompo Health Support Inc., outside the submitted work. Dr. Mori reports personal fees from BackTech Inc. and Sompo Health Support Inc., outside the submitted work. Dr. Fujino has the copyright to WFun with royalties paid from Sompo Health Support Inc., outside from this work. The other authors declare no conflicts of interest associated with this manuscript.
The data that support the findings of this study are available from the corresponding author, Tomohisa Nagata, upon reasonable request.
Conceptualization: Mika Kawasumi
Funding acquisition: Mika Kawasumi, Tomohisa Nagata
Investigation: Mika Kawasumi, Tomohisa Nagata, Koji Mori
Supervision: Yoshihisa Fujino, Koji Mori
Visualization: Mika Kawasumi, Tomohisa Nagata
Writing-original draft: Mika Kawasumi
Writing-review & editing: Mika Kawasumi, Tomohisa Nagata, Kiminori Odagami, Masako Nagata, Shigeyuki Kajiki, Yoshihisa Fujino, Koji Mori
The study protocol was approved by the Ethics Committee of Medical Research at the University of Occupational and Environmental Health, Japan (H26-026)
Participants were asked to check a consent box when completing the questionnaire.
記事関連データはJ-STAGE Dataで利用できます。（産業医科大学雑誌 J-STAGE Data） / The data analysis file and all annotator data files are available in J-STAGE Data. (J-STAGE Data from Journal of UOEH)
