The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.

htm

Eustress, distress and their interpretation in primary and secondary occupational stress management interventions: which way rst?
Mark Le Fevre
Faculty of Health and Environmental Sciences and Faculty of Business, Auckland University of Technology, Auckland, New Zealand

Stress management interventions 547

Received May 2006 Revised June 2006 Accepted June 2006

Gregory S. Kolt
Faculty of Health and Environmental Sciences, Auckland University of Technology, Auckland, New Zealand, and

Jonathan Matheny
College of Business, Massey University, Auckland, New Zealand
Abstract
Purpose To develop an argument for the retention of secondary approaches to stress management (those that focus on the individual within the organization) as rst interventions, prior to the employment of primary approaches (those that focus on the organizations processes and structures). This is based on a reconsideration of eustress versus distress and a review of current empirical evidence on the effectiveness of stress management interventions. Design/methodology/approach Major empirical studies and reviews are critically reviewed and placed within a theoretical framework derived from both early and more recent work in the eld. Findings There is little empirical evidence on which to base recommendations for organization-based stress management interventions as rst or sole approaches and therefore the value of these as rst or sole approaches is questioned. Instead secondary, individual-focused, approaches are recommended as rst-line interventions prior to the adoption of organization-based interventions. Practical implications In practice secondary stress management approaches are currently most common. Broader primary approaches appear to have excellent theoretical support and a growing body of supportive literature and accompanying recommendations for employment. We suggest, however, that secondary approaches be employed prior to the introduction of primary methodologies within a client organization. Originality/value This paper provides a review and framework for interpreting/understanding the research on the effectiveness of stress management interventions and makes recommendations relevant to practitioners in the eld. Keywords Stress, Workplace Paper type Viewpoint
Journal of Managerial Psychology Vol. 21 No. 6, 2006 pp. 547-565 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940610684391

Stress, and occupational stress in particular, has become a major area of interest not just only for academics but also at the industry and government level. This is in terms

JMP 21,6

548

of its potential cost to individuals and organizations and in what may be done to reduce the experience of stress in the workplace. As Cox et al., 2002, p. iii have said . . . work-related stress is currently one of the greatest challenges to the health of working people and to the healthiness of their work organizations. Reducing the cost of occupational stress at the levels of the organization and of national economies is also becoming a major concern. Although individual level stress management interventions still seem to be the most commonly adopted approach to reduce occupational stress in practice, there is an increasingly dominant view in the literature that more broadly focused organizational approaches should be the interventions of choice (Cousins et al., 2004). The reasons for this emphasis seem to be primarily based in a risk management approach wherein the removal or reduction of real or potential sources of stress is preferred to the moderation or amelioration of their effects, or the treatment of manifested damage (Quick et al., 1997). The logic and attractiveness of this stressor reduction approach is obvious, although empirical evidence for its effectiveness is largely lacking (van der Klink et al., 2001). This organizational level approach seems not to fully acknowledge that individuals may react differently to potential stressors, and that these differences in reaction may be rooted in both individual differences and in the context in which these potential stressors exist (Le Fevre et al., 2003). Reintroducing eustress as an entity distinct from distress, we believe, also helps to clarify the process through which external pressure may, or may not, result in distress, and may also shed light on the very mixed results that have been achieved through stress management interventions to date. Some authors also suggest that individual-based stress reduction approaches are less acceptable as they place the responsibility for stress with the individual rather than with the organization and its managers. We have a concern that there is a danger here of throwing the baby out with the bathwater so to speak, in that if the increasing call for organizational level stress management interventions is taken up to the detriment of individual level interventions, then the ultimate result for both organizations and the individuals who work within them may be less advantageous than they could otherwise have been. This paper provides a critical review of the major empirical studies regarding the effectiveness of current stress management interventions. Placing these intervention strategies within a theoretical framework derived from the work of early researchers such as Selye (1936, 1956, 1964, 1983, 1987) and Cannon (1914, 1932) as well as more recent contributors (Le Fevre et al., 2003) we suggest that, on current evidence, individual focused stress management interventions should continue to be considered as a rst line of defense against the potential hazards of occupational stress rather than as mere adjuncts to broad organization-based risk, or stressor, reduction interventions. Stress management interventions Stress management interventions (SMIs) might be dened as any purposeful action taken to reduce or alleviate the stress experienced by organizational citizens in the execution of their work functions. This broad denition excludes all those actions whose primary purpose is other than the management of stress but which may, nevertheless, have impact on it, (i.e. the majority of managerial actions, or inactions). SMIs are thus all those interventions or managerial actions or initiatives that have

 tre, the preferably as their principle purpose, or at least as part of their raison de management or manipulation of environmental stressors or individual stresses in the organization and its members. Stress management interventions can be classied as primary (i.e. those that aim to deal with the source(s) of stress in the workplace (stressors), and focus at the organizational, or group level), secondary (i.e. those that focus at the level of the individual employee), or tertiary (i.e. those that aim to address or ameliorate already existing stress signs and symptoms in individual organizational members) (Quick et al., 1997). Primary interventions, may have a job structure focus including such aspects as task, role, interpersonal and physical demands (Quick et al., 1998), job rotation or design (Hackman and Oldham, 1976, 1980), or organizational restructuring, and organizational development (OD) interventions (De Frank and Cooper, 1987, van der Hek and Plomp, 1997; van der Klink et al., 2001). Primary interventions are relatively long, usually running for over 12 months in the organization. It is noteworthy that non-treatment control groups are usually lacking in reported trials of such interventions (van der Hek and Plomp, 1997; van der Klink et al., 2001) though Randall et al. (2005), have used adapted study designs to help overcome this limitation. By comparison, secondary interventions tend to be relatively brief and can be classied into three main types. The rst, somatic, includes relaxation methods such as progressive relaxation (Jacobson, 1938), bio-feedback (Fentress et al., 1986), and breathing techniques. The second, cognitive, includes mindfulness techniques (Kutz et al., 1985), afrmations, and thought stopping. The third type of secondary intervention, multimodal, combines aspects of both somatic and cognitive technique such as Bensons (1976) relaxation response, transcendental meditation, and comprehensive programs that mix somatic and cognitive methods (Benson, 1976; Benson et al., 1977). Tertiary intervention has the reduction of organizational and individual costs, due to manifest health problems, as its main aim. At the individual level this may often be represented by the provision of medical or psychiatric care, counseling, or employee assistance programs (De Frank and Cooper, 1987; Quick et al., 1997; van der Hek and Plomp, 1997). The typology of stress management interventions as primary, secondary, or tertiary was developed as a framework for preventive stress management based on the public health model of prevention wherein early (primary) intervention to reduce or eliminate health risk factors where possible is considered the ideal approach (Quick et al., 1997): (e.g. the elimination of the smallpox virus (Variola) from the human population). Tertiary interventions, being primarily concerned with the treatment of manifested symptomatic disease (physical or psychological) in affected individuals, are beyond the scope of this paper. Primary stress management interventions Primary SMIs are usually regarded as being proactive (i.e. attempting to remove or alleviate the sources of occupational stress) rather than reactive, (i.e. attempting to relieve or reduce the resulting strain manifested in individuals) (Cooper et al., 2001). Primary SMIs therefore concern themselves with eliciting which aspects of the job and the wider organizational context are real or potential stressors, and making adjustments to the way jobs are performed and structured, their physical environment, formal reporting structure, and policies (Cox et al., 2002). Approaches like these may also perhaps attempt to change or modify the organizational culture. By their nature,

Stress management interventions 549

JMP 21,6

550

primary SMIs tend to be long-term efforts though there are some exceptions as in the recent work of Randall et al. (2005). Though primary SMIs are sometimes suggested as more appropriate and desirable forms of intervention than secondary SMIs (Cooper et al., 2001), there is surprisingly little empirical evidence to support this contention. Jones et al (1988) reported a one-year primary SMI carried out in a medical setting with the aim of reducing medical error and the associated claims. Relative to a matched control group, the client system had signicantly reduced claims over the one-year follow-up period. Other reported primary SMIs (Golembiewski et al., 1987) used no control group. Even Jones et al.s (1988) work cannot be considered to be a pure primary SMI as participants were also trained in relaxation, thus confounding primary and secondary interventions. Primary, organization-based interventions are, however, suggested as the base approach for preventive stress management in organizations, while individual-based approaches are regarded as complementary only (Quick et al., 1997). One of the difculties here may lie in the way stress in the organization and changes in that stress are measured. Work carried out for the UK Health and Safety Executive may help clarify this as a standards-based approach to occupational stress assessment (which includes the use of standardised employee survey instruments) has been developed as part of a development project (Cousins et al., 2004). Randall et al. (2005) recently reported two brief organization-based interventions where simple changes were made to the authority or resourcing available to participants. Using an adapted study design in which participants were assessed on the degree to which they were exposed to the changes rather than starting with specic test and control cohorts, positive responses to the interventions were shown at the individual level using the exhaustion scale of the general well-being questionnaire (Cox and Grifths, 1995; Cox et al., 1983). This approach may help to reduce the current lack of evidence for the effectiveness of job focussed organizational interventions. In an extended evaluation of stress control interventions in hospital staff, however, although improvements were noted in almost all of the specic managerial and organizational areas addressed by interventions, (i.e. inadequate information, lack of praise and recognition, or lack of time, etc.) more general measures of group stress such as feelings of being worn out or the experience of musculoskeletal pain showed little change, or in some cases were increased (Cox et al., 2002). As van der Hek and Plomp (1997) suggested, assessment of the separate sub-components of SMIs to nd out which parts are effective should be a focus for research on stress. Secondary stress management interventions As discussed previously, secondary SMIs are focussed at the level of the individual working within the organization and are commonly classied as being primarily cognitive, somatic, or multimodal in nature. Secondary SMIs are sometimes criticised as placing the onus for managing stress on individuals and so, by implication, releasing management from any responsibility for ensuring that those aspects of the organization within the ambit of their control are structured to reduce their distressful impacts (Cooper et al., 2001). This stance has a certain moral appeal and has been used to justify recommendations that primary SMIs should be the rst choice in managing organizational stress (New Zealand Government, 2003). Current empirical evidence, however, suggests that secondary SMIs have a far better record for showing demonstrable effects (van der Klink et al., 2001).

Secondary SMIs are highly variable, ranging from short one-off relaxation or meditation training sessions to multimodal approaches involving training and feedback over a period of weeks to months. Assessing the effectiveness of this wide range of approaches is also complicated by the heterogeneity of outcome measures used (van der Hek and Plomp, 1997, van der Klink et al., 2001). As van der Klink et al. reported, however, all individual-based interventions showed positive effects and cognitive approaches may be more effective than somatic methods. Further, research making direct comparison between somatic and cognitive techniques may help to clarify or conrm this. Stress distress and eustress The concept of eustress is important in any discussion of stress, especially occupational stress and interventions designed to relieve or reduce occupational stress. As Selye (1956) originally proposed and as later authors (Le Fevre et al., 2003) have supported, a major determinant of whether any given stimulus eventually represents or invokes stress (distress) in an individual depends on their perception and interpretation of that stimulus in the broad context in which it is presented. Aspects important to the interpretive act may include the timing, source, and perceived desirability of the stimulus and the degree to which the individual perceives they have control over it (Le Fevre et al., 2003). Without the useful term eustress to represent positive responses to external stressors, and to differentiate them from the negative or distressful responses that are the frequent sole topic of study, discussion of the broad area of stress becomes rather one-sided, stilted, and difcult to construct. The very fact that eustress seems to have fallen out of discussion may contribute to an overly narrow view of possible stimulus-response and environment-response processes in the occupational stress research agenda. In Selyes (1956, 1964) work good stress or eustress depended not on amount of stimulus as suggested by the frequently illustrated inverted U stress/performance diagram (Certo, 2003; Lussier, 2002; Schermerhorn, 2003), but primarily on how that stressor was interpreted by the individual experiencing it and how they chose to react to it (Selye, 1956, 1964, 1987). In a consistent manner, Harris (1970) equated eustress with pleasurable reactions to stressors and Edwards and Cooper (1988) dened eustress as a positive discrepancy between perceptions and desires (provided that the discrepancy is salient to the individual). Nelson and Simmons (2003) also suggested that eustress and distress are separate and independent aspects of the overall stress response. The point these authors hold in common is that eustress is primarily a result of positive perception of the stressors, and that distress, therefore, is a result of negative perception of the stressors. Whether a particular stressor represents eustress or distress is determined not only by the individuals perception of the amount of demand it represents, but also by their perception of its other characteristics such as its source, timing, the degree to which they have control over it, and the degree to which they consider it desirable. If eustress and distress are considered to be separate constructs it is also possible that both eustressful and distressful responses may occur simultaneously to the same set of environmental stimuli (Le Fevre et al., 2003). The empirical study by Cavanaugh et al. (2000) appears to support the concept of eustress. Cavanaugh et al. differentiated self-report stress into challenge stress and hindrance stress. Challenge stress was dened as . . . self reported work stress associated with

Stress management interventions 551

JMP 21,6

552

challenging job demands (p. 66), and likened to eustress. Hindrance stress was dened as . . . job demands or work circumstances that involve excessive or undesirable constraints that interfere with or hinder an individuals ability to achieve valued goals (p. 67), and likened to distress. This study showed support for the proposition that self-reported stress shows a different relationship to work outcomes, such as job satisfaction and job search behavior, depending on whether the stressor represents challenge (eustress) or hindrance (distress). The emerging interest in positive psychology and its relevance to occupational stress (Quick et al., 2005; Simmons et al., 2004) suggests that eustress may be an old concept that is now due re-examination and re-integration with stress in the wider sense. Terminology in the reporting of stress research is also both confusing and inconsistent. The terms stress and strain are now often used rather than the earlier terms stressor and stress to denote environmental forces and the individuals responses, respectively. Adding to this confusion is the semantic shift that seems to have occurred by which stress, which Selye (1964) originally dened as an inescapable consequence of living, has become synonymous with distress (Le Fevre et al., 2003), the negative, perhaps even maladaptive response to some of the demands of life. This leaves us bereft of any term to use for those responses to the pressures or demands of life and work that are positive or adaptive. In the absence of the eustress concept one is logically left with the single operational alternative of reducing environmental pressure, demand, work, or whatever other variety of potential stressor is relevant, as without eustress, stressors must inevitably lead to distress; where else? If one accepts the eustress concept, as a possible positive response to stressors then, and only then, do the wide range of individual and organization-based stress management interventions make sense since there will always be some irreducible level of environmental stimulus or pressure impinging on organizational citizens to which some kind of response is inevitable. Given that some degree of pressure to perform is an inescapable concomitant of being employed then it seems to us that efforts to assist individuals to achieve more positive interpretations of, and reactions to, their task environment, (i.e. a eustressful rather than distressful interpretation), should be encouraged. Empirical reviews of SMI success in practice Since Newman and Beehr, 1979 early review of the occupational stress literature, which focused primarily on qualitative reports, several major reviews of the empirical literature on str ess management interventions have been published (De Frank and Cooper, 1987; Giga et al., 2003; Murphy, 1996; van der Hek and Plomp, 1997; van der Klink et al., 2001). De Frank and Cooper (1987) reviewed the literature published between 1977 and 1987, and reported the extensive variation on outcome measures used to assess the effectiveness of the interventions reported, and the relative lack of controlled studies. Only 18 of the abstracts reviewed fullled the criteria of having control groups. There was also minimal long-term follow up in any of the reported trials, six months being the longest period reported. van der Hek and Plomp (1997) continued the work, reviewing the literature between 1987 and 1994. Again, of 342 abstracts examined only 37 met the authors criteria of having some kind of evaluation of outcomes, and seven of these were subsequently rejected as their evaluation was purely anecdotal. van der Hek and Plomp (1997) again commented on the lack of long-term follow up.

Murphy (1996), carried out a review of the literature regarding the health effects of stress management interventions covering papers published in the period 1974-1994. Of the 64 papers reviewed, eight included some form of organizational outcome measure. In six papers this was absenteeism. In four of these papers absenteeism improved in the treatment group relative to controls. One of the remaining papers showed no signicant difference between treatment and control groups while the other showed a signicant increase in absenteeism in the treatment group over the measurement period. Overall, a majority of the studies showed some positive effect of stress management intervention on health-outcome measures at the individual level. van der Klink et al. (2001) carried out a more narrowly-focused study, using meta-analytical technique to assess whether stress interventions were effective, and if so, which types were most effective. The literature between 1977 and 1996 was searched. Only 48 studies fullled the authors four main criteria, which were: interventions were designed to reduce psychological complaints related to occupational strain, subjects were drawn from a working population on the basis of already manifested strain, experimental or quasi-experimental design using non-treatment controls was used, and well-dened and reliable outcome measures were employed. The second criterion of van der Klink et al., that subjects should be drawn on the basis of already manifested strain, was a major departure from earlier reviews and had a large role in excluding many trials resulting in the small number of studies included in the analysis. Very few studies had selected subjects on the basis of demonstrated pre-existing strain. Failure to do so may compromise the ability to demonstrate positive change (van der Klink et al., 2001). Any preventative effect of an intervention may also be difcult to demonstrate in the absence of pre-existing strain without longer-term follow up than is common in SMI research (De Frank and Cooper, 1987; van der Hek and Plomp, 1997; van der Klink et al., 2001). van der Klink et al. (2001) found that, in general, employees did benet from stress reduction interventions, that secondary interventions were more effective than primary interventions, that cognitive techniques were more effective than somatic techniques, that all individual-based interventions showed positive effects, and that there appeared to be an inverse relationship between length of intervention and effectiveness of intervention. While generally encouraging, the ndings of van der Klink et al. (2001) do require closer examination. Previous trials and reviews by Barkham and Shapiro (1990) and van der Hek and Plomp (1997) found that brief interventions focused at the individual level do seem to have higher effectiveness than more lengthy programs. In the van der Klink et al. (2001) paper, the relatively low effect reported for primary interventions is confounded with the fact that the primary interventions tend to be those with longer time frames. The nding of lower effect in primary interventions is interesting, however, as it tends to contradict the more general opinion expressed in qualitative reviews that primary interventions should be more effective than secondary as they are designed to reduce the number of environmental stressors rather than treat the resulting stress (Danna and Grifn, 1999). van der Hek and Plomp (1997) found no correlation between the length of interventions and their effects although they conducted a review rather than a quantitative meta-analysis of the data. This lack of clarity in the relationship between length of intervention and potential gain must raise questions in the minds of those who purchase and recommend stress management interventions, in terms of the cost relative to the benet.

Stress management interventions 553

JMP 21,6

554

This variation in type and duration of intervention, and outcome measures employed, may be related to the lack of any universally accepted denition of stress and strain. The majority of outcome measures reported by van der Klink et al. (2001) were subjective, were determined by self-report instruments, and were focused at the individual level. These outcome measures can be summarized under three headings: (1) quality of work life, which includes assessments of job demand, work pressure, job control, working conditions, and social support; (2) psychological resources and response, which include assessments of self-esteem, mastery, and coping skills; and (3) complaints, which include assessments of stress, burnout, somatic symptoms, depression, and anxiety. In terms of some of the major occupational stress theories, the outcome measures used appear to t into the subjective domain of person environment t (Edwards et al., 1998), cybernetic theory (Cummings and Cooper, 1998), and control theory (Spector, 1998). In all of these measures, based at the individual level, positive effects were found. Objective outcome measures reported (van der Klink et al., 2001) included muscle tension, electromyography, catecholamine levels, cholesterol levels, blood pressure monitoring, and absenteeism. All of these except absenteeism were again measured at the individual level. Absenteeism, usually measured through the employers human resource management systems, is the only organization-based measure used, and was the only measure to show no signicant effect in this meta-analytical study (van der Klink et al., 2001). Studies such as that by Murphy and Sorenson (1988), however, that showed positive effects for absenteeism would have been excluded from the meta-analysis as the subjects were not assessed for pre-existing stress. Despite the ndings of van der Klink et al. (2001) it would be premature to assume that stress management interventions have little effect on absenteeism. Murphys (1996), review would tend to reinforce this view. Giga et al. (2003) reviewed the literature from 1990 to 2001 selecting for review only those papers that reached at least three stars according to Murphys (1996) quality evaluation schema (i.e. at least involved an intervention and had some formal evaluation but not necessarily a control group, were drawn from a normal working population, and had a sample size of at least thirty). As the authors intended to review from a UK perspective, only those studies carried out in the UK were included in this review. De Frank and Coopers (1987) typology (individual, individual/organizational, or organizational) was used to classify the studies. This classication approximately equates to Quick et al.s (1997) three level classication in that individual-based interventions correspond to Quicks secondary interventions while individual/ organizational and organizational interventions correspond to primary interventions. Some interventions such as employee assistance programs included under individual interventions may be better classied as tertiary by Quicks schema. The majority of interventions reviewed (. 80 percent) were carried out at the individual level though the authors noted a more recent shift in reported interventions toward organizational level or primary interventions. No rm conclusion was reached concerning the relative effectiveness of the different intervention types though it appeared that organizational and individual/organizational interventions were more likely to lead to improvement in

organizational performance than individual level programs. Individual level programs were, however, often associated with improvements in mental and emotional well-being. van der Klink et al. (2001), van der Hek and Plomp (1997) and De Frank and Cooper (1987) all suggested that stress management interventions can yield positive results in both subjective and objective parameters at the individual level, although long-term effectiveness has not been clearly demonstrated. There is some indication that repeated brief individual-focused interventions may enhance outcomes in the long-term (Barkham and Shapiro, 1990), but there is no clear evidence for the effectiveness of long-term organization-focused interventions at either the individual or organizational level (Briner and Reynolds, 1999), Indeed Briner and Reynolds suggested that authors in the eld tend to an enthusiasm that is not supported by available evidence. Whether stress management interventions in general can yield outcomes that result in clear positive effects at the organizational level has not, as yet, been shown. These ndings are summarized in Table I. Stress management interventions in terms of eustress and distress Primary stress management interventions The main assumption underlying a primary SMI is that the most effective method of reducing stress is to remove or reduce the stressors present in the organizational environment (Cooper et al., 2001). This underlying assumption does not sit comfortably with the idea that responses to environmental stressors may be eustressful or distressful or a combination of both (Le Fevre et al., 2003). This lack of denition of stress as distress or eustress is reected in the underlying assumptions driving primary SMIs, that stressors (assumed to be distress) in the environment can be removed or reduced, and that such removal or reduction will result in reduced stress in organizational members. Primary SMIs often include some kind of organizational stress audit stage in which organizational members are canvassed for their opinions and ideas on what may constitute a stress or stressor for them in their organization and job. We believe this approach is fundamentally weakened, however, as no attention seems to be paid to the individuals interpretation of organizational stressors as distress or eustress, but only to distressful interpretations. In summary, primary SMIs make use of averaged responses from organization members to ascertain the aspects of the organization, and the jobs within it, that are regarded as distressful by organization members. They then engage in various forms of organizational change and restructuring in an attempt to remove or reduce the number and severity of organizational stressors. Where organizational change is used as a means of stress management there is an inherent paradox, as negative emotions such as anxiety, fear, and stress itself are so often regarded as being associated with change that organization behaviour texts (Kreitner and Kinicki, 2004; Schermerhorn, 2003) often treat them in the same chapter. Where organizational change is used as a means of stressor removal one might wonder whether the process of change introduces as much or more stress than is being relieved. Perhaps this may go some way to explain the relatively low effectiveness reported for primary SMIs (van der Klink et al., 2001), and represent an area for further research.

Stress management interventions 555

JMP 21,6

556

Authors Primary interventions Interventions emphasizing changing organizational, job, and role variables Interventions emphasizing organization wide physical tness programs Secondary interventions a Meditation Philosophy of life Behavioral modication Literature review and opinion

Newman and Beehr (1979) Literature review and opinion

De Frank and Cooper (1987)

van der Hek and Plomp(1997)

Table I. Summary of main ndings from stress management intervention reviews Type of study Interventions studied Main ndings Speculative or opinion pieces without empirical support but suggesting positive outcomes for their recommended approaches Speculative or opinion pieces without empirical support but suggesting positive outcomes for their recommended approaches Speculative or opinion pieces suggesting positive outcomes with one evaluation study showing positive outcomes Speculative papers recommending this approach but without empirical evidence Speculative and opinion pieces recommending this approach but without empirical evidence No ndings reported for primary interventions Literature review and opinion Primary interventions No primary interventions were reviewed as none were . . . serious, controlled attempts to evaluate the efcacy of programs . . . (p. 6) Secondary interventions Relaxation Biofeedback Exercise Cognitive coping Meditation Time management Employee assistance programs Primary interventions Organization wide OD intervention A total of 18 studies were evaluated all showing some degree of signicant positive effect. Control groups were often not present, however, and follow-up was brief to nonexistent Burnout, group properties, and staff turnover all improved but no control group used, only change over time was assessed (continued )

Authors

Type of study

Interventions studied

Main ndings

van der Klink et al. (2001) Meta analytical study

Organization wide stress management training Reduced malpractice claims in medical practice against matched control group, but this study includes a secondary intervention component as well Support groups Improvement in group evaluation but no control groups, no follow-up, and a high drop out rate in this study Combined support groups and stress SIT effective in reducing self-report stress, no inoculation training (SIT) reported effect for group support, no control group and SIT, again has secondary intervention components Secondary interventions Relaxation Muscle tension and self-report anxiety reduced vs placebo control group Individual psychotherapy Depression symptoms, anxiety, and self esteem improved but no control group Rational emotive therapy Anxiety, depression symptoms, irrational beliefs, weariness, and assertiveness measures showed improvement, one trial controlled one uncontrolled Experiential group training Improvement in personal accomplishment and emotional exhaustion scales of the Maslach Burnout Inventory, no control group Multimodal approaches All showed positive results for measures such as burnout, anxiety, depression and stress but, no control groups Primary interventions 5 organization-focused trials assessed Effect size overall d 0.08 and is not statistically signicant, total number of participants 1,463b,c (continued )

Stress management interventions 557

Table I.

JMP 21,6

558

Authors Secondary interventions 18 cognitive behavioral trials assessed

Giga et al. (2003)

Notes: asome papers reviewed in Newman and Beehr (1979), refer to coping responses rather than formal interventions. These have been omitted; bThe medical malpractice study of Jones et al. (1988), contributed the largest effect and this study includes a signicant secondary intervention component; c d refers to Cohens d Source: Cohen (1988)

Table I. Type of study Interventions studied Main ndings Literature review and opinion Effect size overall d 0.68 p , 0.05, total number of participants 858 17 relaxation trials assessed Effect size overall d 0.35 p , 0.05, total number of participants 982 8 multimodal programs assessed Effect size overall d 0.51 P , 0.05, total number of participants 470 The cognitive behavioral approaches had a signicantly greater effect size than the relaxation approaches but there was no signicant difference between relaxation and the multi modal programs 7 individual focused programs 3 trials had control groups, four had no control group Three of four EAP programs showed improvements in individual measures of anxiety, depression, self esteem or work satisfaction, one showed no effect on absenteeism or health 7 individual/organizational focused programs Six of seven programs either showed positive changes in such variables as health, depression, attitude, and self-esteem at the individual level or differences in awareness of stress management and coping 2 organizational focused programs One program showed positive effects on job satisfaction and one showed that increasing workload had a negative impact on both speed and accuracy of working

Secondary SMIs The main assumption underlying secondary SMIs is that the complete removal of stressors inherent in the job and organization is unlikely to be possible, so it is wise to focus on the organizational members reactions to and ways of coping with those stressors (Cooper et al., 2001). These assumptions t far more comfortably with the idea that individuals may react positively or negatively to environmental pressures and demands depending on their interpretation of them and the methods they adopt to cope. Somatic approaches aim to modify the individuals physiological response to environmental stressors by reducing sympathetic activation in response to stressors perceived as negative, threatening, or harmful (distress). Cognitive approaches have a broader aim, intending to modify both the individuals appraisal of the nature of the stressor as distressful or eustressful, and their psychological reaction to that stressor whether it be ultimately perceived as distressful or eustressful. This idea is, to some extent, echoed in the oft-repeated management mantra that threats should be seen as opportunities. The former represents distressful appraisal, the latter eustressful. One might assume that the combination of somatic and cognitive approaches in multimodal secondary SMIs should be able to capitalise on the strengths of both approaches to have the greatest stress management potential. This conclusion was not supported, however, by van der Klink et al. (2001). Their meta-analytical study of 48 trials showed that cognitive interventions were signicantly more effective than somatic techniques, while multimodal approaches were not signicantly different from somatic approaches. Again, research that directly compares these different approaches may help to clarify these issues. Cognitive techniques such as visualization, meditation, and afrmation may be used to modify the individuals perceptions of both the quantum and quality of stressors. This places the individual in the role of prime actor in the stress process. It is at this perceptual and interpretive act that primary cognitive SMIs take their major aim. The major relationships between primary and secondary SMIs and the constructs of eustress and distress are summarised in Table II. Discussion The current emphasis in the occupational SMI literature toward primary interventions may be driven to some extent, perhaps, by efforts to reduce potential legal liability for harm caused by stress. In some cases it may also be driven by legislation and governmental guidelines based on the philosophy of risk minimisation espoused by occupational safety and health practice (Health and Safety Unit, 1998 as cited in New Zealand Government, 2003, Sutherland and Cooper, 2000). In itself it is no bad thing to make management take responsible action to reduce the stressful nature of the workplace, and any concerted move to reduce the potential harm of employment must be applauded. As alluded to earlier though, primary SMIs inevitably involve organizational change and this has itself been identied as a potential source of occupational stress (Sutherland and Cooper, 2000). Further, the initial stages of well-designed primary SMIs involve some form of organizational stress audit to ascertain those aspects of the job and organization that are seen as stressors by the organizations members. There is strong evidence that individual-based stress management interventions, especially those based on cognitive skill training can

Stress management interventions 559

JMP 21,6

560

Intervention type The OD intervention showed positive outcomes for burnout, group properties, and staff turnover

Primary interventions Active organization development

Support groups

Comprehensive multimodal

Secondary interventions Somatic Relaxation Biofeedback Exercise

Table II. Summary of intervention types in relation to eustress and distress Brief summary of ndings Relationship to eustress and distress Primary SMIs give primacy to the principle that stress is the response to stressors in the environment but do not appear to take into account any differentiation of stress into distress or eustress. The timing, source, and control over stressors are all legitimate targets of primary SMIs but perception of desirability is not addressed and the individuals interpretation of stressors is not directly addressed in primary SMIs There were no signicant positive outcomes from the support group approach The comprehensive approach, which included both primary and secondary techniques (Jones et al., 1988), showed improvement in the target variable of medical malpractice claims. The overall effect size from meta analysis (van der Klink et al., 2001) for primary interventions was small and not signicant A majority of trials showed positive effects in the most frequently measured physiological variables. Positive effects were also shown for psychological variables such as anxiety and stress. The overall effect size from meta analysis (van der Klink et al., 2001) was small but signicant Somatic approaches do not differentiate between distress and eustress. Their emphasis is on reducing the sympathetic activation from distress. The timing, source, and desirability of the stressor are also not addressed by somatic approaches though perceived control over the stressor may be enhanced by the ability to alter physiological responses (continued )

Intervention type A majority of trials show some positive effect over a wide range of variables. The overall effect size from meta analysis (van der Klink et al., 2001) was moderate and signicant and signicantly greater than somatic interventions

Brief summary of ndings

Relationship to eustress and distress

Cognitive Meditation/visualization Psychotherapy Cognitive behavioral

Multi modal Combinations of the above approaches

A majority of trials show some positive effects over a wide range of variables. The overall effect size from meta analysis (van der Klink et al., 2001) was moderate and signicant, but not signicantly different from somatic interventions

As generally described cognitive approaches do not directly acknowledge the existence of eustress and distress as aspects of the stress construct, however, cognitive approaches do seek to alter the perception and interpretation of stressors in terms of desirability, control, and source. Realization of the internal (self) origin rather than external (other) origin of much stress is a common aspect of cognitive approaches. Given that cognitive approaches work primarily with perception and interpretation, perceptions of timing may also be addressed Multi modal approaches do not differentiate between eustress and distress. Although the combination of cognitive and somatic approaches should allow more effective treatment of perceptions of control than cognitive approaches alone this is not born out by empirical evidence

Stress management interventions 561

Table II.

JMP 21,6

562

produce positive effects for the individuals so trained. It seems premature, therefore, to conduct such a stress audit prior to training organization members in effective stress management techniques. Such training would enable them to not only better identify those aspects of the organizational milieu that remain distressful after training, but also to provide some protection or resilience against the rigours of organizational change to come. As Semmer (2003) has suggested one should not pit primary and secondary approaches one against the other. What we are suggesting here, therefore, is that there may be advantage in employing individual focused, secondary, approaches as a rst step in interventions designed to reduce organizational job stress at the individual level, rather than as complements to an initial organizational-based approach. If eustress as originally dened by Selye (1956, 1964, 1987) and more recently Le Fevre et al. (2003) is a real entity, then one would predict that individual focused cognitive approaches to stress management, with their emphasis on the individuals perceptions and interpretations of stressors, would produce the greatest positive effects in trials of SMIs. This appears to be in agreement with the arguments presented here and the results summarized in Table II. There appears to be very little evidence on which to base the current emphasis on organizational focused, primary SMIs and little clear information on the relative effectiveness of the various components of SMIs in general. As amply illustrated in the UK Management Standards approach (Cousins et al., 2004; MacKay et al., 2004), however, there is excellent theoretical reason to suppose that organization-based interventions should be effective at both the individual and organizational level. As Cousins et al. and MacKay et al. also show there is an increasing weight of opinion supporting this stance from both employer and employee groups. This is, indeed, a source of concern for us as stated in the beginning of this paper, that, by shifting the focus too strongly toward organizational level approaches, well-proven individual approaches may be inappropriately relegated to mere add-ons to be employed once organization-based approaches are underway. As Cooper and Dewe (2004) indicated, future research needs to address the broad area of stress management intervention, and we believe that cooperative research between academics and practitioners is needed to clarify this situation. In the meantime, however, we suggest that individual-based approaches be the rst line of defence against the potential hazards of organizational job stress, to be followed by broader organizational interventions rather than being regarded as mere supplements to organization-based interventions, at least until there is evidence that suggests otherwise.
References Barkham, M. and Shapiro, D.A. (1990), Brief psychotherapeutic interventions for job-related stress: a pilot study of prescriptive and exploratory therapy, Counseling Psychology Quarterly, Vol. 3, pp. 133-48. Benson, H. (1976), The Relaxation Response, Collins, London. Benson, H., Kotch, J.B., Crassweller, K.D. and Greenwood, M.M. (1977), Historical and clinical considerations of the relaxation response, American Scientist, Vol. 65, pp. 441-5. Briner, R.B. and Reynolds, S. (1999), The costs, benets, and limitation of organizational level stress interventions, Journal of Organizational Behaviour, Vol. 20 No. 5, p. 647.

Cannon, W.B. (1914), The emergency function of the adrenal medulla in pain and the major emotions, American Journal of Physiology, Vol. 33, pp. 356-72. Cannon, W.B. (1932), The Wisdom of the Body, Norton, New York, NY. Cavanaugh, M.A., Boswell, W.R., Roehling, M.V. and Boudreau, J.W. (2000), An empirical examination of self-reported work stress among U.S. managers, Journal of Applied Psychology, Vol. 85 No. 1, pp. 65-74. Certo, S.C. (2003), Supervision: Concepts and Skill Building, 4th ed., McGraw-Hill, New York, NY. Cohen, J. (1988), Statistical Power Analysis for the Behavioral Sciences, 2nd ed., Lawrence Erlbaum Associates, Hillsdale, NJ. Cooper, C.L. and Dewe, P. (2004), Stress: A Brief History, Blackwell, Oxford. Cooper, C.L., Dewe, P. and ODriscoll, M. (2001), Organisational Stress: A Review and Critique of Theory, Research and Applications, Sage, London. Cousins, R., MacKay, C.J., Clarke, S.D., Kelly, C., Kelly, P.J. and McCaig, R.H. (2004), Management Standards and work-related stress in the UK: practical development, Work and Stress, Vol. 18 No. 2, pp. 113-36. Cox, T. and Grifths, A. (1995), The nature and measurement of work stress: theory and practice, in Wilson, J. and Corlett, N. (Eds), The Evaluation of Human Work: A Practical Ergonomics Methodology, Taylor & Frances, London. Cox, T., Randall, R. and Grifths, A. (2002), Interventions to Control Stress at Work in Hospital Staff, The Institute of Work, Health and Organisations, Nottingham. Cox, T., Thirlaway, M., Gotts, G. and Cox, S. (1983), The nature and assessment of general well-being, Journal of Psychosomatic Research, Vol. 27, pp. 353-9. Cummings, T.G. and Cooper, C.L. (1998), A cybernetic theory of organizational stress, in Cooper, C.L. (Ed.), Theories of Organizational Stress, Oxford University Press, New York, NY, pp. 101-21. Danna, K. and Grifn, R.W. (1999), Health and well being in the workplace: a review and synthesis of the literature, Journal of Management, Vol. 25, p. 357. De Frank, R.S. and Cooper, C.L. (1987), Worksite stress management interventions: their effectiveness and conceptualisation, Journal of Managerial Psychology, Vol. 2, pp. 4-10. Edwards, J.R., Caplan, R.D. and Van Harrison, R. (1998), Person-environment t theory: conceptual foundations, empirical evidence, and directions for future research, in Cooper, C.L. (Ed.), Theories of Organizational Stress, Oxford University Press, New York, NY, pp. 28-67. Edwards, J.R. and Cooper, C.L. (1988), The impacts of positive psychological states on physical health: a review and theoretical framework, Social Science & Medicine, Vol. 27, pp. 1447-59. Fentress, D.W., Masek, B.J., Mehegan, J.E. and Benson, H. (1986), Biofeedback and relaxation response training in the treatment of pediatric migraine, Developmental Medicine & Child Neurology, Vol. 28, pp. 139-46. Giga, S.I., Noblet, A.J., Faragher, B. and Cooper, C.L. (2003), The UK perspective: a review of research on organisational stress management interventions, Australian Psychologist, Vol. 38 No. 2, pp. 158-64. Golembiewski, R.T., Hilles, R. and Daly, R. (1987), Some effects of multiple OD interventions on burnout and work site features, The Journal of Applied Behavioral Science, Vol. 23, pp. 295-313.

Stress management interventions 563

JMP 21,6

564

Hackman, J.R. and Oldham, G.R. (1976), Motivation through the design of work: test of a theory, Organizational Behavior and Human Performance, Vol. 16, pp. 250-79. Hackman, J.R. and Oldham, G.R. (1980), Work Redesign, Addison-Wesley, Reading, MA. Harris, D.V. (1970), On the brink of catastrophe, Quest Monograph, Vol. 13, pp. 33-40. Jacobson, E. (1938), Progressive Relaxation, 2nd ed., University of Chicago Press, Chicago, IL. Jones, J.W., Barge, B.N., Steffy, B.D., Fay, L.M., Kuntz, L.K. and Wuebker, L.J. (1988), Stress and medical malpractice: organizational risk assessment and intervention, Journal of Applied Psychology, Vol. 73, pp. 727-35. Kreitner, R. and Kinicki, A. (2004), Organizational Behavior, 6th ed., McGraw-Hill, New York, NY. Kutz, I., Boreysenko, J.Z. and Benson, H. (1985), Meditation and psychotherapy: a rationale for the integration of dynamic psychotherapy, the relaxation response, and mindfulness meditation, American Journal of Psychiatry, Vol. 142, pp. 1-8. Le Fevre, M., Matheny, J. and Kolt, G.S. (2003), Eustress, distress, and interpretation in occupational stress, Journal of Managerial Psychology, Vol. 18, pp. 726-44. Lussier, R.N. (2002), Human Relations in Organizations: Applications and Skill Building, 5th ed., McGraw-Hill, New York, NY. MacKay, C.J., Cousins, R., Kelly, P.J., Lee, S. and McCaig, R.H. (2004), Management standards and work-related stress in the UK: policy background and science, Work and Stress, Vol. 18 No. 2, pp. 91-112. Murphy, L.R. (1996), Stress management in work settings: a critical review of the health effects, American Journal of Health Promotion, Vol. 11 No. 2, pp. 112-35. Murphy, L.R. and Sorenson, S. (1988), Employee behaviors before and after stress management, Journal of Organizational Behavior, Vol. 9, pp. 173-82. Nelson, D.L. and Simmons, B.L. (2003), Health psychology and work stress, in Quick, J.C. and Tetrick, L.E. (Eds), Handbook of Occupational Health Psychology, American Psychological Association, Washington, DC. Newman, J.E. and Beehr, T. (1979), Personal and organizational strategies for handling job stress: a review of research and opinion, Personnel Psychology, Vol. 32, pp. 1-43. New Zealand Government (2003), Healthy Work: Managing Stress and Fatigue in the Workplace, Occupational Safety and Health Service, Department of Labour, Wellington. Quick, J.C., Quick, J.D., Nelson, D.L. and Hurrell, J.J. (1997), Preventive Stress Management in Organizations, American Psychological Association, Washington, DC. Quick, J.C. et al. (2005), Positive strengths to meet the challenges . . . hope, vigour, and self-reliance, paper presented at British Academy of Management Annual Conference 2005, Said Business School, Oxford. Quick, J.D., Quick, J.C. and Nelson, D.L. (1998), The theory of preventive stress management in organizations, in Cooper, C.L. (Ed.), Theories of Organizational Stress, Oxford University Press, New York, NY, pp. 246-68. Randall, R., Grifths, A. and Cox, T. (2005), Evaluating organizational stress-management interventions using adapted study designs, European Journal of Work and Organizational Psychology, Vol. 14 No. 1, pp. 13-41. Schermerhorn, J.R. (2003), Organizational Behavior, 8th ed., Wiley, New York, NY. Selye, H. (1936), A syndrome produced by diverse nocuous agents, Nature, Vol. 138, p. 32. Selye, H. (1956), The Stress of Life, McGraw-Hill, New York, NY.

Selye, H. (1964), From Dream to Discovery, McGraw-Hill, New York, NY. Selye, H. (1983), The stress concept: past, present, and future, in Cooper, C.L. (Ed.), Stress Research, Wiley, New York, NY, pp. 1-20. Selye, H. (1987), Stress without Distress, Transworld, London. Semmer, N.K. (2003), Job stress interventions and organisation of work, in Quick, J.C. and Tetrick, L.E. (Eds), Handbook of Occupational Health Psychology, American Psychological Association, Washington, DC. Simmons, B., Nelson, D.L. and Phillips, M. (2004), Hope at work: evidence from four studies, paper presented at British Academy of Management 2004 Conference, St Andrews. Spector, P.E. (1998), A control theory of the job stress process, in Cooper, C.L. (Ed.), Theories of Organizational Stress, Oxford University Press, New York, NY, pp. 153-69. Sutherland, V.J. and Cooper, C.L. (2000), Strategic Stress Management, Macmillan Press, Basingstoke. van der Hek, H. and Plomp, H.N. (1997), Occupational stress management programmes a practical overview of published effect studies, Occupational Medicine, Vol. 47, pp. 133-41. van der Klink, J.J.L., Blonk, R.W.B., Schene, A.H. and van Dijk, F.J.H. (2001), The benets of interventions for work-related stress, American Journal of Public Health, Vol. 91, p. 270. Corresponding author Mark Le Fevre can be contacted at: mark.lefevre@aut.ac.nz

Stress management interventions 565

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.