Help-Seeking among Male Employees in Japan: Influence of Workplace Climate and Distress.

OBJECTIVES: Although using mental health services is an effective way to cope with work-related stressors and diseases, many employees do not utilize these services despite service improvements in recent years. The present study aimed to investigate the interaction effects of workplace climate and distress on help-seeking attitudes, and elucidate the reasons for mental health service underutilization in Japan.
METHODS: A questionnaire was distributed to 650 full-time male Japanese employees. Hierarchical multiple regression analysis was used to investigate interaction effects of workplace climate and distress on help-seeking.
RESULTS: Results showed that the association between workplace climate and help-seeking attitudes differed depending on employee distress level. For employees experiencing low levels of distress, openness to seeking treatment increased with a higher evaluation of the mental health services available at the workplace. However, the same did not hold true for employees experiencing high levels of distress. Instead, openness to seeking treatment decreased with perceived risk for career disadvantage for high distress employees. Additionally, negative values for seeking treatment in highly distressed employees decreased only when services were perceived as valuable, and the risk to their career was perceived as low.
CONCLUSIONS: Overall, these findings indicate that distress distorts the perception of social support, which may lead to underutilization of available services. Assessing employees' distress levels and tailoring adequate interventions could facilitate help-seeking in male employees.

Introduction

Work stress is a factor that significantly influences the mental health of employees[1)]. Common stressors in the workplace include long work hours, workload pressure, feelings of lack of control, and poor workplace support[2)]. Failure to cope with work-related stressors causes a broad range of distressing consequences, such as abnormal heart rate, substance dependence, and depressive mood, which are in turn associated with cardiovascular disease and mental illness[3)]. Additionally, work stress contributes to increased sick leave and lowered work performance resulting in a considerable economic burden. This is true for both Asian and Western countries[4)].

Undertaking psychotherapy, which includes behavioral therapy, cognitive behavioral therapy, psychoanalysis, and psychodynamic psychotherapy, has proved to be an effective strategy to cure psychological symptoms[5)]. Furthermore, it has been shown that mental health interventions are significantly effective in reducing economic burdens resulting from work stress, even when we take into account the cost of the psychological treatment[4)]. Thus, seeking help from mental health services and getting professional support could be an effective way of dealing with work stress.

However, the majority of those who suffer psychological problems do not utilize mental health services[6)] despite the increasing awareness and improvement of mental healthcare in the workplace in recent years[7)]. Moreover, the utilization of mental health services among employees has been found to be even lower when compared to non-employees, including unemployed individuals, students, and retired individuals[8)]. Previous studies show that only 20% of employees who suffer from major depressive disorder receive adequate treatment in the U.S.[9)], and similar results were obtained in European studies[8)] and Asian populations[10)]. Similarly, very few employees utilize adequate medical services in Japan[11)], while more than 50% of them suffer from severe work stress[12)]. Moreover, neither the rate of employees who suffer from work stress nor the usage of mental health services has improved, in spite of an increase in mental healthcare access in working environments in recent years[12)]. In other words, there seems to be a gap between service improvement and continued underutilization, which needs to be filled by facilitating help-seeking toward mental health services by employees.

In investigating help-seeking toward mental health services, numerous studies have examined attitudes toward seeking professional psychological help, the major predictors of help-seeking behavior[13)]. Attitudes Toward Seeking Professional Psychological Help: A Shortened Form (ATSPPH-SF[14)]) is a well standardized and most widely used scale, which consists of ten items and one factor. On the other hand, the recent study has indicated the scale with a two-factor structure, ATSPPH-SF consists of "Openness to seeking treatment for emotional problems" and "Value and need in seeking treatment" [15)].

Studies have shown that employees' emotional openness is influenced by the support climate of workplace[13)], and personal values are interrelated with organizational values[16)]. These findings imply that perceived support or workplace climate, rather than the availability of services, has great influence on seeking psychological help. Workplace Mental Health Climate (WMHC) or "shared social attitudes and norms towards mental health among organization members," is a scale that measures mental health care and climate in the workplace from the employees' viewpoint, based on the theory that an organizational climate is composed of the aggregation of individual psychological climates[17)]. WMHC consists of three factors: Evaluation of Services, Risk for Career Disadvantage, and Understanding Mental Health. Of those three, Evaluation of Services: "cognition and evaluation of mental health care effort in the workplace," and Risk for Career Disadvantage: "perceived risk and concern about damage of one's career or losing face and future opportunities because of suffering from mental illness" represent perceived support climate and value for mental health at the workplace. On the other hand, Understanding Mental Health represents "employee's personal values regarding mental health," and previous study have indicated qualitative differences based on biased distribution and low correlations between this scale and the other two subscales[17)]. Although Evaluation of Services and Risk for Career Disadvantage may link to attitudes toward psychological help, the association among them has not been revealed in previous studies.

Moreover, the studies have found that physical and psychological distress, which is a prior condition to seeking help[18)], is associated with less perception of available support[19)]. Other studies have found that increased prior stress symptoms had a negative influence on the perception of an organizations' support (e.g.,[20)]). These studies imply that distress may lead to a failure in cognitive appraisal of available resources, such as underestimating Evaluation of Services or overestimating Risk for Career Disadvantage, leading to underutilization of services.

Taken together, it is assumed that positive WMHC (high Evaluation of Services and low Risk for Career Disadvantage) is associated with positive attitudes toward psychological help, although the association may be moderated by individual distress levels. Regarding gender differences, previous studies have revealed that perception of social support[25)] and help-seeking[8],[14)] is lower in males compared to females. Moreover, male help-seeking behavior relies more on organizational norms when compared to female help-seeking[27)]. Therefore, the present study focused on male employees in Japan. The purpose of this study was to investigate the interaction effects of WMHC and distress on attitudes toward professional psychological help, and elucidate the reasons for the underutilization of mental health services among distressed employees. Understanding the complexity may contribute to exploring how to facilitate help-seeking in employees.

The hypotheses were: (H1) in employees with low distress levels, positive WMHC (high Evaluation of Services and low Risk for Career Disadvantage) is associated with positive attitudes toward professional psychological help, (H2) in employees with high distress levels, positive WMHC (high Evaluation of Services and low Risk for Career Disadvantage) is not associated with positive attitudes toward professional psychological help.

Methods

Subjects

Questionnaires were distributed to 1,100 Japanese working males. Of those 757 were returned, which means that the response rate was 68.8%. By excluding part-time employees and not fully answered questionnaires, of these 650 full-time employees were used for the subsequent analysis. The participants' ages ranged from 20 to 65 years (M = 42.04, SD = 9.66), 44.5% were employed in manufacturing, and 55.5% were employed in other industries. Regarding job title and role, 29.4% were entry-level, 13.7% were assistant managers, 13.4% were subsection chiefs, 22.2% were managers, 13.4% were general managers, 4.9% were presidents, and 13.5% were "other." Company size was indicated by the total number of employees: 17.2% worked in a company with less than 30 employees, 30.9% worked with 30-99 employees, 24% worked with 100-999 employees, and 27.8% worked in a company with more than 1,000 employees.

Measures

Attitudes Toward Seeking Professional Psychological Help

Help-seeking towards mental health services was assessed via the Japanese version of ATSPPH-SF[23)]. The original scale consists of ten items (rated on a 6-point Likert scale from 1 = absolutely disagree to 6 = strongly agree) and a single factor, while two-factor structure has also been revealed[15)]. Since the factor structure depends on religion, illness attribution, and culture[24)], the present study assumed one or two factors, and analyzed exploratory. Confirmatory factor analysis with one factor showed a rather low fit (GFI = 0.863, AGFI = 0.785, CFI = 0.785, RMSEA = 0.133), thus exploratory factor analysis, employing maximum likelihood estimation with promax rotation was conducted. The analysis yielded a two-factor structure (GFI = 0.933, AGFI = 0.892, CFI = 0.902, RMSEA = 0.091), which is congruent with Elhai et al.[15)]. Therefore, the factors were named Openness to Seeking Treatment (OST) and Negative Value for Seeking Treatment (NVST) in accordance with the previous study[15)]. OST represents "openness to seeking treatment for emotional problems," and consists of five positive statements such as "If I believed I was having a mental breakdown, my first inclination would be to get professional attention." NVST represents "negative values and lack of needs in seeking treatment" and consists of five negative statements such as "A person should work out his or her own problems; getting psychological counseling would be a last resort." The Cronbach's alpha was 0.799 for OST and 0.680 for NVST.

Workplace Mental Health Climate

The Workplace Mental Health Climate Scale[17)] was used to measure WMHC. The scale originally consisted of 16 items (rated on a 5-point Likert scale from 1 = disagree to 5 = agree), and 11 items of Evaluation of Services and Risk for Career Disadvantage were used in the present study. However, confirmatory factor analysis showed a rather low fit (GFI = 0.870, AGFI = 0.825, CFI = 0.840, RMSEA = 0.095), possibly because the present study targeted only males. Therefore, exploratory factor analysis, employing unweighted least squares estimation with promax rotation was conducted. The result showed a two-factor structure, while one item which had factor loadings under 0.350, and another item which was cross-loading onto two factors were excluded (GFI = 0.920, AGFI = 0.886, CFI = 0.899, RMSEA = 0.081). Because only one item was altered besides the excluded factors, the original names of the subscales were maintained. The Cronbach's alpha was 0.867 for Evaluation of Services and 0.724 for Risk for career disadvantage.

Hopkins Symptom Checklist

The Japanese version of the Hopkins Symptom Checklist (HSC;[25)]) was used to measure distress levels. The scale was originally constructed by Derogatis and colleagues[26)]. The Japanese version adopted the Psychosomatic score, Depressive score, and Anxiety score, which had high reliability among the original five subscales. The scale consists of 30 items (rated on a 5-point Likert scale from 1 = never to 5 = frequent). In order to investigate the effect of distress comprehensively, subsequent analysis used 30 items, integrating Psychosomatic, Depressive, and Anxiety scores. The proportion of variance explained for one factor was 38.8%, and the Cronbach's alpha was 0.947.

Procedure

The data were collected in May-June of 2013 at the following two locations: 1) classes on a campus at a junior high school and a high school, and 2) 41 companies in central Japan. At the schools, teachers distributed questionnaires to students. The students were asked to hand the questionnaire to a working male friend or family member who was aged 20 to 65 years and ask them to complete it. At the different companies, delegates distributed the questionnaires, and employees were asked to return them by post when completed.

The study design was approved by the Ethical Review Board, Graduate School of Education and Human Development, Nagoya University. The questionnaires were anonymous and submitted in sealed envelopes for confidentiality. Participants answered the questionnaire upon agreement to the gist of the study.

Statistical analysis

Data were analyzed by using IBM SPSS Statistics 18 software. Firstly, Pearson's correlation coefficients were calculated between the total score of ATSPPH-SF, WMHC, distress, and demographic variables in order to select factors to be controlled. The total scores were adopted in order to unify the control factors between the analyses. The results showed that age, title, and company size were significantly correlated with the total score of ATSPPH-SF and WMHC. Therefore, age, title, and company size were included in the analysis to control confounding effects. Secondly, independent variables were converted into a z-score, and hierarchical multiple regression analysis was conducted separately on OST and NVST. In order to confirm the adequacy of the model by examining the change of coefficient of determination (ΔR2), age, title, company size, Evaluation of Services, Risk for Career Disadvantage, and distress were entered as independent variables at Step 1, and the multiplied variables of Evaluation of Services, Risk for Career Disadvantage, and distress were entered at Step 2. When the result showed significant interaction effects, further analyses were conducted to draw the regression lines by plugging the values of ± 1SD.

Results

Correlations between the variables

Pearson's correlation coefficients were calculated to investigate correlations between the variables (Table 1). The results showed significant negative correlations between subscales of help-seeking attitude (r = -0.396, p< 0.001) and WMHC (r = -0.427, p < 0.001). Additionally, distress showed significant positive correlations with OST (r = 0.152, p < 0.001) and Risk for Career Disadvantage (r = 0.273, p < 0.001), and a negative correlation with Evaluation of Services (r = -0.225, p < 0.001).

Table 1.

Correlation between the variables

	M	(SD)	Help–seeking attitude		WMHC		Distress
			OST	VNST	Evaluation	Risk
*p<0.05, **p<0.01, ***p<0.001
OST	2.93	(0.951)	–	0.396***	0.086*	–0.037	0.152***
VNST	3.47	(0.903)		–	–0.085*	0.105**	0.030
Evaluation of Services	2.89	(1.061)			–	–0.427***	–0.225***
Risk for Career Disadvantage	2.92	(0.773)				–	0.273***
Distress	2.06	(0.653)					–

Hypothesis testing

Interaction effects on OST

The results showed a significant two-way interaction effect for Evaluation of Services × distress (β = -0.116, p < 0.01; Fig. 1) and Risk for Career Disadvantage × distress (β = -0.088, p < 0.05; Fig. 2). The change of coefficient of determination between Step 1 and Step 2 was significant (ΔR2 = 0.011, p < 0.05) and multicollinearity was not present (Table 2). Thus, adding interaction terms contributed to better explanation of OST. Further analysis showed regression coefficients differ by level of distress: high distress (+1SD) and low distress (-1SD). The adjusted coefficient of determination was R'2 = 0.060.

Fig. 1.

Two-way interaction of "Evaluation of services" × "distress" on "OST"

Fig. 2.

Two-way interaction of "Risk for career disadvantage" × "distress" on "OST"

Table 2.

Multiple regression analysis on OST

	Step1						Step2
Predictor	β		95% confidence interval		collinearity statistics		β		95% confidence interval		collinearity statistics
			Lower	Upper	Tol.	VIF			Lower	Upper	Tol.	VIF
Tol. =Tolerance*p<0.05, **p<0.01, ***p<0.001
Age	0.010	*	0.001	0.019	0.753	1.32	0.011	**	0.002	0.019	0.745	1.34
Title	0.013		–0.034	0.060	0.763	1.31	0.014		–0.033	0.061	0.762	1.31
Company size	0.038		–0.020	0.096	0.689	1.45	0.031		–0.027	0.089	0.681	1.47
Evaluation of Services	0.058		–0.036	0.152	0.582	1.72	0.052		–0.043	0.147	0.563	1.78
Risk for Career Disadvatage	–0.044		–0.127	0.040	0.731	1.37	–0.075		–0.163	0.013	0.650	1.54
Distress	0.177	***	0.102	0.252	0.909	1.10	0.149		0.068	0.230	0.772	1.30
Evaluation×Risk							–0.052		–0.125	0.021	0.737	1.36
Evaluation×Distress							–0.116	**	–0.201	–0.032	0.720	1.39
Risk×Distress							–0.088	*	–0.173	–0.004	0.641	1.56
Evaluation×Risk×Distress							–0.052		–0.126	0.021	0.566	1.77
ΔR2	0.049***						0.011*
Adjusted R2	0.049***						0.060***

Interaction effects on NVST

The results showed a significant three-way interaction effect for Evaluation of Services × Risk for Career Disadvantage × distress (β = -0.113, p < 0.01; Fig. 3). The change of coefficient of determination between Step 1 and Step 2 was significant (ΔR2 = 0.019, p < 0.01), and multicollinearity was not present (Table 3). Thus, adding interaction terms contributed to better explanation of NVST. Further analysis showed regression coefficients differed by level of distress and Risk for Career Disadvantage; high distress (+1SD) combined with high/low Risk for Career Disadvantage (± 1SD), and low distress (-1SD) combined with high/low Risk for Career Disadvantage (±1SD). The adjusted coefficient of determination was R'2 = 0.056.

Fig. 3.

Three-way interaction of "Evaluation of services" × "Risk for career disadvantage" × "distress" on "NVST"

Table 3.

Multiple regression analysis on NVST

	Step1						Step2
Predictor	β		95% confidence interval		collinearity statistics		β		95% confidence interval		collinearity statistics
			Lower	Upper	Tol.	VIF			Lower	Upper	Tol.	VIF
Tol. =Tolerance*p<0.05, **p<0.01, ***p<0.001
Age	–0.013	**	–0.021	–0.005	0.753	1.33	–0.013	**	–0.021	–0.005	0.745	1.34
Title	–0.017		–0.062	0.028	0.763	1.31	–0.015		–0.060	0.029	0.762	1.31
Company size	–0.049		–0.105	0.006	0.689	1.45	–0.055		–0.110	0.000	0.681	1.47
Evaluation of Services	0.017		–0.072	0.107	0.582	1.71	0.000		–0.091	0.090	0.563	1.78
Risk for Career Disadvatage	0.084		0.005	0.164	0.731	1.37	0.137	**	0.053	0.221	0.650	1.54
Distress	–0.001		–0.072	0.071	0.909	1.10	0.029		–0.048	0.106	0.772	1.30
Evaluation×Risk							0.124	***	0.055	0.194	0.737	1.36
Evaluation×Distress							–0.069		–0.150	0.011	0.720	1.39
Risk×Distress							0.047		–0.034	0.127	0.641	1.56
Evaluation×Risk×Distress							0.113	**	0.043	0.182	0.566	1.77
ΔR2	0.037***						0.019**
Adjusted R2	0.037***						0.056***

Discussion

The aim of the present study was to investigate the reasons for the underutilization of mental health services among male employees in Japan. The main results showed that the relationship between WMHC and help-seeking attitudes differed depending on distress level.

The results supported our first and second hypotheses. OST increased along with higher Evaluation of Services, only if distress was low. As perception of available support facilitates psychological service use[27)], we hypothesize that employees are more likely to seek help if they hold the available services in high regard. By contrast, OST of the high distress group did not improve regardless of their Evaluation of Services. It is known that depressed individuals tend to a have a negative perception of their environment and future, resulting in a negative view of social situations and a belief that their current situation will not improve, even with effort to change it[28)]. Other studies have found that depressive symptoms are associated with subjective cognitive complaints and actual cognitive impairments (e.g.,[29)]). Baumeister and colleagues[30)] explained that cognitive processes will be impaired when individuals have to manage negative emotion because executive resources will be monopolized by regulating affect. Thus, it is possible that distress inhibits either recognizing available support or evaluating coping options, leading to no improvement in OST. On the other hand, OST was relatively high with the high distress group and a positive correlation was shown between distress and OST, which corresponds to what other previous studies found (e.g.,[31)]). Taken together, these results imply that highly distressed individuals retain a potential need for professional psychological help. However, current mental health care usage[6],[12)] implies that high levels of distress will not necessarily lead to help-seeking behavior in real-life situations.

Results from Risk for Career Disadvantage contrasted with our original hypothesis. OST rose only with the high distress group, as Risk for Career Disadvantage decreased. As it positively correlates with Risk for Career Disadvantage, distress makes people particularly sensitive to social exclusion or rejection[32)]. Concerns about treatment or evaluation in the social group present a significant barrier to utilization of mental health services[33)]. On the other hand, highly distressed individuals are most in need of professional psychological help. Thus, they would be willing to open their emotional problems when the workplace climate was supportive and free from risk of social exclusion. However, OST remained low with the low distress group. This result implies that Risk for Career Disadvantage is a significant factor only when employees have a need for mental health care, and the concern becomes a pressing issue with newfound reality.

The association between WMHC and NVST was also incongruent with our initial hypothesis. NVST declined only when Evaluation of Services was high and Risk for Career disadvantage was low with the high distress group, whereas the other three groups remained at a high level. This implies that negative values for seeking professional help is widely shared at least among male employees. Walton[34)] revealed that employees are concerned with negative judgments from their managers or colleagues for seeking mental health services. In addition, previous studies have found that seeking psychological help tends to create a conflict with traditional male stereotyped roles[35)]. For these reasons, male employees, especially those with high perceived risk for their career, may hesitate when seeking mental health services. On the other hand, results showed that employees who experience high distress do not refuse services that are perceived as valuable and when the risk of social judgment is low. This implies that potential need for help may take priority over resistance arising from negative affect and cognition if the workplace environment is favorable.

Piecing our results together, the mechanism of mental health service underutilization by employees could be described as follows: with the low distress group, negative perspective and value for seeking professional psychological help seem deep-rooted and persists even with highly evaluated services and low risk to one's career. Although inclination to open emotional problems to psychological help rose with evaluation of workplace mental health care, it remained relatively low compared to the high distress group. Thus, employees with low distress are likely to have a conflict when seeking professional psychological help, which inhibits them from seeking help in the early stages of distress. However, when employees have help needs such as high distress, service evaluation is no longer a facilitator of service use. Openness to seeking psychological help is suppressed by perceived risk for career disadvantage, and negative value for help-seeking do not decline until perceived risk for career is low and service evaluation is high. As a result, very few male employees seek support from mental health services. These findings indicate that distress impairs accurate perception of available support and activates excessive sensitivity to fears of negative evaluation for seeking professional psychological help. This distorted cognition may lead to a negative spiral of failure of coping and worsened outcomes.

As a result of these findings, the different associations between workplace support and help-seeking point to the necessity for interventions tailored to each individual's distress levels. An approach for employees with low distress should include improving service evaluation, which would facilitate intention to seek help. Moreover, it is important to find a solution to reduce feelings of resistance towards mental health services. For employees with high distress, on the other hand, intervention needs to be more aggressive. The workplace should endeavor to reduce concerns about career disadvantage while enhancing standards of mental health services. For instance, showing managers' or chiefs' positive attitudes toward mental health care may be effective, as it has been shown that employees are more likely to use counseling services if their workplace managers are more supportive of the service[34)]. In addition, the results indicate the importance of early service use, since appropriate appraisal and selection of coping strategies will be difficult in later stages of distress. Therefore, informing employees of the importance of early help-seeking behavior and describing the negative spiral mechanism that may occur because of distress, may be an effective means to prevent further deterioration of employee mental health.

In spite of these findings, some limitations have to be noted in this study. First, the causal relationships of workplace climate, distress, and help-seeking attitudes cannot be determined, since the present study was a cross-sectional design. Although it was a meaningful step to investigate our hypothesis with a large sample, other study designs or analytical methods that can clarify precise relationships of the factors should be conducted in the future.

Second, the workplace mental health services were evaluated by the employees' subjective views, and the actual effort of the workplace healthcare support systems were not measured. Since WMHC is said to correlate with the actual degree of implementation of mental health services in the workplace[17)], actual conditions may influence WMHC and, therefore, help-seeking. Future research should investigate the effect of both an objective and subjective workplace environment.

Third, although results were statistically significant, the multiple regression coefficient and coefficient of determination were low. As previous studies have pointed out, help-seeking is influenced by a number of factors, such as stigma, past experiences with help-seeking, and knowledge etc.[27)]. Based on this complexity, the present study focused on the interaction effects of workplace climate and distress in particular. Additionally, Cronbach's alpha for ATSPPH-SF was not very high; thus, reliability and validity among male employees may not be adequate. Therefore, it is necessary to consider the smallness of the value and limitations of the measurement when the findings in this study are utilized.

Forth, distress states may have biased responses and results. Given that high distress impairs accurate perception, it may impair the validity of the responses of highly distressed individuals. In contrast, individuals experiencing low distress may not need help or ever experience conflict relating to seeking help. Therefore, the results must be interpreted carefully, and further study is needed to clarify the impact of distress.

Conclusions

The present study aimed to investigate the interaction effects of workplace climate and distress on help-seeking attitudes, and elucidate the reasons for underutilization of mental health service by employees. The results indicate that the associations between workplace climate and attitudes toward professional psychological help differ, depending on individuals' distress levels. Moreover, the findings imply that individuals who experience low levels of distress are inhibited by negative value for seeking help, while high distress individuals have difficulty in perceiving available support. Developing interventions tailored to individuals' distress levels may be effective in facilitating seeking out support from mental health services.

Acknowledgments: This research was supported, in part, by a grant from the Japan Society for the Promotion of Science under Grant 11317.

Conflicts of interest: The authors declare that there are no conflicts of interest.