Tsutsumi, Akizumi MD; Nagami, Makiko MA; Yoshikawa, Toru MD; Kogi, Kazutaka MD; Kawakami, Norito MD
* Review the theoretical background and characteristics of participatory approaches to workplace intervention, including evidence to date on their use.
* Outline the process of the participatory intervention evaluated in this study and the types of changes that were designed and implemented as a result.
* Discuss the intervention’s effects on mental health and job performance outcomes, along with factors that promote and hinder successful intervention.
Observational studies suggest that improving the psychosocial work environment may improve mental health,1 and that mental health in the workplace is closely related to productivity.2 Based on the relevant literature, studies on the effectiveness of improving the work environment are called for.3–9
Recent trends in workplace intervention include comprehensive and integrative approaches.10 An important element in these interventions is the participatory approach, which implies control and empowerment for those involved.11 Employees in the workplace well understand both problems and their solutions.12–14 A participatory approach is likely to ensure an appropriate risk assessment, which is an important prerequisite for a focused intervention.13,15 Participation is also likely to improve the chances that changes will be accepted as meaningful, and so be integrated into the organizational culture.16 Similar approaches have also been successfully introduced into the workplace to improve productivity.17,18
Several previous studies have outlined the theoretical background of the participatory approach to mental health issues. The most fruitful evidence has been derived from studies concerned with musculoskeletal disorders including low back pain as the outcome.19–23 Findings on participatory ergonomic interventions on musculoskeletal disorders revealed that greater participation in the process was associated with increased levels of job control and communication and consequently reduced musculoskeletal symptoms and time loss at work.24 Other than worker “control,” participation by those directly involved is likely to increase a “sense of democracy or fairness in the workplace,” as well as “support,” all of which are relevant psychosocial job conditions,25,26 and improvements in such psychosocial job conditions are expected to improve workers’ mental health1 and productivity.27 Recently, a controlled trial indicated that a worker participatory organizational intervention may have beneficial effects on psychosocial job conditions and psychological distress among white-collar women.28
Although participatory workplace improvement appears to be a promising measure not only for workers’ good mental health but also for organizational profit, its effectiveness in a real workplace setting has not been investigated in a randomized controlled trial. We conducted such a trial to explore the effects of participatory intervention for workplace improvement on mental health and job performance among blue-collar workers manufacturing electronic equipment.
The study was a cluster randomized controlled trial in which the effect of participatory intervention for workplace improvement on workers’ mental health and productivity was evaluated. The intervention was applied at the cluster level because of the nature of team-based intervention, and clustering was on the assembly line level. Eleven assembly lines producing electrical devices were randomly allocated to intervention and control lines. The workers in the intervention lines identified and prioritized their specific needs and developed action plans to improve their work environments. Self-report measures were used at pre- and post-intervention, and the same survey was administered at both time points. Workers in the control groups completed both questionnaires, but no organized activities were provided. We did not obtain written informed consent from the workers; however, all workers were told about the study procedure and were clearly informed that completion of the questionnaire was entirely voluntary, and all responses were confidential. The responses to the questionnaire surveys were considered to be an agreement to participate in the study. The study design and procedure were reviewed and approved by the Research Ethics Committee of the Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences.
The study was conducted in a medium-sized company producing electrical devices. Because of the recent highly competitive trend in their marketplace, the administration took several steps to improve business: ie, an annual salary payment plan (April 1999), a voluntary retirement plan to enforce staff reduction (between October and December 1999), streamlining the production line, transferring some of the engineering staff from head office to a plant site (December 1999), and the sale of a factory (March 2000). As a result, the business appeared to improve in 2000. However, the administration was concerned about the increase in employee workload due to the anticipated increase in production in the forthcoming years. The administration also recognized a recent decrease in employees’ morale and expected that stress reduction could lead to a rise in performance.
Several meetings were held with administrators and human relations personnel at the company to discuss the needs for work environment improvements. During these sessions, the researchers sought to ensure that the president and factory manager were provided with the rationale behind the participatory activities. The feasibility of the study protocol, including the study procedure (worker-participatory style) and the production of units for comparison by randomization of departments, was discussed. It was decided that the trial should be conducted among blue-collar workers employed in the company factory lines (n = 139). There were 14 assembly lines producing electrical devices in the factory, involved in machinery and personal computer production, in addition to the quality control division. Three lines involved in machinery (n = 42) were excluded from the study because of the planned re-streamlining of these lines. As a result, a total of 97 workers on eleven assembly lines were chosen as the study population.
Randomization and Blinding
Randomization was performed using the SPSS random sampling procedure by a researcher who was provided with a list of the eleven lines. The list included only the nicknames of the lines to ensure that the researcher was blind to the identity of the lines selected. The lines were allocated to six intervention (n = 47) and five control (n = 50) lines. All workers within the study population were included, and strategies were implemented to minimize the possibility of selection bias over the process of allocation. However, because of the nature of the study, neither the workers nor the researchers were blind to their study line.
Two primary outcomes were adopted—mental health and productivity. They were measured using the self-administered questionnaires before (July 2005) and at the end of the intervention (August 2006). As a mental health outcome, minor psychiatric morbidity was assessed using the General Health Questionnaire (GHQ) 28-item version.29 The GHQ is widely used in psychiatric epidemiology as a measure of emotional distress in community samples. The questionnaire referred to the respondent’s experience over the past week. The GHQ includes items related to anxiety, sleep disturbance, depression, and cognitive function. It was coded in a conventional fashion with scores potentially ranging from 0 to 28. Internal consistency reliability coefficients were obtained for the company employees (n = 944). In our study, Cronbach alpha coefficients were 0.93 and 0.94, at pre- and post-intervention surveys, respectively.
To assess the behavioral outcome, a self-reported job performance checklist was given to individual workers. The checklist was taken from the WHO Health and Work Performance Questionnaire (HPQ) and included the following: quantitative and qualitative efficiency during the last 30 days (10 items), self-evaluated job outcomes relative to those of other workers (1 item), and special work success or failure during the last 30 days (2 items).30 Although a categorical rating is employed in the instrument, a single index of job performance was derived by summing the individual ratings for this study.31 In our study, Cronbach alpha coefficients were 0.82 and 0.80, at pre- and post-intervention surveys, respectively.
Stress Surveillance for Hazard Identification.
Exposure to psychosocial job conditions was measured before the intervention as a prior hazard identification and at the end of the intervention to evaluate implementation of the intervention. For this, we employed the Japanese version of the Job Content Questionnaire (JCQ).32 The Japanese version of the JCQ is based on Karasek’s demand-control model.33 This questionnaire has the following subscales: job demands regarding quantitative and qualitative workloads (5 items), job control regarding decision-making authority and skill discretion (9 items), and social support from supervisor and coworkers (4 items each). Each item was scored based on a four-point response ranging from 1 (strongly disagree) to 4 (strongly agree). The alpha coefficients estimated at the pre- and post-intervention surveys were 0.67 to 0.71 for job demands, 0.80 to 0.77 for job control, 0.91 to 0.91 for supervisory support, and 0.79 to 0.80 for coworker support, respectively. The sum of the weighted item scores was used as a scale score.34 The scale scores from the JCQ were manipulated so that participants could compare their stress levels with Japanese national summary data.35
Relevant sociodemographic variables were determined at the pre-intervention survey via a standardized questionnaire that included information on gender, age, career years, years of education, and employment status (general worker or supervisor).
Outline of the Intervention
The team-based, problem-solving intervention used in the current study was based on active employee involvement, shared work-related goals, and action planning to improve the work environment for stress reduction. Consistent with the concept of healthy work organization, the intervention focused on problems in existing practice, was process oriented, and was carried out in a joint cooperative effort between employees and researchers.36 The intervention also focused on environment improvement or job redesign rather than behavioral change in symptoms of ill health.26,37 A participatory approach has several principles which include: 1) to build on local practice (starting from real problems of the enterprise instead of the priorities of outsiders, problems that need improvement exist at the site where employees work, so researchers’ knowledge has several limitations), 2) to focus on achievements (good practices already available), 3) to link working conditions with other management goals (ie, productivity), 4) to encourage exchange of experience, 5) to promote employee involvement, and 6) to use learning-by-doing.38 The details of the outline of our intervention are described below (Fig. 1).
We trained key persons who could take the initiative and facilitate participatory activities at their workplaces. We held a half-day training workshop for the facilitators in May 2005. The workshop participants, including one from the study company, were provided with comprehensive information on mental health in the workplace and listened to a lecture on the participatory approach for stress reduction. They also experienced the employee’s role via a workshop based on real cases using the results of stress surveillance and the checklist for risk assessment (see set-up workshop). Several key points for successful facilitating were stressed.39 These included; 1) organization skills, ie, planning effective workshops and recruiting appropriate participants, 2) technical knowledge, ie, understanding and applying stress surveillance and the basic rules of the technical areas to be improved, 3) group work skills, ie, guiding and facilitating active group discussions, 4) presentation skills, ie, speaking in a clear and concise manner using audio visual aids effectively, and 5) effective follow-up and evaluation activities to sustain local improvement actions.
Human resource personnel and the factory section chief undertook roles as facilitators. In October 2005, facilitators and researchers provided a supervisory education program for stress reduction in which the supervisors were informed of the significance of positive mental health, improvement in the work environment and given examples of good practices. We also tried to ensure that the participants were equipped with the knowledge and skills needed to identify occupational stressors. We then helped the supervisors design initiatives which could clarify ambiguity, reduce conflict and overcome the sources of stress.
To initiate activity for improvement, the set-up workshop was held by the workers in the intervention group in November 2005. At the beginning of the workshop, the factory manager declared that the factory would deal with mental health as a primary concern and participatory workplace improvements for stress reduction with the help of external experts (the researchers). The results of pre-intervention stress surveillance was visualized to compare national data and then used as a reference for hazard identification and an index of improvement.32 In addition, a checklist was introduced for organizing workplace-level discussion to identify immediate, low-cost improvements in the workplace. The checklist was developed through a review of related references and the collection of examples of workplace improvements to produce a concrete action plan.40 Using these tools, workers listed and prioritized issues to be improved in their workplace based on the results of their stress surveillance. The facilitators led group discussions to assist the workers in listing activities for workplace improvements. After the workshop, each group proposed improvements (action plans) at their workplaces, and considered the effectiveness, feasibility, priority, and cost of those improvements.
Implementation of Work Environment Improvements.
After the set-up workshop, workers started to improve their work conditions based on the action plans proposed at the set-up workshop. The role of the facilitators was to support and sustain employee autonomous activities. In April and August 2006, workplace observation and follow-up workshops were held by the study group including the researchers. The activity processes were presented by the line foremen and problems of the implemented improvement and barriers to the activities were discussed. The researchers gave necessary suggestions for further improvement and encouraged workers to sustain the autonomous activities for workplace improvement.41
Evaluation included both outcome and process evaluations. The post-intervention questionnaire survey was administered for the primary outcomes and psychosocial job conditions. We inferred whether the intervention was implemented from comparing the changes of psychosocial job conditions in the intervention lines with those in the control lines. To further evaluate implementation of the intervention in the intervention lines, we conducted post-intervention interviews with the factory manager and facilitators.
All participants were assigned an identification number and were treated anonymously in all analyses. We conducted the analyses according to the intention-to-treat principle. As the point of inference in our study is on the individual workers level, we present outcomes of analyses on the individual workers level. Continuous variables are listed with means (SD) and categorical variables as frequencies and percentages. The mean of each measure was compared using t-tests. The differences in the categorical variables were assessed using the χ2 test. To assess the interventional effects on minor psychiatric morbidity and job performance, we used multilevel modeling to take account of the fact that individual workers are nested in units comprised of assembly lines. Linear mixed models were used. Gender, age, career years, years of education, and employment status were adjusted for in the model. The statistical significance for the interactive effects between groups and time was assessed. To test for simple main effects, paired t-tests for the intervention and control groups were computed. To assess whether the lost-to-follow-up data have caused bias, we conducted additional analyses. Imputation of missing values was performed with the mean for the entire series of continuous variables and the modal category for the small number of categorical covariates. Values of P < 0.05 (two-tailed) were considered statistically significant. Statistical analyses were undertaken using SPSS 15 for Windows.
A total of 97 workers were eligible for the study because they were employed in the target assembly lines at baseline. There were 27 (57%) and 18 (36%) female workers in the intervention and control groups, respectively; the mean age was 48 years and 44 years in the intervention and control groups, respectively; the mean number of career years was 19 and 16 in the intervention and control groups, respectively; the mean years of education was 12 and 14 in the intervention and control groups, respectively; there were 42 (89%) and 42 (84%) general workers in the intervention and control groups, respectively; and there were 5 (11%) and 8 (16%) supervisors in the intervention and control groups, respectively. With the exception of gender (χ2 = 4.48, df = 1, P = 0.034), relevant socio-demographic data were evenly distributed between the groups. After the exclusion of workers who were transferred to other lines or retired during the intervention period, 35 workers on the intervention lines and 42 workers on the control lines were analyzed. Table 1 shows the baseline characteristics of intervention and control groups. Gender, employment status, and educational attainment were evenly distributed between the groups, although the intervention group included more veteran workers than the control group. For the mental health analyses, we excluded 9 workers (7 in the intervention group and 2 in the control group) for whom information on mental health from the survey responses was incomplete. For the job performance analyses, we excluded 7 workers (4 from the intervention group and 3 from the control group) for whom information on job performance from the survey responses was incomplete. Thus, the final sample size for the analysis of mental health was 68 workers (28 in the intervention group and 40 in the control group), and the final sample size for the analysis of job performance was 70 workers (31 in the intervention group and 39 in the control group) (Fig. 2).
Table 2 shows a summary of the improvements implemented during the intervention period. Across three technical areas—securing safety, ergonomic improvement and communication, and performance improvements—several improvements were planned and implemented. Some improvements were carried out as planned at the set-up workshop, whereas others were carried out with appropriate corrections during the intervention period.
Table 3 shows the changes in mean scores for the GHQ and HPQ in the intervention and control groups. GHQ scores remained the same in the intervention group, whereas the GHQ scores deteriorated (increased) in the control group; the change was statistically significant (t = −2.43, P = 0.020). The interaction effect was not statistically significant. HPQ scores increased in the intervention group (t = −0.95, P = 0.348), but decreased in the control group (t = 2.13, P = 0.040). The interaction effect was statistically significant.
Table 4 shows the results of the additional analyses. During the follow-up period, GHQ scores improved (decreased) in the intervention group, but deteriorated (increased) in the control group; the paired t test revealed that both changes were statistically significant (t = −2.37, P = 0.022 and t = 2.18, P = 0.035, respectively). The interaction effect was statistically significant. HPQ scores increased in the intervention group, whereas decreased in the control group. However, either the changes or the interaction term was not statistically significant.
With the exception of two lines (b and c), scores on the psychosocial job conditions scale generally improved in the intervention group, in particular, increased job control at lines a and f, and improved supervisor or coworker support at lines e and f were noted. Although there was a line in which the psychosocial job conditions scores slightly improved in the control group (line i), the scores worsened in the remaining lines (g, h, j, and k) (Table 5).
Post-intervention interviews with the factory manager and facilitators revealed some of the reasons why psychosocial job conditions did not improve in the intervention lines. The workers in line b had relatively low self-esteem, and the human relationships among those on the line had not been very good, so an active commitment from these workers in group practices like the current participatory activities was not obtained. The researchers also considered from the workshop presentations that no progress was realized in this line in the two consecutive reports during the intervention. Line c produced many small pieces and because of the small profits, quick returns were always requested. Furthermore, the majority of veteran workers, who were near retirement, appeared to resist change. Although they had the opportunity of part-time employment after retirement, their salaries were expected to decrease. A few improvements were attempted, such as clear labeling and storage for larger numbers of products, but the line leader admitted to the poor commitment of the workers, and thus improvements were not effective.
Nevertheless, at the lines where improvements in psychosocial job conditions were observed, we found several reasons for these improvements. For example, at line a, where job control was improved, the workers dealt with a variety of small parts and the tasks had not been standardized. The work involved the workers thinking for themselves, and the small units they worked on were formulated according to the specific task. Participatory workplace improvement could have contributed to improvement in job conditions (job control). At line d, the workers proposed many improvement plans by mutual agreement. For example, dust was a problem in the workplace, so they decided to cleanup their workplace regularly with all members of the group carrying out the plan. At line f, the preparatory table was moved nearer the task table so that frequently-used materials and tools would be placed within easy reach, thus substantially shortening preparation time. Manuals for working procedures were revised by marking specific places, so that workers could refer to them more easily. Such redesigns appeared to contribute to improvements in coworker support and job performance. The line leader of line e produced several improvement plans such as labeling materials, inventory control, and gauge instrument improvement, which were then executed. Supervisor support was substantially improved in this line. Although this may not have been due strictly to the results of participatory workplace improvement, the intervention program appeared to provoke these activities.
To our knowledge, our study is the first report to demonstrate the significant beneficial effects of participatory workplace improvement using a randomized controlled design. The workers improved their work environment by hazard identification based on stress surveillance, risk evaluation including the application of an action checklist, group discussion on workplace improvements, and actual planning to realize these goals. Although employees’ self-administered questionnaires were used, the results indicated that participatory intervention for workplace environment improvements was effective in preventing deterioration of mental health. The findings also imply that an organizational benefit, ie, productivity, a valuable outcome for management personnel, can be brought about using this participatory approach.
Although the response rate for each survey was satisfactory, only half of the target population was analyzed in this study. The dropout analysis indicated that slightly more women than men, and older rather than younger workers were lost from the follow-up data, which might have exaggerated or reduced the associations according to the outcomes. However, individual transfer to other lines and retirement took place or had already been planned irrespective of the intervention. Furthermore, the overall changes in the outcomes were in the expected direction.
The intervention was feasible, and several visible improvements were achieved. Workplace improvements which aim to exclude obstacles are in accordance with the participants’ ordinary work activities. Workers in Japanese manufacturing industries, such as the population in this study, are familiar with Total Quality Management, an organization managerial approach centered on quality which is based on the participation of all members, and aimed at long-term success through customer satisfaction. They are also familiar with Kaizen—a process step which focuses on continuous process improvement to make processes visible, repeatable, and measurable. The present factory study was small, and therefore interpretation of the results may be limited to these samples. To maximize the generalizability of the findings, further studies are necessary among diverse work sites and a wider range of occupations including white-collar occupations.
The process evaluation provided further valuable insights. If properly implemented, intervention does seem to have beneficial effects. However, in cases where a poor function is identified, early detection of the barrier and the causative conditions are essential for successful intervention. At an early stage, constant observation and appropriate assistance from facilitators and researchers might be needed. In addition, the significant role of foremen was inferred; leadership is necessary for workplace improvements to be undertaken.
The approach we adopted was multi-faceted. Strictly speaking, it is difficult to know which parts of the improvements were really effective. It is necessary to know how, to what extent, and under which conditions it is possible to improve working conditions.42 As for outcome measurements, only self-reported indices were employed, which raises the issue of a possible response bias.43 Imprecision in measurements (ie, job demands scale) and imputation of missing data may increase random error and weaken apparent associations. Some of the original questionnaires which we developed will need further standardization and validation. In particular, the sensitivity of the job performance checklist needs to be evaluated in a Japanese setting. Objective outcomes such as biological markers are necessary to strengthen the values of the activities.
Information leakage was possible in this small factory. In addition, we cannot ignore the fact that a participatory approach brings with it a certain burden on the participants. Such effects would most likely lead to an underestimation in the results. The lack of longer follow-up data makes it impossible to reach any conclusions on the long-term effects of the activity. Future study should evaluate a more permanent effect and cost-effectiveness.
Before conclusion, we would like to make a few comments to improve this kind of research. The realization of the above-mentioned interventional research requires the understanding and cooperation of employers or administrators, because these measures depend largely on organizational redesigns. Researchers should ensure that administrators and other key stakeholders (ie, health workers, human resource personnel, managers and supervisors, and union delegates) have a clear understanding of their roles and responsibilities.44 Intervention should be theory based45 and the study protocol should be developed after taking into account the most suitable moment to apply the protocol (not disturbing the ordinal activities of the organization) and expected barriers for the implementation of the intervention.46 Training of facilitators could be tailor-maid for each workplace. They should encourage workers in active commitment, as the greater participation of workers, the more likely the successful implementation of the intervention.24,28
Despite the limitations, our findings suggest that it is worthwhile to facilitate participatory workplace improvements to ensure workers’ good mental health as well as organizational benefits.
This study was partly supported by Grand-in-Aid for Scientific Research (C), from the Japan Ministry of Education, Culture, Sports, Science and Technology (project number 16590476) and by Health and Labour Sciences Research Grants (Research on Occupational Safety and Health; Research number H20-Rodo-Ippan-006), from the Japan Ministry of Health, Labour and Welfare.
1.Stansfeld S, Candy B. Psychosocial work environment and mental health—a meta-analytic review. Scand J Work Environ Health. 2006;32:443–462.
2.Kalia M. Assessing the economic impact of stress—the modern day hidden epidemic. Metabolism. 2002;51:49–53.
3.Caulfield N, Chang D, Dollard MF, Elshaug C. A review of occupational stress interventions in Australia. Int J Stress Manag. 2004;11:149–166.
4.EASHW. How to Tackle Psychosocial Issues and Reduce Work-Related Stress. Luxembourg: European Agency for Safety and Health at Work (EASHW); 2002.
5.Giga S, Noblet A, Faragher B, Cooper C. Organisational stress management interventions: a review of UK-based research. Aust Psychol. 2003;38:158–164.
6.Kompier MAJ, Kristensen TS. Organizational work stress interventions in a theoretical, methodological and practical context. In: Dunham J, ed. Stress in the Workplace Past, Present and Future. London: Whurr Publishers; 2001:164–190.
7.Michie S, Williams S. Reducing work related psychological ill health and sickness absence: a systematic literature review. Occup Environ Med. 2003;60:3–9.
8.Semmer NK. Job stress interventions and the organization of work. Scand J Work Environ Health. 2006;32:515–527.
9.van der Klink JJL, Blonk RWB, Schene AH, van Dijk FJH. The benefits of interventions for work-related stress. Am J Public Health. 2001;91:270–276.
10.Noblet A, Lamontagne AD. The role of workplace health promotion in addressing job stress. Health Promot Int. 2006;21:346–353.
11.Aust B, Ducki A. Comprehensive health promotion interventions at the workplace: experiences with health circles in Germany. J Occup Health Psychol. 2004;9:258–270.
12.Heaney CA. Worksite health interventions: targets for change and strategies for attaining them. In: Quick JC, Tetrick LE, eds. Handbook of Occupational Health Psychology. Washington, DC: American Psychological Association; 2002:305–323.
13.Kompier MAJ, Geurts SAE, Grundemann RWM, Vink P, Smulders PGW. Cases in stress prevention: the success of a participative and stepwise approach. Stress Med. 1998;14:155–168.
14.Schurman SJ, Israel BA. Redesigning work systems to reduce stress: a participatory action research approach to creating change. In: Murphy LR, Hurrell JJ, Sauter SL, Keita GP, eds. Job Stress Interventions. Washington, DC: American Psychological Association; 1995:235–263.
15.Beehr TA, O’Driscoll MP. Organizationally targeted interventions aimed at reducing workplace stress. In: Thomas JC, Hersen M, eds. Handbook of Personality and Health in the Workplace. Thousand Oaks, CA: Sate; 2002:103–119.
16.Reicherts A, Wanous JP, Austin JT. Understanding and managing cynicism about organizational change. Acad Manage Exec. 1997;11:452–467.
17.Sirota D, Mischkind LA, Meltzer MI. The Enthusiastic Employee: How Companies Profit by Giving Workers What They Want. New Jersey: Wharton School Publishing; 2005.
18.Ulrich D, Kerr S, Ashkenas R. The GE Work-Out. New York: McGraw-Hill; 2002.
19.Hess JA, Hecker S, Weinstein M, Lunger M. A participatory ergonomics intervention to reduce risk factors for low-back disorders in concrete laborers. Appl Ergon. 2004;35:427–441.
20.Hignett S, Wilson JR, Morris W. Finding ergonomic solutions—participatory approaches. Occup Med (Lond). 2005;55:200–207.
21.Poosanthanasarn N, Lohachit C, Fungladda W, Sriboorapa S, Pulkate C. An ergonomics intervention program to prevent worker injuries in a metal autoparts factory. Southeast Asian J Trop Med Public Health. 2005;36:512–522.
22.Poosanthanasarn N, Sriboorapa S, Fungladda W, Lohachit C. Reduction of low back muscular discomfort through an applied ergonomics intervention program. Southeast Asian J Trop Med Public Health. 2005;36(Suppl 4):262–270.
23.Rosecrance JC, Cook TM. The use of participatory action research and ergonomics in the prevention of work-related musculoskeletal disorders in the newspaper industry. Appl Occup Environ Hyg. 2000;15:255–262.
24.Rivilis I, Cole DC, Frazer MB, Kerr MS, Wells RP, Ibrahim S. Evaluation of a participatory ergonomic intervention aimed at improving musculoskeletal health. Am J Ind Med. 2006;49:801–810.
25.Karasek RA. An analysis of 19 international case studies of stress prevention through work reorganization using the demand/control model. Bull Sci Technol. 2004;24:446–456.
26.Mikkelsen A, Saksvik PØ, Landsbergis P. The impact of a participatory organizational intervention on job stress in community health care institutions. Work Stress. 2000;14:156–170.
27.Baruch-Feldman C, Brondolo E, Ben-Dayan D, Schwartz J. Sources of social support and burnout, job satisfaction, and productivity. J Occup Health Psychol. 2002;7:84–93.
28.Kobayashi Y, Kaneyoshi A, Yokota A, Kawakami N. Effects of a worker participatory program for improving work environments on job stressors and mental health among workers: a controlled trial. J Occup Health. 2008;50:455–470.
29.Goldberg D, Williams P. A User Guide to the General Health Questionnaire. Windsor: NFER-Nelson; 1991.
30.Kessler RC, Barber C, Beck A, et al. The World Health Organization Health and Work Performance Questionnaire (HPQ). J Occup Environ Med. 2003;45:156–174.
31.Tsutsumi A, Takao S, Mineyama S, Nishiuchi K, Komatsu H, Kawakami N. Effects of a supervisory education for positive mental health in the workplace: a quasi-experimental study. J Occup Health. 2005;47:226–235.
32.Kawakami N, Kobayashi F, Araki S, Haratani T, Furui H. Assessment of job stress dimensions based on the job demands-control model of employees of telecommunication and electric power companies in Japan: reliability and validity of the Japanese version of the Job Content Questionnaire. Int J Behav Med. 1995;2:358–375.
33.Karasek R, Baker D, Marxer F, Ahlbom A, Theorell T. Job decision latitude, job demands, and cardiovascular disease: a prospective study of Swedish men. Am J Public Health. 1981;71:694–705.
34.Karasek RA, Gordon G, Pietroskovsky C, Frese M, Pieper C, Schwartz J. Job Content Instrument: Questionnaire and User’s Guide. Los Angels/Lowell, MA: University of Southern California/University of Massachusetts; 1985.
35.Kawakami N. Worksite mental health program plan and mental health care practice. In: Japan Industrial Safety and Health Association, ed. Worker Mental Health Program—Guideline and Commentary. Tokyo: Japan Industrial Safety and Health Association; 2001:55–88.
36.Peters M, Robinson V. The origin and status of action research. J Behav Sci. 1984;20:113–124.
37.Karasek R, Theorell T. Healthy Work: Stress, Productivity, and the Reconstruction of Working Life. New York: Basic Books; 1990.
38.Kogi K. Training in practical ergonomics improvements. J Hum Ergol. 1989;18:147–150.
39.Kogi K. Participatory action training for workplace improvements in small enterprises in developing countries. Environ Sci. 1996;4:123–133.
40.Yoshikawa T, Kawakami N, Kogi K, et al. Development of a mental health action checklist for improving workplace environment as means of job stress prevention. Sangyo Eiseigaku Zasshi. 2007;49:127–142.
41.Itani T, Tachi N, Takeyama H, et al. Approaches to occupational health based on participatory methodology in small workplaces. Ind Health. 2006;441:17–21.
42.Kristensen TS. Intervention studies in occupational epidemiology. Occup Environ Med. 2005;62:205–210.
43.Crowne DP, Marlowe D. A new scale of social desirability independent of psychopathology. J Consult Psychol. 1960;24:349–354.
44.World Health Organization. Mental Health Policies and Programmes in the Workplace. Geneva: World Health Organization; 2005.
45.Goldenhar L, LaMontagne A, Katz T, Heaney C, Landsbergis P. The intervention research process in occupational safety and health: an overview from the National Occupational Research Agenda Intervention Effectiveness Research team. J Occup Environ Med. 2001;43:616–622.
46.van Oostrom SH, Anema JR, Terluin B, Venema A, de Vet HC, van Mechelen W. Development of a workplace intervention for sick-listed employees with stress-related mental disorders: Intervention Mapping as a useful tool. BMC Health Serv Res. 2007;7:127.