24
Causal relationship between occupational dysfunction and depression in healthcare workers: A study using structural equation model Mutsumi Teraoka, Makoto Kyougoku Purpose: The purpose of this study is to identify the impacts of occupational dysfunction on depression in healthcare workers (nurses, physical therapists, and occupational therapists) in hospitals. Methods: Healthcare workers responded to a questionnaire based on the Classification and Assessment of Occupational Dysfunction (CAOD) and Center for Epidemiologic Studies Depression Scale (CES-D). CAOD and CES-D were examined using the following methods: descriptive statistics, exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and a causal sequence model. Results: CFA of CAOD had 16 items and 5 factors (CFI=0.958, TLI=0.946, RMSEA0.092). CFA of CES-D had 20 items and 4 factors (CFI=0.950, TLI=0.942, RMSEA=0.060). The results suggest that occupational dysfunction had positive causal effects on depression (CFI=0.926, TLI=0.920, RMSEA=0.059). Conclusion: This model refers to the relationship between depression and occupational dysfunction. Therefore, assessment and intervention on classification of occupational dysfunction for healthcare workers would be beneficial in the prevention of depression. PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015 PrePrints

Causal relationship between occupational dysfunction and ... · Causal relationship between occupational dysfunction and depression in healthcare workers: A study using structural

Embed Size (px)

Citation preview

Causal relationship between occupational dysfunction anddepression in healthcare workers: A study using structuralequation modelMutsumi Teraoka, Makoto Kyougoku

Purpose: The purpose of this study is to identify the impacts of occupational dysfunctionon depression in healthcare workers (nurses, physical therapists, and occupationaltherapists) in hospitals.Methods: Healthcare workers responded to a questionnaire based on the Classificationand Assessment of Occupational Dysfunction (CAOD) and Center for Epidemiologic StudiesDepression Scale (CES-D). CAOD and CES-D were examined using the following methods:descriptive statistics, exploratory factor analysis (EFA), confirmatory factor analysis (CFA),and a causal sequence model.Results: CFA of CAOD had 16 items and 5 factors (CFI=0.958, TLI=0.946, RMSEA=0.092).CFA of CES-D had 20 items and 4 factors (CFI=0.950, TLI=0.942, RMSEA=0.060). Theresults suggest that occupational dysfunction had positive causal effects on depression(CFI=0.926, TLI=0.920, RMSEA=0.059).Conclusion: This model refers to the relationship between depression and occupationaldysfunction. Therefore, assessment and intervention on classification of occupationaldysfunction for healthcare workers would be beneficial in the prevention of depression.

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

1

Title: Causal relationship between occupational dysfunction and depression in

healthcare workers: A study using structural equation modeling

Authors: Mutsumi Teraoka 1, 2, Makoto Kyougoku 3

Affiliations:

1 Doctor Course, Graduate School of Health Sciences, KIBI International University,

Okayama, Japan

2 Oosugi Hospital, Okayama, Japan

3 Department of Occupational Therapy, School of Health Sciences, KIBI International

University, Okayama, Japan

Location:

8, Iga-machi, Takahashi city, Okayama, 716-8508, Japan

Corresponding Author

Name: Mutsumi Teraoka

Email: [email protected]

Phone: +81-866-22-9091

Competing Interests

The authors have declared that no competing interests exist.

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

2

Introduction

Occupational dysfunction is recognized as a major health-related problem in

workers by occupational therapists (Maynard 1986; Scaffa & Reitz 2013). Occupational

dysfunction is defined as a category of difficulties faced by individuals when

performing daily activities in the work environment, including occupational

marginalization, occupational imbalance, occupational alienation, and occupational

deprivation (Teraoka 2014). Occupational marginalization is defined as not being

afforded the opportunity to engage in daily activities (Townsend & Wilcock 2004).

Occupational imbalance is defined as a loss of balance in the engagement of daily

activities (Anaby et al. 2010). Occupational alienation is defined as not satisfying one’s

inner needs through daily activities (Bryant et al. 2004). Occupational deprivation is

defined as a loss of choice and opportunities in daily activities, which are beyond the

control of the individual(Whiteford 2000). These problems are recognized as

health-related risk factors for workers.

It has been pointed out that occupational dysfunction arises without apparent

medical disease (Kyougoku 2010). According to an estimate of an observational study

on workers without obvious medical disease, 36% of workers have some occupational

dysfunction (Akiyama 2010). Regarding occupational alienation, 43% of workers have

reported experiencing a serious problem (Akiyama 2010). In other words, workers have

reported experiencing psychological stress. Moreover, a report found that occupational

dysfunction was observed in 75.4% of rehabilitation therapists in hospitals without

obvious medical disease, and occupational dysfunction showed a correlation with job

stress (Miyake 2014). Previous study indicates that healthcare workers experience

occupational dysfunction and various levels of stress more frequently than other

professionals (Akiyama 2010; Miyake 2014).

Healthcare workers have high rates of anxiety, burnout, depression, substance

abuse, and suicide related to strong stress levels on the job (Dyrbye et al. 2008; Harry

2013). In particular, depression is caused by an increase in job stress, and is recognized

worldwide as a major health-related problem (Irvine 1997; Van Praag 2004). In Japan,

more than 60% of workers are reported to suffer from stress (Honda et al. 2014). In

Japanese society, there is a recognized association between depressive mood and

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

3

subsequent suicide among workers (Takeuchi & Nakao 2013; Tamakoshi et al. 2000).

One of the causes of depression in workers is attributed to difficult working conditions,

such as heavy overtime work, understaffing, time pressure, relationship problems, and

cost-cutting (Denton et al. 2002; Kato et al. 2014; Schaefer & Moos 1993; Seki &

Yamazaki 2006). Many Japanese workers are workaholics, which leads to fatigue, and

this is also one of the causes of depression (Matsudaira et al. 2013; Seki & Yamazaki

2006). There has been concern about depression, especially among healthcare workers,

because depression is one of the most common work-related health problems in

healthcare (Health 2008).

However, no previous study has examined the impact of occupational

dysfunction on depression. A case study and theoretical study on occupational therapy

have suggested a causal linkage between occupational dysfunction and depression (Ishii

2010; Kyougoku 2010). As stated above, it has been pointed out that occupational

dysfunction arises without medical disease (Kyougoku 2010). In fact, workers without

medical disease have presented with occupational dysfunction (Akiyama 2010; Miyake

2014). Therefore, occupational dysfunction has the possibility to antedate the

appearance of depression in workers. We hypothesize that occupational dysfunction as

assessed by the Classification and Assessment of Occupational Dysfunction (CAOD) is

associated with the Center for Epidemiological Studies Depression Scale (CES-D). The

hypothesized model is shown in Figure 1. In other words, our model posits that the

occurrence of occupational dysfunction in healthcare workers facilitates depression.

Moreover, we surmise that occupational dysfunction in healthcare workers influences

depression-related factors, including opportunities for opportunities for refreshing

changes and methods of spending leisure time, and work relationships. In addition, we

postulate that occupational dysfunction in healthcare workers is influenced partly by

personal factors (such as age, years of work experience, and job category). The

significance of this study will be providing insights into the causes of depression.

In summary, this study aims to test the hypothesis that depression is

influenced by occupational dysfunction in healthcare workers.

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

4

Methods

Ethics statement

The Ethics Committee of Kibi International University approved the research

protocol (Nos. 13−30). We provided the participants with a letter explaining the outline

and purposes of the study and obtained their informed consent. Participants had the

right to drop out of the study without reason. We regarded return of the survey form as

consent for participation in this study. Survey forms were sent back anonymously in

sealed envelopes.

Participants

There were a total of 911 participants (463 nurses, 239 physical therapists,

and 209 occupational therapists).

Measures

Participant Profiles: Demographic data were obtained from participants. We assessed

gender, age, years of work experience, job category, taking a leave of absence, vacation,

work relationships, marital status, work schedule, drinking, and smoking.

CAOD (Miyake 2014): CAOD was measured using occupational dysfunction,

including occupational marginalization (6 items), occupational imbalance (4 items),

occupational alienation (3 items), and occupational deprivation (3 items), based on

OBP2.0. CAOD contains 16 items on a 7-point Likert scale (1 = strongly disagree, 7 =

strongly agree). CAOD has been widely used as an assessment tool for occupational

dysfunction.

CES-D (Shima 1985): CES-D was measured based on the level of depression

experienced within the past week using 20 items on four subscales - depressed affect (7

items), negative affect (4 items), interpersonal difficulties (2 items), and somatic

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

5

symptoms (7 items) - with a 4-point response (0 = none of the time, 3 = all of the

time). In epidemiologic studies, CES-D has been used worldwide as an assessment tool

for depression. Among the negative affect-related items, 4 were originally regarded as

related to a positive affect. In the present study, the 4 items were inversely scored to

make this point more comprehensible.

Statistical Analysis

SPSS Statistics (http://www.spss.com) was used for the sample characteristics.

Mplus 7.3 (http://www.statmodel.com) was used for structural equation modeling

(SEM). SEM is a comprehensive statistical analysis for the integration of path analysis

and factor analysis (Ullman & Bentler 2003). SEM offers the advantage of

identification of causal relationships (Ullman & Bentler 2003).

Sample Characteristics: The demographics of participants were summarized using

descriptive analyses. The normal distribution of all scores was examined using the

Kolmogorov-Smirnov test (p>0.05).

Structural validity: The factor structure of CAOD and CES-D were determined by

confirmatory factor analysis (CFA) of SEM, using a robust weighted least squares

factoring method (WLSMV) with missing data (Asparouhov & Muthén 2010). If an

unacceptable model fit was found due to CAF, we performed EFA using WLSMV with

missing data. EFA adopted the same model-data fit assessment as CFA. Based on the

results of EFA, once again, we performed a CFA using WLSMV with missing data. We

used three indexes for assessment of model-data fit (Kline 1986; Tabachnick & Fidell

2007). The first index was the root mean square error of approximation (RMSEA).

Critical values of RMSEA from 0.08 to 0.10 showed a mediocre fit, and below 0.08

indicated a good fit (MacCallum et al. 1996). The second and third indexes were the

comparative fit index (CFI), and the Tucker–Lewis index (TLI), both with critical

values above 0.95 (Kline 2011).

Testing the causal relationship: We analyzed the relationship using SEM. The

analyses examining the effects of occupational dysfunction on depression were

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

6

performed for causal structures as indicated in Figure 1. We assessed the model fit of

the hypothesized relationship between latent variables (occupational dysfunction,

depression) to data by SEM. To account for the contexts, personal variables (such as

age and gender) and depression-related variables (such as opportunities for refreshing

changes and methods of spending leisure time, work relationships) were included in the

model. The indirect effect estimates were calculated to test whether or not occupational

dysfunction was indirectly associated with subscales of CES-D (depressed affect,

negative affect, interpersonal difficulties, and somatic symptoms) via depression. When

modifying the model based on the results of SEM, the modification indexes, model fit,

standardized estimates, 95% confidence intervals, and standard errors were considered

while taking the hypothesis into account. SEM was performed without special

limitations. All items were treated as categorical data. An estimator was used for

WLMSV with missing data. Goodness of model fit index was evaluated using CFI, TLI,

and RMSEA. Critical values of RMSEA from 0.08 to 0.10 showed a mediocre fit, and

below 0.08 indicated a good fit (MacCallum et al. 1996). The second and third indexes

were the comparative fit index (CFI), and the Tucker–Lewis index (TLI), both with

critical values above 0.95 (Kline 2011).

Results

Sample Characteristics

Table 1 shows the results of sample characteristics. In this study, a total of

674 participants (388 nurses (including 11 public health nurses and midwives and 63

assistant nurses), 155 physical therapists, 123 occupational therapists, and 8 unknowns)

participated (73% response rate), including 159 males, 509 females, and 6 unknowns,

with an average age of 33.6 ± 10.2 years. The Kolmogorov-Smirnov test showed that all

scores had normal distribution.

Structural validity of CAOD

Figure 2 shows the results of CFA of CAOD. First, CFA was estimated to be a poor

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

7

estimate of RMSEA of model fit (RMSEA = 0.104, CFI = 0.943, and TLI = 0.931).

Therefore, we performed EFA, we performed EFA, and found that the CAOD consisted

of 16 items of 5 factors, including occupational marginalization of shared environment

(2 items), occupational marginalization of unshared environment (4 items), occupational

imbalance (4 items), occupational alienation (3 items), and occupational deprivation (3

items). The indexes for this model were RMSEA = 0.066, CFI = 0.988, and TLI = 0.972.

Based on EFA, CFA of CAOD was determined a good estimate of model fit (RMSEA =

0.092, CFI = 0.958, and TLI = 0.946).

Structural validity of CES-D

Figure 3 shows the results of CFA of CES-D. The CFA model for the latent

factors of CES-D exhibited good fit of depressed affect, negative affect, interpersonal

difficulties, and somatic symptoms (RMSEA = 0.060, CFI = 0.950, and TLI = 0.942).

Hypothesized model

Figure 4 shows the results of the final model. The hypothesized model

exhibited excellent fit (RMSEA=0.059, CFI=0.926, TLI=0.920). In this model,

occupational dysfunction was significantly positively associated with depression

(standardized direct effect = 0.745, S.E. = 0.021, Est./S.E. = 34.634, P-Value = 0.000,

95% CI = 0.703; 0.787). Indirect effect of occupational dysfunction on depression

symptoms was significant for the following indicators: depressed affect (standardized

indirect effect = 0.715, S.E. = 0.024, Est./S.E. = 29.618, P-Value = 0.000, 95% CI =

0.667; 0.762), negative affect (standardized indirect effect = 0.404, S.E. = 0.033,

Est./S.E. = 12.070, P-Value = 0.000, 95% CI = 0.526; 0.645), interpersonal difficulties

(standardized indirect effect = 0.585, S.E. = 0.030, Est./S.E. = 19.301, P-Value = 0.000,

95% CI = 0.526; 0.645), and somatic symptoms (standardized indirect effect = 0.681,

S.E. = 0.024, Est./S.E. = 28.175, P-Value = 0.000, 95% CI = 0.634; 0.729). In other

words, the symptoms of depression increased with the deterioration of occupational

dysfunction. In addition, the effects of occupational dysfunction on depression-related

factors were significant for the following indicators: opportunities for refreshing

changes (standardized direct effect = 0.552, S.E. = 0.030, Est./S.E. = 18.515, P-Value =

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

8

0.000, 95% CI = 0.494; 0.611), methods of spending leisure time (standardized direct

effect = 0.597, S.E. = 0.027, Est./S.E. = 22.112, P-Value = 0.000, 95% CI = 0.544;

0.650), and work relationships (standardized direct effect = 0.526, S.E. = 0.031, Est./S.E.

= 16.740, P-Value = 0.000, 95% CI = 0.464; 0.588). However, the effects of personal

factors (gender, age, years of work experience, job category, marital status, work

schedule, drinking, and smoking) on occupational dysfunction were not found to be

statistically significant. Therefore, these factors were deleted from the final model (see

Figure 4).

Discussion

We tested a hypothesized causal relationship model for predicting depression

in healthcare workers (Ishii 2010; Kyougoku 2010). This study is the first to test a

hypothesized model related to the relationship between occupational dysfunction and

depression in healthcare workers. Our statistical results provide evidence of the

applicability of a conceptual model of the effect of occupational dysfunction on

depression. Our findings might well warrant further study of the hypothesized model

about the effect of occupational dysfunction on depression in healthcare workers.

Previous studies have been limited to empirical data supporting the validity of this

hypothesized model.

Our findings demonstrated that occupational dysfunction factors are

significant contributors to depression. In Figure 4, our results show that occupational

dysfunction caused a direct effect on depression. Moreover, occupational dysfunction

caused an indirect effect on depressed affect, negative affect, interpersonal difficulties,

and somatic symptoms. In other words, healthcare workers with occupational

dysfunction have experienced depression, including depressive mood, negative

emotions, deterioration of interpersonal relationships, and a variety of physical

symptoms. Moreover, occupational dysfunction appears to be a significant contributing

condition to taking a leave of absence, vacation, and work relationships, as shown in

Figure 4. In other words, when there is occupational dysfunction, healthcare workers

experience deterioration in work relationships, rest, and diversion. These finding are

consistent with those of other studies (Akiyama 2010; Ishii 2010; Kyougoku 2010;

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

9

Miyake 2014).

Moreover, the results of CFA indicated a good factor structure for CAOD and CES-D.

In Figure 2, the results of CFA of CAOD indicated good model fit on the five-factor

structure, unlike a previous study (Miyake 2014). In Figure 2, major factors of

occupational dysfunction contributing to the development of depression were

occupational marginalization of shared environment, occupational marginalization of

unshared environment, and occupational deprivation. These kinds of occupational

dysfunction might be matters of special importance in the prevention of depression in

healthcare workers. In Figure 3, CFA also showed that the factor structure of CES-D

was appropriate. Thus, the factor structure of the CES-D was similar to those shown in

previous studies (Shima 1985).

Therefore, these results suggest it might be possible to prevent the occurrence

of depression by early detection and rapid treatment of occupational dysfunction. For

example, CAOD can be applied to healthcare workers to identify occupational

dysfunction. After completing CAOD, healthcare workers and occupational therapists

can discuss the responses to gain a better understanding of the occupational dysfunction.

When there is occupational dysfunction, occupational therapists can provide healthcare

workers with counseling and advice (Law et al. 1998; Law et al. 2002; Scaffa et al.

2010). Consequently, they may be able to treat the depression to a certain extent.

The first limitation of this study is an inferred causal effect with low power by

a cross-sectional design. Future research is necessary to confirm this relationship by a

longitudinal design. A second limitation is that the conclusions are limited to some

extent, as the participants were chosen from among nurses, physical therapists, and

occupational therapists, owing to some constraints. Future research with different

subjects, such as physicians, psychologists, and pharmacists, is necessary.

Conclusions

We conducted a comprehensive examination on occupational dysfunction and

depression in healthcare workers. The statistical results supported the hypothesis that

occupational dysfunction influences depression. In addition, the results of this study

suggest that occupational dysfunction influences taking a leave of absence, vacation,

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

10

and work relationships. The findings of this study have beneficial value for industrial

health.

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

11

Figures

Figure 1. Hypothesized model

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

12

Note: CFI = 0.958; TLI = 0.946; RMSEA = 0.092

Figure 2. CFA of CAOD (5 factors)

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

13

Note: CFI = 0.950; TLI = 0.942; RMSEA = 0.060

Figure 3. CFA of CES-D

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

14

Note: CFI = 0.926; TLI = 0.920; RMSEA = 0.059

Figure 4. Causal model of CES-D on CAOD

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

15

Tables

Table 1. Sample characteristics

M ± SD

Age

Total 33.6±10.2

Nurses 37.4±11.2

Physical therapists 29.3±6.4

Occupational therapists 27.6±4.1

Years of work

experience

Total 9.67±9.2

Nurses 12.8±10.3

Physical therapists 5.64±5.3

Occupational therapists 4.84±3.5

Work hours

(hours) Total 9.15±3.3

Commute time

(minutes) Total 26.4±18.8

Total N %

Gender

Male 159 23.5

Female 509 75.4

Unknown 7 1.0

Job category

Nurse 326 48.2

Health nurse, Midwife 12 1.7

Assistant nurse 63 9.3

Physical therapist 155 22.9

Occupational therapist 123 18.2

Unknown 8 1.1

Drinking

Total 324 48

Nurses 166 42.5

Physical therapists 85 54.8

Occupational therapists 71 57.7

Smoking Total 122 18.1

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

16

Nurses 65 16.6

Physical therapists 32 20.6

Occupational therapists 23 18.6

Taking a leave

of absence

Very good 71 10.5

Good 364 54

Neither good nor bad 91 13.5

Fair 56 8.3

Poor 62 9.1

Unknown 30 4.4

Vacation

Very good 53 7.8

Good 285 42.2

Neither good nor bad 141 20.9

Fair 123 18.2

Poor 43 6.3

Unknown 29 4.3

Work

relationships

Very good 85 12.6

Good 356 52.8

Neither good nor bad 162 24

Fair 29 4.3

Poor 13 1.9

Unknown 29 4.3

CAOD M±SD Skewness Peakedness

Total 52.4±17.8 0.21 -0.231

Nurses 55.7±18.4 0.03 -0.139

Physical

therapists

46.6±16.5 0.45 0.244

Occupational

therapists

48.6±15.9 0.16 -0.56

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

17

References Akiyama E, Kyougoku, M. 2010. Examination of occupational dysfunction in workers:

using occupational self assessment version2 (OSA2). Sogo Rehabilitation

38:373-379.

Anaby D, Jarus T, Backman CL, and Zumbo BD. 2010. The role of occupational

characteristics and occupational imbalance in explaining well-being. Applied

Research in Quality of Life 5:81-104.

Asparouhov T, and Muthén B. 2010. Weighted least squares estimation with missing

data. MplusTechnical Appendix.

Bryant W, Craik C, and McKay EA. 2004. Living in a glasshouse: exploring

occupational alienation. Canadian Journal of Occupational Therapy

71:282-289.

Denton M, Zeytinoglu IU, Davies S, and Lian J. 2002. Job stress and job dissatisfaction

of home care workers in the context of health care restructuring. International

Journal of Health Services 32:327-357.

Dyrbye LN, Thomas MR, Massie FS, Power DV, Eacker A, Harper W, Durning S,

Moutier C, Szydlo DW, and Novotny PJ. 2008. Burnout and suicidal ideation

among US medical students. Annals of internal medicine 149:334-341.

Harry E. 2013. Stress and the healthcare worker. As complicated or as simple as fear

and hope. The Journal of medical practice management: MPM 30:28-30.

Health NIfOSa. 2008. Exposure to stress: occupational hazards in hospitals. Cincinnati:

National Institute for Occupational Safety and Health.

Honda A, Date Y, Abe Y, Aoyagi K, and Honda S. 2014. Work-related Stress,

Caregiver Role, and Depressive Symptoms among Japanese Workers. Saf

Health Work 5:7-12.

Irvine D. 1997. The performance of doctors. I: Professionalism and self regulation in a

changing world. BMJ: British Medical Journal 314:1540.

Ishii Y, Kyougoku, M., Nagao S. 2010. Occupational therapy of mental health. Tokyo:

Chuohoki.

Kato R, Haruyama Y, Endo M, Tsutsumi A, and Muto T. 2014. Heavy overtime work

and depressive disorder among male workers. Occupational Medicine:kqu120.

Kline P. 1986. A handbook of test construction: Introduction to psychometric design:

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

18

Methuen.

Kline RB. 2011. Principles and practice of structural equation modeling: Guilford

press.

Kyougoku M. 2010. Unstructured assessment for occupational therapy, 4 condition

method. Tokyo: Seisin-shoboo.

Law M, Steinwender S, and Leclair L. 1998. Occupation, health and well-being.

Canadian Journal of Occupational Therapy 65:81-91.

Law MC, Baum CM, and Baptiste S. 2002. Occupation-based practice: Fostering

performance and participation: Slack Incorporated.

MacCallum RC, Browne MW, and Sugawara HM. 1996. Power analysis and

determination of sample size for covariance structure modeling. Psychological

methods 1:130.

Matsudaira K, Shimazu A, Fujii T, Kubota K, Sawada T, Kikuchi N, and Takahashi M.

2013. Workaholism as a Risk Factor for Depressive Mood, Disabling Back Pain,

and Sickness Absence. PLoS One 8:e75140.

Maynard M. 1986. Health promotion through employee assistance programs: a role for

occupational therapists. The American journal of occupational therapy: official

publication of the American Occupational Therapy Association 40:771-776.

Miyake Y, Teraoka, M., Ogino, K., Kyougoku, M. 2014. Survey of the classification of

occupational dysfunction among japanese rehabilitation therapists and the

association of occupational dysfunction with job strain. Journal of Preventive

Medicine 9:93-100.

Scaffa ME, and Reitz SM. 2013. Occupational Therapy Community-Based Practice

Settings: FA Davis.

Scaffa ME, Reitz SM, and Pizzi M. 2010. Occupational therapy in the promotion of

health and wellness: FA Davis Company Philadelphia, PA.

Schaefer JA, and Moos RH. 1993. Relationship, task and system stressors in the health

care workplace. Journal of Community & Applied Social Psychology 3:285-298.

Seki Y, and Yamazaki Y. 2006. Effects of working conditions on intravenous

medication errors in a Japanese hospital. Journal of nursing management

14:128-139.

Shima S, Shikano, T, Kitamura, T., Asai, M. 1985. New self-rating scales for

depression. Clinical Psychiatry 27:717-723.

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

19

Tabachnick BG, and Fidell LS. 2007. Using multivariate statistics, 5th ed. Toront:

Allyn and Bacon.

Takeuchi T, and Nakao M. 2013. The relationship between suicidal ideation and

symptoms of depression in Japanese workers: a cross-sectional study. BMJ

Open 3:e003643.

Tamakoshi A, Ohno Y, Yamada T, Aoki K, Hamajima N, Wada M, Kawamura T,

Wakai K, and Lin YS. 2000. Depressive mood and suicide among middle-aged

workers: findings from a prospective cohort study in Nagoya, Japan. Journal of

epidemiology/Japan Epidemiological Association 10:173-178.

Teraoka M, Kyougoku, M. 2014. Development of occupation-based practice 2.0 that

integrates occupation based-practice and dissolution approach for belief conflict.

Japanese occupationai therapy researh 33:249-258.

Townsend E, and Wilcock AA. 2004. Occupational justice and client-centred practice: a

dialogue in progress. Canadian Journal of Occupational Therapy 71:75-87.

Ullman JB, and Bentler PM. 2003. Structural equation modeling: Wiley Online Library.

Van Praag HM. 2004. Can stress cause depression? Progress in

Neuro-Psychopharmacology and Biological Psychiatry 28:891-907.

Whiteford G. 2000. Occupational deprivation: Global challenge in the new millennium.

The British Journal of Occupational Therapy 63:200-204.

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.787v1 | CC-BY 4.0 Open Access | rec: 13 Jan 2015, publ: 13 Jan 2015

PrePrin

ts