INTRODUCTION
The purpose of this chapter is to review and discuss literature searches that addresses the link between `stress' and personal control in work settings. The review is limited to three topical areas: participative decision making and related managerial styles, research on the job characteristics model of Hackman and Oldham (1975, 1976) and locus of control. First a review of the relevant literature is presented. Then, a conceptual framework is presented for examining the role of control in work settings. The framework highlights the need to consider stress symptoms as responses to environmental control, and the need to consider attempts to increase control as responses to experienced stress.
 

1. Karasek’s ‘Job Demand-Control’ model

　　Initial attempts to relate stress in the workplace with health-related behaviours, highlighting heart disease date back to the early 1960s (Theorell & Karasek, 1996). Robert Karasek was one of the early contributors in the study of workplace social associations, stress-inducing causes and their effects on an individual’s health. In the 1970s, Karasek (1979) developed a bi-dimensional conceptual model that linked two aspects, demand and control in the workplace, to the risk of disease. The Demand-Control (DC) Model combines elements of two prior research traditions: the occupational stress tradition (e.g., Caplan, Cobb, French, van Harrison, & Pineau, 1976; Kahn, 1981) and the job redesign tradition (e.g., Hackman & Oldham, 1980). Both research traditions associated psychosocial job attributes to employee health. The occupational stress tradition focused on “stressors” at work, such as high workload, role conflict, mentally taxing work and role ambiguity (e.g., French & Kahn, 1962). The job redesign tradition centered mostly on job control aspect as its primary intend was to provide information on job redesign with the intention of augmenting the motivation, satisfaction, and performance at work. As suggested by Karasek (1979), the relationship between job demands placed on the employee and the employee’s ability/freedom to choose how to meet those demands (i.e., job control) is an important first step in the prediction of strain and of the employee’s active learning and growth.

　　From this point, these two fundamental principles, work demands and job control or decision latitude, is appropriate for further investigation, as it can provide insight in predicting mental strain. According to Karasek’s (1979) Job Demand-control model, work demands are occupational stressors where work inflicts on the worker and they can include working with speed and efficiency, the volume of work, the speed of work, conflicting tasks and demands, inadequate time and insufficient resources. Decision latitude includes decision authority and unrestricted control that workers can exercise control how their work will be completed, as well as skill discretion, which ascribes to the occasion to utilize an array of skills on the job (Karasek, 1979). Mental strain can be a result of heavy work load and low decision latitude or control (Karasek, 1979). In essence, heavy work demands induce worker’s level of arousal, typically accompanied by an increased heart rate, increased adrenalin, increased breathing rate and higher blood pressure (Karasek, 1979). According to Karasek, 1979, in a high control job, the worker has the ability to channel his or her own coping response to effectively reduce the arousal level. In a low job control, the employee’s unresolved strain may in turn accumulate and as it builds up can eventually manifest themselves as cardiovascular disease, emotional exhaustion, anxiety, depression, or psychosomatic related ailments (Fox et al., 1993; Karasek, 1979; Schaubroeck and Merritt, 1997, Cole et al., 2002; Perbellini, 2004).#p#分頁標題#e#

　　Karasek et al. expanded the model in 1982, to include social support as a third element and as an important aspect of today’s work environment. This review found that social and emotional support through colleagues, supervisors, social activities and informal rituals have beneficial effects in behavioral responses to stress. It provides a buffer between the stressors and the individual and is consistent with the idea that it may also facilitate coping patterns within the worker. Accordingly, the increased threat of illnesses is found in jobs with high work demands, low decision latitude (described as low intellectual discretion and low personal scheduled freedom) and little social support (Karasek et al., 1982.

　　Considerable empirical support of the model has been found. For example, Landisbergis (1988) showed that unrewarding work, depression, and symptoms of mental health were high in a study of New Jersey healthcare workers when a heavy workload and where their decision latitude was compromised. Similarly, Bromet et al. (1988) demonstrated a link between work demands and lack of control to a higher family-work conflicts and alcohol consumption. A study of 90 male manufacturing employees by Dwyer and Ganster (1990) showed that heavy work requirements and perceived low control were linked with absenteeism and tardiness.

　　Karasek's model of job strain has also been related to work satisfaction (Landsbergis, Schnall, Deitz, Friedman, & Pickering, 1992; Munro, Rodwell, & Harding, 1998; Parkes, Mendham, & von Rabenau, 1994). For example, Dollard and Winefield (1998) study among correctional officers used the Demand-Control/ Support model to understand the causes of the intention to leave their employment. Further endorsement for the model transpires from Fox et al.’s (1993) research of 136 registered nurses. The relations between workload with low control were positively related to increased blood pressure and cortisol levels and are a useful measure to predict job dissatisfaction (Fox et al.’s (1993).

　　These studies are consistent with the idea that passive jobs, with little opportunity for control correlates significantly to emotional exhaustion, depersonalization and lack of productivity (Landsbergis, 1998; Dollard and Winefield 1998; De Jonge J, Kompier; 1997, Kristensen; 1999, Van der Doef & Maes, 1999). Taken together, these results present apparent support of the consequences of adverse work environments on a workers health and well-being.
 

2.Empirical Literature - Main Findings

　　The Karasek Job Demand-Control model (Karasek, 1979, Karasek and Theorell, 1990) clearly confirms the stress hypothesis which has been linked to ill-health, such as heart- and vascular diseases (e.g. de Jonge & Kompier, 1997; van der Doef & Maes, 1998). To test Karasek’s hypothesis, Verhofstadt, Hans De Witte & Omey highlighted the research work of van der Doef and Maes (1999) where the authors examined 20 years of data on the Job Demand Control model. Van der Doef and Maes (1999) concluded that workers in the high-strain category experienced more distress, job dissatisfaction, burnout, hopelessness, tension, and anger. Moreover, Van der Doef and Maes (1999) reported that social support buffered negative health-related outcomes if it conformed to the specific demands of the job. However, as pointed out by Cohen & Wills (1985), that strong social support defined as instrumental help from others as work demands or time pressures are increased, is effective in preventing and reducing the negative effects of stress. Van der Doef and Maes (1999) concur that both job-related control and job-related social support provide positive and critical stress buffers. In this view, instrumental support from colleagues can help meet work deadlines (Van der Doef & Maes, 1999) and may consequently improve and lower the impact of stress (Cohen & Wills, 1985). Interestingly, there is substantial evidence that autonomy can also buffer the stressful impact of work demands on exhaustion, cynicism, depression and anxiety (Van der Doef & Maes, 1999). For example, a literature review has correlated psychological ill health including anxiety, depression, and emotional exhaustion and absenteeism to a variety of job aspects, including management effectiveness, excessive work load and pressures, control at work, and role ambiguity (Michie & Williams, 2003). These reviews have reaffirmed job stress as a risk factor for adverse physical and mental health outcomes.#p#分頁標題#e#

　　There has also been an emphasis on the role of higher workloads in physical health of workers (Netterstrom & Juel, 1988). These complex questions to the increase of disease and death rates were answered in Netterstrom & Juel (1988) Danish study which followed2465 bus drivers for a period of seven years. This study found that high traffic congestions that were interpreted by the workers as lack of control were more like to suffer with heart attacks (Netterstrom & Juel, 1988). In their study, they hypothesize that the incidenceof both death and hospitalization in those with excessive workloads (driving in heavy traffic) and negligible social support from colleagues were more than two times that in the group with low workloads (Netterstrom & Juel, 1988 cited in Michie & Cockcroft, 1996 ). In line with the literature, in a study of British bus drivers tackling heavy traffic, time pressures, multiple stops, passenger complaints and irregular working hours experienced higher job strain, which, in turn, increased stress hormones, cardiovascular risk factors, mental health complaints and musculoskeletal disorders (Evans & Carrere, 1991).

　　Evans, G, W., Carrere, & S. (1991). Traffic congestion, perceived control, and psychophysiological stress among urban bus drivers. Journal of Applied Psychology, 76, 658-663.

　　Evans (1994)concluded that more than half of all urban bus drivers retired prematurely with some form of medical disability. Consistent findings were revealed to be in line with Karasek’s job strain model which suggest that high job control may help alleviate stress (Evans & Carrere, 1991)

　　Falk, Hanson, Isacsson & Ostergren, (1992, cited in Michie & Crawford, 1996) tracked 500 Swedish men for seven years, concluded that the pace of work combined with low social interaction with colleagues increased the risk of heart attacks. In this contribution, Falk et al. (1992) found that job strain foretold mortality. In the research, job strain was affiliated with low personal freedom and exposure to high demands and the results attested that the mortality rate was notablyabaded in individuals who had good social networks and support outside the workplace, implying a buffering function on the negative side effect of stress (Falk, Hanson, Isacsson & Ostergren, 1992).

　　A further study found a correlation between physical activity at work and job responsibilities and mortality (Menotti & Seccareccia, 1985). The cohort consisted of 100,000 railway workers, aged between 40-59 years. The data collected included the men’s exercise habits and subjective ratings of workload, responsibility (accountability) stress and fatigue (Menotti & Seccareccia, 1985). Interestingly, the representative nature of the study permitted the researchers to conclude that lack of exercise at work and higher work responsibilities were linked to increased risk of heart attacks and other chronic illnesses (Menotti & Seccareccia, 1985).#p#分頁標題#e#

　　In similarity, a longitudinal one year review of health care personnel is also worth noting with respect to stress and heart disease (Doncevic, Theorell & Gianpaolo, 1988). Researchers noted that the nurses with the highest workload, defined by emotional exhaustion, job dissatisfaction and limited support networks, had thehighest levels of plasma cortisol in the morning, elevated bloodpressure and heart rate during working hours, and significant sleep disturbances (Doncevic, Theorell & Gianpaolo, 1988). In this study, the authors sought to demonstrate that the nurses with the highest autonomy, decision latitude and good social support exhibited the lowest stress (Doncevic, Theorell & Gianpaolo, 1988).

　　Finally, in testing the job strain hypothesis, Susan Michie and Anne Cockcroft, (1996) hypothesize in their article “Overwork can kill”, that high job demand may diminish performance and adaptationdue to poor memory and attentiveness and prolonged ineptnessand disorganisation, leading to a build-up of errors, accidents,and occupational hazards at work. The former and latter results seem to suggest that time pressures and energy committed to higher job demands may also impede the protective effects of social support and to information and coachingon evading accidents at work (Michie & Cockcroft, 1996). The combination of heavier workloads and diminished adaptation may also be linkedwith a work environment that is not conducive to a positive health and safety culture (Di Martino, 2003.)

　　Ten years of study have correlated high jobdemands (workload and pace) and low decision latitude (individualcontrol at work, decision making, and skill training) to diverse threats comprising (Michie & Cockcroft, 1996) everything from heart attacks, strokes, musculoskeletal disease and mortality to increased absenteeism, turnover and low moral (De Jonge & Kompier, 1997).

　　In summary, Michie & Cockcroft (1996) confirm that these outcomes are conformable with Karasek's job strain hypothesis (Karasek & Theorell, 1990) which demonstrates that the more job demands are stressful, whereas more control in the job leads to a decrease in stress among workers. Furthermore, the evidence also suggests that work stress is rapidly emerging as a consistent factor to work-related disease and injury, and as a considerable deleterious contributor to the overall burden of disease in society.
 

3. Criticism of the Karasek model

　　Karasek’s ‘Job Demand-Control’-model (1979; also see Karasek & Theorell, 1990) has rapidly moved into a dominant position in the field of work and occupational health. One of the reasons for this may be its simplicity and broad applicability: the model integrates various older research traditions such as Hackman & Oldham (1980) work motivation theoretical framework and the Michigan model of organizational stress (Kahn, Wolfe, Quinn, Snoek & Rosenthal, 1964) to identify role conflict, role ambiguity, and role overload as critical sources of stress in organizations, and reduces these stressors to only two basic dimensions, namely the psychological demands and job control when analyzing jobs. The model’s simplicity has been criticized, however, (e.g. Johnson & Hall, 1988) as it disregards some crucial job dimensions. Johnson and Hall (1988) show a strong association involving a much larger number of variables between demands and decision latitude. Controlling and deciphering these environmental work factors and job circumstances create a wide variety of uncontrollable variables (e.g. social support, physical stressors, job characteristics, meaningfulness of the job, social class, and personality). These inconsistencies which incorporate personality, organizational and societal aspects may manipulate the outcomes of perceived autonomy and control (Johnson & Hall, 1988). Karasek’s operationalization of control, more particularly, his assessment of decision latitude has been further criticized as including work factors that do not typify control, such as utilization of skills, job enrichment, the use of creativity, or leaning opportunities (Ganster 1988, Smith, Tisak, Hahn & Schmeider, 1997; Wall & Jackson, 1996). This limitation was validated in a study by Wall and Jackson, in which the moderating role of control was explored utilizing two divergent measures of control: Karasek’s decision latitude and a more focused measure of timing (freedom to determine the pace of work) and method control (control of how the work is done). The interface between job demand and the focused control measure forecasted job strain (job dissatisfaction, depression and anxiety) whereas the interaction between job demand and Karasek’s decision latitude did not. As suggested by Ganster (1988) this can be addressed in other ways. First by demonstrating that for control to be a factor, it must be conceptualized as a multi-faceted framework to correspond to the explicit stressors confronted in the workplace. He developed measures that addressed seven work domains in which individuals may or may not have control: work scheduling, physical work environment (air quality, temperature, and humidity), work pacing, decision-making, task demands (work pressures, overtime work, shift work), social environment at work, and career mobility. In addition, Ganster (1988) included predictability of work activities and work demands as an ameliorator which arguably provides security and certainty in people’s coping mechanism.#p#分頁標題#e#

　　Similarly, Bacharach, Bamberger, Conley and Bauer (1990) have found empirical support for the multiple construct of decision participation, which provides workers a feeling of efficacy and job satisfaction by offering a sense of control in managing work stress. These researchers also acknowledge personality characteristics and individual differences in the desired levels of participation in the workplace, as one employee might view decision participation as a stressor in itself, while another might find that participation provide the tools to successfully cope with job demands. They optimally attribute decision participation as a discrepancy between perceived and expected or desired levels of participation in decision-making (Bacharach et al, 1988). They strategically recommend a typology of participation domains supported by two dimensions of decision-making: strategic versus operational and individual versus organizational (Bacharach et al, 1988).

　　A further critique of research attempting to provide empirical support for the moderating role of control is the assumption of linearity. A curvilinear relationship between level of control and strain has been suggested in that too little control may lead to stress, but too much control is damaging when it induces a sense of too much responsibility and decision-making (de Jonge & Kompier, 1997).

　　In another study, Schaubroeck and Merritt (1997) found that job control was predicated to buffer the negative effects of demands on health among the more efficacious workers, whereas it had the opposite effects among the more inefficacious ones. Schaubroeck and Merritt's (1997) obtained data indicating that self-efficacy may play an important role in reducing cardiovascular consequences of job related strain. They further conclude that individuals with low self-efficacy tend to feel threatened with increased responsibilities and challenges, will easily discourage themselves, self-blame, suffer anxiety and as shown in Seligman’s (1975) study, intensify their “learned helplessness”. In a similar vein, Bandura (1992, 1997) asserts that individuals with high-efficacy appreciate challenges; demonstrate better judgment, intensify their efforts to succeed in a particular situation and recover quickly from setbacks and disappointments.

　　This perspective is in keeping with Jimmieson’s (1984) framework. According to Jimmieson’s (1984), an individual with high self-efficacy are able to implement strategies to cope with job demands. Evidence shows that self-efficacy reduces the negative health consequences of stress (Jimmieson’s, 1984). As such, both self-efficacy and social support are seen as effective characteristics to prevail over stressful demands (Jimmieson’s, 1984). This “personal control” enables an individual to feel confident during the course of their daily lives and buffers the negative impact of workplace stressors (Jimmieson’s, 1984). Extending the findings from Schaubroeck and Merritt’s (1997) study, these individuals demonstrate better coping behaviours to handle demanding work tasks. Similarly, Welch & West (1995) make a very strong case that highly efficacious individuals who are unable to exercise control in the workplace, will face many struggles and negative health consequences. In such cases, efficacy values are confronted and frustration sets in (Fisher, 1984). Thus, low control is harmful for efficacious workers because they are powerless to resolve their job problems (Fisher, 1984).#p#分頁標題#e#

　　Models of occupational stress contain a number of alternative explanations for the role of self-efficacy in the demands-control interaction (Karasek, 1970). Karasek utilizes the term “job decision latitude” to symbolize control (Karasek, 1970). Conversely, efficacious individuals who lack job control can experience dangerous health effects (Averill, 1973). As noted by Schaubroeck and Merritt’s (1997) Karasek's (1979) demands-control hypothesis only pertain to individuals with high self-efficacy.

　　Turning to a more general mental health outcome, a UK study investigated a connection between personality, negative affectivity and the risk of poor mental health from organizational stress (Blomkvist, Eriksen, Theorell, Ulrich & Rasmanis, 2005). Surprisingly, the results indicated that personality had little effect on mental health in relation to job control (Blomkvist et al. 2005).

　　In the stress measurement literature, researchers have identified some possible antecedent constructs with lower socio-economic status. For example, consistent evidence demonstrate that low job control and high physical demands are more common among lower status occupations, particularly with machine-paced production or scheduled performance requirements, whereas higher psychologic demands combined with greater job control are more common among high-status occupations, which have professional entry criteria and higher compensation (Belkic et al. 2000). Evidence is increasing that demonstrate the growing risks in coronary heart disease, high blood pressure, and cholesterol, as well as other health outcomes (Belkic et al. 2000). Belkic (2000) argues on whether social disadvantage confound the relationships observed between high job stress, which are characterized with the unhealthy combination of high demands and limited discretionary autonomy with adverse health outcomes.

　　Strictly speaking, her observational studies have addressed these challenges in two ways. First, by assessing whether the job stress is associated with health outcome within the lower disadvantage group and secondly, she adds that most positive studies of job stress and heart disease have controls for social class (Belkic et al. 2000). She further makes her point, by stressing a 2004 well-conducted Australian research realized by Strazdins et al (2004). Their sample was drawn from a collection of advantaged professionals and managers aged 40-44, where they reported a fairly strong association between job stress and adverse physical and mental health outcomes. They concluded that the negative health outcome associated with stress should not be confounded by lower social disadvantage (Strazdins et al. 2004).

　　We have now set the stage for our conceptualization of control and work behaviours as a special case of job stress. It is important to distinguish between formal and informal control, as it is to distinguish between actual and illusory control. Although the formal organizational structure may limit control, informal norms may arise that allow for increased or even decreased employee control.#p#分頁標題#e#

　　However, the effects of demands on health are complex and vary considerably according to individual differences. Research supports the view that having high levels of control over one’s job responsibilities can rebound if an individual lacks the confidence on the job or has a predisposition to take the blame for negative results at work (Schaubroeck, Jones & Xie, 2001). More accurately, job control develops into recurring disappointment and a validation for self-blame (Schaubroeck, Jones & Xie, 2001).

　　It seems clear from the research and that of Schaubroeck (1997, 2001) and his colleagues, that for an individual to manage stress, more self-efficacy in the workplace is generally better than having none.

　　The mounting evidence suggests that job stress is quickly looming as the single greatest threat of injury and work-related illnesses, and as a significant contributor to diseases in our society. Adjustment for personality traits, inefficacious individual, self-efficacy and lower status occupations may attenuate the approaches on how some individual handle stress, but it does not eliminate its negative health outcomes. Today’s organizations are adversely affected through the consequences of absenteeism, antecedent to commitment, inadequate job performance, turnover, lack of productivity, and other human and financial tolls.

　　Although Karasek’s model has attracted criticism in some quarters, others have conjured that coupled with knowledge on the effect of stress (mental health, sickness, absenteeism) it can depict a useful, rapidly administered diagnostic tool. The model also offers a basic intervention tool for organizations to execute organizational changes and intervention programs to address identified sources of workplace stress.
 

4.The Effects of Occupational Stress Management Intervention Programs

　　“Stress is the number one disease of the 21st century.”

　　Rob Weingust

　　The topic of occupational stress has received considerable research attention over the last decade and has emerged as an important occupational safety and health concern. Workers' compensation claims for stress-related illnesses, for example, are the fastest growing type of claim, comprising more than 11% of all such claims: (Statistics Canada, 2002).

Stress-related absences cost Canadian employers approximately over $3.5 billion each year. (Statistics Canada, 2002)

Health care costs are almost 50% greater for individuals who disclose elevated levels of stress. (Statistics Canada, 2002)

83.1% of Canadian workers acknowledged stress as the most important health worries within their company. (Canadian Mental Health Association, 2002)

Stress causes 19% of absenteeism expenditures, 40% of turnover costs, 55% of EAP expenses, 30% of STD and LTD costs, 60% of accidents at the workplace, and 10% of cost of prescriptions. (Health Canada, 2002)#p#分頁標題#e#

　　Occupational stress has become a real and pervasive challenge in the workplace: stress-related costs are currently estimated at more than $33 billion annually (Statistics Canada) and the Canadian Centre for Occupational Health and Safety rates stress as one of the ten leading work-related diseases (CCOHS, Forum 2005). Over a period of five years in the United States, stress-related disability claims have continued to grow by almost 700%, with the unequivocal cost to settle a stress claim averaging between $10,000 and $15,000 (Stevens, 1992.)

　　Of the many proposed by those authors studying in the field, three common stress intervention strategy frameworks have been identified as relevant to the understanding of stress management in the workplace (Mullen, 1989; Newman & Beehr, 1979; Murphy, 1988). They are classified as primary, secondary (both preventative measures) and tertiary (curative) (Cahill, 1996; Hurrell & Murphy, 1996).

　　The nature of interventions at primary level is preventative. The target is a reduction in the number and/or intensity of stressors so that stress can be prevented or eliminated (Clarke & Cooper, 2004). This approach aims at work environments, technologies or organizational structures (Clarke & Cooper, 2004). Examples of interventions are job redesign, clear job descriptions, employee participation, increasing employee control, or flexible working arrangements (Clarke & Cooper, 2004).

　　Another ameliorative secondary intervention is of a preventative/reactive nature (Clarke & Cooper, 2004). They aim at modifying individual responses to stressors (Clarke & Cooper, 2004) by helping employees improve their coping capacity. Examples are stress management training, communication and information sharing (Clarke & Cooper, 2004; Cox & Cox, 1990).

　　Finally, tertiary interventions focus on treatment of employees already suffering stress-related illness (Clarke & Cooper, 2004; Cox, Leather & Cox, 1990). They aim at minimizing the damaging consequences of stressors by helping individuals to cope better. Examples are employee assistance programs or stress counseling (Clarke & Cooper, 2004; Cox, Leather & Cox, 1990). Tertiary interventions are not usually regarded as preventative programs (Cox, Leather & Cox, 1990). According to Cox et. al (1990), in the average workplace, it may be more effectual to create a comprehensive stress prevention program which would include all three interventions. Cooper et al. (2001) noted that organizations mainly focus their stress management strategies on individual worker. The authors underscore that effective stress management interventions include organizational level interventions as compared to focusing on the individual (Cooper et al, 2001). Giga et al (2003) presents a framework that recommends a more comprehensive coverage focused on integrating management support, strategic solutions and commitment combined with work participation. Models focusing on the physical work environment (noise level, job redesign, workload reduction) and other aspects of the organization (coping skills training, employee assistance programs, conflict management training) have greater preventive potential as effective intervention approaches than concentrating on individual employees (Hurrell & Murphy, 1996; Melamed & Froom, 2000). Consequently, primary prevention is more effectual than secondary (they handle problem at the source), and secondary (proactive), is more helpful than tertiary (van Dierendonck, Schaufeli & Buunk, 1998). Essentially, combining both approaches (primary and secondary) are more valuable to the employee and organization (Halperin, 1996). The systems approach (dealing with problems at their source) typology described is generally synonymous with most other ‘best practice’ models, all of which acknowledge the need to address both work organisations and individual levels (Karasek, 2004; Kompier & Cooper, 1999; Hurrell & Murphy, 1996; Bond, 2004). Yet a number of studies have concluded that the effectiveness of the different approaches to these interventions are difficult to evaluate because of the heterogeneity of the methods, study designs and even target groups (Van der Hek & Plomp, 1997). According to Ivancevich et al. (1990), most professionals choose to focus their intervention efforts on changing individual behaviours because most are more comfortable with changing the individual rather than their organizations. As noted by Fleishman & Quaintance (1984), and as expected, psychologists have been more interested in evaluating human responses and actions than situational factors.#p#分頁標題#e#

　　Van der Hek and Plomp (1997) reviewed 24 studies on the effects of occupational stress management programs published between 1987 and 1994. Only two of them evaluated interventions on the individual-organizational interface level (restructuring of jobs, coping skills and support groups) and two on the organizational level (1 year organization-wide stress management program) (van der Hek & Plomp, 1997). These results provide firm evidence that an organization-wide stress intervention approach provides the most favourable outcome on the individual and the entire organization (van der Hek & Plomp, 1997).

　　An in-depth analysis of 90 cost reduction studies showed the impact of the growth in ‘lean production’ management methods with respect to associated effects on job stress (Landsbergis, Cahill & Schnall, 1999). Lean production is an assembly-line manufacturing methodology developed originally for Toyota and the manufacture of automobiles. In its most basic form, lean production is the methodical eradication of misuse – overproduction, waiting, shipping/haulage, inventory, wasted time, over-processing, substandard products – and the execution of the model of continuous flow and customer pull (customer tells a company what to produce) (Landsbergis et all, 1999). Landsbergis et al. (1999) established little substantiation that lean production environments empowered workers or reduced their work stress. Conversely, they seemed to intensify work pace and demands. Increases in decision authority and skill are very reticent and temporary, and decision latitude remains low. Therefore, the scheme of “lean" work ideology (e.g., an understaffed, high demands, repetitive work, overtime) is more likely to amplify the occurrence of stress-related physical illnesses (e.g., hypertension, cardiovascular disease) and mental health (e.g., depression) outcomes (Landsbergis et al., 1999).

　　In Cox et al (2000) collaborative efforts regarding organizational interventions on work stress, recommend a balance approach tailored to the problems of each organization. They suggest a detailed risk assessment practices to create solutions best suited for the organization. As Cox (1993) indicates, selecting a solution, should be based by analysing the current problems instead of using cookie cutter designs which may be projected at the individual rather than the organization.
 

5.Concluding Comments

　　Over the past two decades, research into worker’s well-being and stress at work has been greatly influenced by the demand-control-support models of Robert Karasek (1979; Karasek & Theorell, 1990). These models propose that worker strain and active learning are determined by particular combination of job demands, job control and social support. Specifically, incumbents of jobs that are high in demands, low in control, and low in support are expected to show high levels of strain, while incumbents of jobs that are high in all three job factors are expected to display high levels of activity, learning and participation, both on the off the job. The models also propose that prolonged exposure to combinations of these job conditions influence workers accumulated anxiety and sense of mastery.#p#分頁標題#e#

　　Kristensen (1995, p.18) notes that it is “no exaggeration to say that (his) has been the most influential model in the research on psychosocial work environment, stress and disease for the past ten year”. However, Karasek’s model has also been criticized for their breadth, simplicity, and lack of conceptual clarity. A number of writers have argued that in the field of stress research requires distinguishing between specific types of job demands and control. Their argument is that different dimensions of the work environment have different affects upon strain, for example, personality, work environment, the culture of the organization, and social and economic environments. I agree with the argument that the psychosocial work environment extends beyond the characteristics of the job itself, and stress management consultants need to focus on the nature and quality of workplace atmosphere. The quality of your work environment extends beyond the characteristics of the job itself, but evolves around social and interpersonal relationships and it is the result of these relationships with your peers and management which will influence how an individual will handle and manage their stress level. Positive and supportive relationships will nonetheless contribute to favourable work attitudes and behaviours. As simplistic as the model is, the theory has profound implications for the design of healthy work environments, seeing that Karasek clearly indicates that stress poses a real risk to staff’s health and wellbeing. Depression, anxiety, work absences and other work related diseases were clearly linked with the presence of greater work demands and low job control.

　　Occupational stress is a complex issue. Stress intervention approaches that incorporate organizationally focused primary and secondary interventions are likely to be more successful in reducing the level of stress in the workplace.

　　Developing and maintaining a good work culture with a reasonable amount of stress should not be about managing absenteeism or increasing productivity but should be about quality of life issues, such as work-life home balance, manageable work hours and workloads, clarity in one’s career direction, security at work and being a valuable member to the company. Therefore, it is important to acknowledge stress as an important issue and it should be a strong focus on occupational health and reduction of sickness in the workplace.