See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/237075031

McVicar, A., Munn-Giddings, C., Seebohm, P., 2013. Workplace Stress

Interventions Using Participatory Action Research Designs. International
Journal of Workplace Health Management...

Article · January 2013

CITATIONS READS

0 2,203

1 author:

Carol Munn-Giddings
Anglia Ruskin University
91 PUBLICATIONS 903 CITATIONS

SEE PROFILE

All content following this page was uploaded by Carol Munn-Giddings on 27 May 2014.

The user has requested enhancement of the downloaded file.

 The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1753-8351.htm

IJWHM
6,1
Workplace stress interventions
using participatory action
research designs
18 Andrew McVicar, Carol Munn-Giddings and Patience Seebohm
Faculty of Health & Social Care, Anglia Ruskin University, Chelmsford, UK

Abstract
Purpose – Complex collaborative interventions are increasingly applied for stress management but
outcomes are inconsistent. “Collaboration” is most highly developed in participatory action research
(PAR). Future research might be guided by understanding features integral to successful PAR designs.
The purpose of this paper is to present a review of PAR studies which had predominantly positive
outcomes, in order to identify features of their designs.
Design/methodology/approach – In total, 48 collaborative intervention studies (1982-2010)
were identified, and filtered according to positive outcomes (improved working environment, job
performance, absenteeism, and stress levels), and PAR criteria for stakeholder engagement: 11 studies
from six countries were selected for scoping review.
Findings – Organization size and sector was not important for PAR, but the extent of uptake of an
intervention/change is and a “unit” of up to 100 employees was engaged in most of the studies reviewed.
Study aims should not be over-ambitious. Long-term involvement of “change agents” or “action groups” in
close communication with a steering group appears most effective in engaging employees over a long period
of time, ideally 12 þ months. Self-report scales dominated evaluations (21 different scales; range 1-7 per study)
but this strategy is challenged by impacts of organizational change and staff turnover on response rates.
Comparison with a non-intervention group appears to strengthen the evaluation, but PAR also provides an
opportunity to implement an innovative strategy sensitive to the workplace situation. PAR provides scope to
engage managers as participants. The participatory process was least effective where this was unsuccessful.
Research limitations/implications – PAR has high potential for the engagement of management, and
identification of a rigorous evaluation strategy, that would facilitate the efficacy of collaborative designs.
Originality/value – Insights are provided into characteristics of highly collaborative, and
demonstrably effective, PAR designs.
Keywords Organizational change, Stress, Wellness interventions, Research methods, Workplace wellness
Paper type Research paper

The scale of the problem of workplace stress is well recognized: for example, in the UK
around 13.5 million working days were lost in 2007/2008 as a consequence of stress,
depression or anxiety (Health and Safety Executive (HSE), 2009a). Individual-focused
approaches continue to dominate UK strategy for stress management despite growing
evidence that systems approaches, involving collaborative action to change the
workplace environment, may offer a more beneficial means of addressing the impacts of
job stress for both individuals and organizations. With the systems approach, translation
of research findings into actions is a significant barrier, and empowering staff and their
participation is a key factor (e.g. Coffey et al., 2009). Participatory action research (PAR), a
form of action research that explicitly has a high degree of participant engagement in
International Journal of Workplace part or all of the research process, specifically embraces the principle of empowerment
Health Management
Vol. 6 No. 1, 2013
and collective action, and this paper presents a review of PAR studies which had
pp. 18-37 predominantly positive outcomes, in order to identify features of their designs.
r Emerald Group Publishing Limited
1753-8351
Definitions of workplace stress reflect the position of the individual within
DOI 10.1108/17538351311312303 a demanding environment. For example, the UK HSE (2009b) define it as “the process
that arises where work demands of various types and combinations exceed the Workplace stress
person’s capacity and capability to cope”. The emphasis on the individual’s appraisal interventions
of potential stressors, and their capacity to cope, challenges attempts to understand
issues for the wider employee population, illustrated, for example, in the shifting
perceptions apparent over time when nurses were asked to rank the main sources of
their stress (McVicar, 2003). Evaluating categories or dimensions of work, rather than
specific sources of stress, reduces situational specificity (Van Veldhoven et al., 2005) 19
and, in that context, the UK HSE has put forward a national strategy for a management
“standards” approach to stress assessment[1] based on six job dimensions (Mackay
et al., 2004): job demands, job control, support, relationships, role and change.
The complexity of stress assessment within the workplace is clearly apparent, as is the
emphasis on work “environment”. A further complication is the lack of independence
of the dimensions as a cause of stress. Thus, vulnerability to job-related stress seems to
relate to a complex mix of social gradient, job control, effort-reward imbalance, social
support and health behaviours (Lunt et al., 2007), whilst combinations appear to have
a stronger effect than sources of stress alone (Rick et al., 2002).
Intuitively, the pluralistic basis to workplace stress suggests that stress management is
most likely to be successful for the majority of employees if intervention comprehensively
addresses sources of stress within the work environment, and their interactions (Bond,
2004; Michie and Williams, 2003). Organizational-focused interventions seem logical to
achieve this but, in contrast, strategies predominantly have been individual-focused with
the aim of raising the individual’s resilience (Murphy and Sauter, 2003; Caulfield et al.,
2004; Lamontagne et al., 2007). In the UK, an influential review by Seymour and Grove
(2005) concluded that only individual-focused therapy is backed up by strong evidence as
an effective solution, but gave greatest weight to quantitative studies that applied
randomized controlled trials, a design associated with a “medical” model of stress that has
dominated recent stress research (Kelloway et al., 2008), and which arguably is less
relevant to the uniqueness of causative factors in any given workplace.
Hurrell and Murphy (1996) suggest that workplace interventions fall into three
categories: “primary” (aimed at eliminating the cause of stress), “secondary” (aimed at
enabling employees to manage their stress) and “tertiary” (aimed at aiding employees
requiring rehabilitation). Organizational-focused interventions fall into the “primary”
category as they engage the wider employee population, and have the potential to deal
more effectively with the complexity of the workplace itself, as part of a preventative
strategy. However, complex, collaborative, psychosocial interventions within the
workplace, that is, collective actions aimed at improving the psychological well-being
of individuals through changing aspects of the working environment, have been
reported to have inconsistent outcomes (Seymour and Grove, 2005), though some
reviews have also suggested that when effective they are more likely to have a positive
outcome than individual-focused ones (Caulfield et al., 2004; Hatinen et al., 2007).
Recent trends suggest that psychosocial interventions are increasingly being
applied (Lamontagne et al., 2007), and can be of benefit to both individuals and
the organization provided that there is a “moderate/high” level of collaborative
involvement ( Jordan et al., 2003; Lamontagne et al., 2007), although a background of
widespread organizational reforms is detrimental to the process (Haukka et al., 2010).
Contributory factors to collaborative involvement include a presence of a social climate
of learning, the provision of opportunities for multi-level participation in the design of
interventions, awareness of tacit behaviours that may be undermining, and definition of
roles and responsibilities (Nytro et al., 2000). Various models of collaboration are available,
IJWHM and in their review Kelloway et al. (2008) identify a need for further research on
6,1 interventions appropriate to the complexity of organizations. Studies that have applied
highly collaborative approaches and have achieved demonstrable positive outcomes
therefore provide an opportunity to identify features that may inform future research.
Aims and objectives
20 Participatory approaches to changing the workplace clearly must operate within the
organizational context in which the interventions take place. Whereas participatory
approaches can have successful outcomes (Jordan et al., 2003; Lamontagne et al., 2007),
the workplace introduces a range of potential variables in studying the efficacy
of intervention, for example in terms of the size of organization, level of engagement
of employees and managers in the process, or duration of the intervention, that will
potentially influence the likelihood of a successful outcome. The aim of the present
study therefore was to elucidate the characteristics of successful collaborative
interventions through a review of empirical studies that specifically, and
successfully, applied a PAR approach to stress management. PAR has the greatest
potential for high-level collaboration as it is founded on engagement by most or all
stakeholders in the workplace, and so goes beyond simple consultation to integrally
involve those facing the situation in the research process itself (Winter and Munn-
Giddings, 2001). The approach has distinguishable organizational characteristics
through its focus on system development, co-learning processes, participation, democratic
process, empowerment and a search for a balance between research and intervention
(Schurman, 1996). The uniqueness of any given workplace means that even restricting the
review to PAR will still identify variations in study design. Accordingly a non-systematic
scoping review (Anderson et al., 2008) of relevant issues was conducted to identify “the
current state of understanding (and) the sorts of things we know and do not know [y]
within policy and practice contexts” (Davis et al., 2009, p. 10) and to extract “the essence of
a diverse body of evidence” (Davis et al., 2009, p. 13).

Methods
Scoping reviews generate a breadth of coverage of literature, with subsequent elimination by
post hoc criteria (Arksey and O’Malley, 2005). In line with this, interventional PAR studies
which predominantly had demonstrably positive impacts on workplace stress outcomes (see
“Sample” section below) were identified that highlighted a range of designs with sufficient
breadth to identify their most pertinent features. Several databases of literature related to
mental or physical health, psychology and organizational research, were searched to provide
the broad basis to the review. Those searched were Business Source Premier, Emerald, Assia,
PubMed, RDN, Cochrane, Expanded Academic ASAP (Gale), National Electronic Library
for Health (NHS), CINAHL, PSychFIRST, NMAP, Medline, PsychInfo and IBSS. The
search was conducted using the terms “stress” and “participatory” within the title and
abstract, for papers published between 1982 and 2010 Other papers were extracted from
the BOHRF review of stress interventions (Seymour and Grove, 2005).
Sample
A total of 57 papers were identified as potentially having apparent (but variable)
relevance to the aim of the review:
. Four papers related to PAR and safety at work/improved working conditions.
Although linked to stress they were not explicitly concerned with stress
management, and so were not considered further.
 . In all, 23 papers related to PAR outside the workplace. Whilst specifically Workplace stress
related to stress research, the focus was more concerned with community interventions
development or public health than with workplace stress management, and so
the papers were rejected.
. Four studies were not empirical per se, but were either discussion papers related
to participation (two studies), or were evaluation studies which did not involve
an intervention (two studies). 21
This filtering left 26 papers related to PAR and work-related stress. An initial analysis
of the value to this study was made by one of the authors (PS), and the decision ratified
in discussion with the other two. In 13 papers either the link to stress was weak, or the
level of participant engagement was not considered sufficiently “active”, based on
the degree of shared ownership of decision making in the evaluation and intervention
processes (Munn-Giddings et al., 2008). Some were linked to the notion of a learning
organization and explored the benefits of that but not specifically in the context of
stress management; many were also outside the selected timeframe.
The remaining 13 papers had brought together members of the workforce with the
aim of reducing workplace stress and/or improving well-being. All had positive
outcomes, considered to be those evidenced as having had an impact on conditions and
processes in the workplace: better working environment, improved job performance,
improved absenteeism rate and reduced stress. Although mainly positive, some
studies also reported some negative outcomes. This might be expected in complex,
multi-factorial situations, but it can complicate evaluation of the efficacy of the
intervention (e.g. Briner and Reynolds, 1999; Lamontagne et al., 2007; Nytro et al., 2000).
In the selected studies, the positive outcomes broadly were in accordance with the
individual study’s aims, and so were considered appropriate for this review (see Table I).
Four of the papers actually referred to just two studies, but for two papers (Lavoie-
Tremblay, 2004; Lavoie-Tremblay et al., 2005) the double reporting each provided
particular insights and so both were retained, i.e. the final sample was 12 research
papers (see Table II) related to 11 empirical studies. Three studies were from the USA,
two from the UK, two from Canada, two from Norway, one from Germany and one from
Sweden. All brought together members of the workforce with the aim of changing
work conditions, reducing stress and/or improving well-being.
Participants in PAR studies additionally gain various generic benefits from the PAR
process itself (Winter and Munn-Giddings, 2001), and several studies in the sample
identified that the introduction of new skills and new collaborative ways of working
brought such benefits even before any intervention had started. Examples are:
. gaining problem-solving skills which can then be integrated into all aspects of
their life (Mikkelsen et al., 2000);
. new learning inspired and supported participants in their personal development
(Fricke, 1983); and
. the process provided a platform for participants’ to be heard and respected, and
taken seriously (Fricke, 1983; Munn-Giddings et al., 2005).
Analysis
A standardized approach to analysis is recommended for scoping reviews (Davis et al.,
2009) and the papers were analysed by applying a framework based on that put
forward by Arksey and O’Malley (2005), which contrasted the process of scoping and
 6,1

22

Table I.
IJWHM

Study aims and outcomes

Study aims Study outcomes: positive Study outcomes: negative or other

Improved employee Improved management style (Lavoie-Tremblay et al., 2005) Difficult to evaluate long-term effects due to
participation More employee participation/decision making (Fricke, 1983; uncertainties about future work practice
Heaney et al., 1993; Gold, 1998; Boivie et al., 2003; (Boivie et al., 2003)
Munn-Giddings et al., 2005) Conflict with, and resistance to change by, managers
(Fricke, 1983)
Participants felt “at risk” if collective action was
undertaken (Gold, 1998)
Reluctance to approach managers (Boivie et al., 2003)
Some managers/chief executive declining to take part
(Mikkelsen and Gundersen, 2003; Munn-Giddings
et al., 2005)
Improved job performance/ Performance/productivity improved (Fricke, 1983; Bond and Job satisfaction unchanged (Fricke, 1983; Bond and
productivity Bunce, 2001) Bunce, 2001)
Reward-effort balance improved (Lavoie-Tremblay et al., 2005) Demand/job strain unchanged (Lavoie-Tremblay
Learning/social environment and working conditions improved et al., 2005)
(Fricke, 1983; Golembiewski et al., 1987; Mikkelsen and Support unchanged or reduced (Heaney et al., 1993;
Gundersen, 2003; Lavoie-Tremblay et al., 2005) Lavoie-Tremblay et al., 2005)
Increased worker autonomy/responsibility (Boivie et al., 2003;
Mikkelsen and Gundersen, 2003)
Reduced absenteeism Rates decreased (Golembiewski et al., 1987; Bond and Bunce, Turnover intentions unchanged (Halbesleben et al., 2006)
and turnover 2001; Lavoie-Tremblay et al., 2005)
Reduced stress/improved Improved mental health (Bond and Bunce, 2001) Effects limited (Mikkelsen and Gundersen, 2003)
mental health Reduced stress (Mikkelsen and Gundersen, 2003;
Lavoie-Tremblay, 2004)
Reduced exhaustion and disengagement/burnout
(Golembiewski et al., 1987; Halbesleben et al., 2006)
Integration and addressing of health risk aspects into model
for job change (Boivie et al., 2003)

Note: See also Table III for details of surveys when used
Sector, type of organization,
study author and country Target participants
of location Design (unit size engaged in study) Participatory process

Statutory sector
Community health care Quasi-experimental intervention in All supervisors and all Introduction seminar: employees and managers to
Mikkelsen et al., 2000 healthcare institutions on workers’ employees in two healthcare identify need for improvement in their work unit.
Norway job stress and other job institutions. Randomly Included plenary, individual and small group sessions,
characteristics allocated to an intervention and facilitated by consultant
and to a control group Five work groups: nine meetings of each to increase
detail on problems, suggest solutions and report to
steering group. Development
of an action plan for sustaining the process
of change
Steering group: met weekly throughout. Implement
organizational changes
Government office Quasi-experiment to test whether a 121 people of whom 97 were Steering Committee (PAR group): five meetings
Bond and Bunce, 2001 work reorganization intervention administrative employees: facilitated by in-house occupational psychologist, over a
UK can improve stress-related 49 allocated to control three month period
outcomes by increasing civil group, and 48 to PAR Wider consultation: all employees offered opportunities
servants’ job control intervention group to discuss and influence the proposed strategies
Government An exploratory study to reduce Six administrators/VDU Individual interviews/non-participant observation.
Boivie et al., 2003 health risk (from excessive operators Assessment of the problem
Sweden workload) within a National Tax Case analysis using scenarios. Evaluate organizational
Board office in the analysis, design and job requirements
and evaluation of Two design workshops, one including users.
a prototype software Identification of design solutions, design of prototype
software
Test sites: testing of prototype designs at employees
workstations
Care unit (long-term care) Evaluation study of the 60 healthcare workers in Commitment contract: meetings with each key group of
Lavoie-Tremblay, 2004; effectiveness of a participatory direct staff/management/unions/HR – commitment contract
Lavoie-Tremblay organizational intervention to contact with patients: 14 agreed. Work team identified of staff, unions,
et al., 2005 improve the psychosocial work nurses, management, researchers
Canada 25 nursing assistants, 12 Three focus groups (17 participants in total) to identify
(continued)

and participants
Settings, designs
interventions
Workplace stress

23

Table II.
 6,1

24

Table II.
IJWHM

Sector, type of organization,

study author and country Target participants
of location Design (unit size engaged in study) Participatory process

environment support work constraints

in a long-term care unit staff, nine from a multi- Work team meetings: met six times, each time for one
disciplinary team day, over seven months. Prioritize constraints, develop
action plans
Implementation teams: two or three members in each,
set up to implement other action plans
Regular meetings between unit head and staff
maintained communication, for project evaluation and
follow-up
Health and social services Exploratory study to develop a Health Trust: practitioners Introduction meetings: between
Munn-Giddings et al., mental well-being strategy in a and managers. Social lead researcher and senior
2005UK healthcare organization and a services: middle/senior managers to agree access
social services organization (SSD) managers Participatory workshops: five
sessions to identify problem areas,
and suggest action to address them
Fire service Exploratory study to develop a 95 staff – firefighters, Initial meetings: key stakeholders
Halbesleben et al., 2006 strategy to reduce job stress management and admin Meetings across the organization: to establish
USA amongst fire fighters staff expectations, trust and rapport
Problem and opportunity identification phase:
informally with all workforce, then formally with
2  top managers. Discussions with staff on each shift at
each fire station, and with managers to develop solutions
and agree timeframe: development of a multisource
training tool. Two management sessions held on this
Renewal of mission: discussion sessions between top
management and all staff
Commercial organizations
Screw factory Prospective interventional study 47 “shop floor” workers Initial department meeting: workers, and
Fricke, 1983 (a “demonstration experiment”) representatives of the work council, management and
Germany to change working conditions experts e.g. safety engineers, to define the problems
for employees in a factory Twice-yearly seminars for four years: designing action
plans, reviewing progress
Project groups (four to six members): implementation
(continued)
Sector, type of organization,
study author and country Target participants
of location Design (unit size engaged in study) Participatory process

Corporate Human Exploratory study for an 31 members of staff in the Individually: all HR staff asked to list the three best
Resources department organizational development to human resources unit things about their dept and the three concerns they most
Golembiewski et al., 1987 reduce burnout in an HR wanted to change
USA department Group feedback and disclosure: findings shared and
presented to the vice
president of HR
Four interest groups of volunteers to gather additional
info and make recommendations for change to career
progression plans
Manufacturing plant Quasi-experimental evaluation Whole workforce of 1,100 Stress and Wellness Committee: 26 workforce volunteers
Heaney et al., of the influence of a participatory people to develop a “local stress theory”; four problem areas
1993 intervention on the impact of considered by four subcommittees. Action plans
USA labour-management relations on presented to joint meetings of management and unions.
workplace stress Implement, evaluate and modify
Post office Quasi-experimental evaluation Total 89 people As for Mikkelsen et al. (2000)
Mikkelsen and of a participatory organizational 40 people in intervention
Gundersen, 2003 intervention during a period of group
Norway organizational restructuring 49 people in control group
within the postal service
“Not-for-profit” sector
Children’s welfare Exploratory study to help women 40 female workers: 30 Six focus groups of female social workers, supervisors:
Gold, 1998 doing child welfare work develop frontline workers and ten identification of positive and negative aspects of their
Canada strategies for the protection and supervisors (approximately work, effects of their job on physical and mental health,
promotion of their health half of workers employed at and how they coped/protected their health
the agencies) Follow-up meetings over several months: action
planning
interventions
Workplace stress

25

Table II.
IJWHM systematic reviewing. Regarding analysis, the emphasis was on “charting” data rather
6,1 than “data extraction”; the former relates to decisions taken as to what the “key items”
to be recorded and collated are, whereas the latter is usually related to meta-analysis
often involving statistical method. The identification of “key items” was based on
issues noted in the initial reading of the selected papers, and in subsequent discussions
within the research team, and on previous published discussions (Jordan et al., 2003;
26 Lamontagne et al., 2007) of factors related to collaborative practice. Features of the
studies were collated according to:
(1) Context:
. Type of organization involved in the study: to identify if there was
any pattern to the kind of organization that might be receptive to PAR
approaches;
. Nature of stress targeted, and the participatory process: to establish
commonality or variation in the problem and approach used; and
. Methods employed: to gain insight into the means of data collection, and the
intervention strategy, across the studies.
(2) Issues arising:
. Evaluation: to establish issues for evaluation and impact;
. Collaboration issues: to establish promoting or restricting issues produced
by the collaborative process. Participatory approaches are recognized as
being potentially challenging as so many staff groups may be involved.

Findings: study designs, and issues

Context
Types of organization/employees involved in the study/initiation of the study. Six of the
11 studies took place in the statutory sector (i.e. provision is funded by the State
nationally or at local level), four in the independent (commercial) sector and one in the
not-for-profit sector (see Table I). The variety of settings suggests that the economic
sector and nature of business are not important factors when selecting a site for
participative research into workplace stress. This is also supported by a review by
Kompier et al. (1998) of interventional studies which, though not necessarily having a
level of participation normally associated with PAR, also encouraged the participation
of workers in the intervention process: the sites of those studies were a government
office, prison, cigarette factory, oil company, scaffolding company, bricklaying,
hospital, home for the elderly, telecommunications and a home care institution.
No obvious pattern was evident in the type of employee recruited to the study,
and participants tended to reflect the nature of the organization’s activities (e.g.
by involving firemen, healthcare workers or VDU operators). Across all studies, six
engaged administrative staff, seven engaged frontline service providers or shop floor
workers and five engaged supervisors/managers, within the participatory process. Not
all studies provided a demographic profile of the participants but where details
were available many studies mostly involved women (e.g. 100 per cent in Gold, 1998;
94 per cent in Mikkelsen et al., 2000; 78 per cent in Lavoie-Tremblay, 2004,
Lavoie-Tremblay et al., 2005; 63 per cent in Mikkelsen and Gundersen, 2003), but in the
studies of Bond and Bunce (2001) and Halbesleben et al. (2006) 63 and 98 per cent
of participants were male, respectively. Ethnicity was even less reported but in some Workplace stress
studies it was identified that most participants were “white” (Heaney et al., 1993; interventions
Munn-Giddings et al., 2005).
Most studies were initiated either by a direct approach to the organization from an
academic institution, or by the organization management contacting those institutions.
The study by Fricke (1983) is unusual in this sample as it applied a “bottom-up”
approach to initiating the study: researchers and employee representatives at the 27
works council came to an agreement about doing the project before seeking
management approval. That project achieved a number of its aims, in particular
reduction in factory noise, establishment of a worker project group and a model for
continuing worker education, but wider structural aims, such as break times and
a process for decision making regarding workplace change, were not realized.
The author suggests that established hierarchical decision-making processes, and
inequality of power in the company were critical factors, the implication being that
engagement of management and the perceived benefits of change, are more important
than how an interventional project is initiated.
Nature of stress targeted and the participatory process. Common features of the
research sites were evidence of past or continuing re-organization and re-structuring,
and concern with excessive workplace stress. Regarding the latter, most studies refer
to this in the context of mental health problems arising from stress, though some
(Gold, 1998; Lavoie-Tremblay, 2004; Lavoie-Tremblay et al., 2005) also refer to physical
health problems. The nature of the mental health problem is frequently described as
“burnout” (Golembiewski et al., 1987; Halbesleben et al., 2006; Lavoie-Tremblay, 2004;
Lavoie-Tremblay et al., 2005; Munn-Giddings et al., 2005), a term used to describe
emotional exhaustion and disengagement from work, and predictive of absenteeism
and staff turnover (Leiter and Maslach, 2009).
The studies involved interventions targeted at features of the work environment or
culture, and broad sources of workplace stress are suggested. One emergent theme
from the review is the importance of limiting study ambitions to fit available time and
resources, and so reduce the risk of disappointing results, although this potentially
may challenge the credibility of the PAR approach if it restricts the extent of
participation (Mikkelsen and Gundersen, 2003). In this respect there are some
prominent factors across the studies:
. poor relationships between management and workers;
. inadequate communication;
. lack of support; and
. perceived lack of control by employees of their workload.
For most, the focus of the research was on specific dimensions, such as job (workload)
control (Bond and Bunce, 2001); indeed those authors suggested that an intervention is
likely to be most effective if specific issues are targeted. Conversely, the aims of some
were less specific, for example “to improve the psychosocial work environment”
(Lavoie-Tremblay et al., 2005).
In accordance with PAR principles, a feature of all of the studies was the application
of collaborative processes involving stakeholders to identify the “local” issues. The hub
of subsequent interventions generally was a collaborative forum where problems and
solutions were discussed by participating stakeholders, including workshops, work
or project groups, steering groups and various meetings/focus group activities
IJWHM (see Table I). These usually drew on the findings from earlier phases of the study and
6,1 informed the next stage, such as a wider consultation phase or implementation phase.
Forums were mainly action or project groups of some sort that met periodically (five to
ten meetings) over the duration of the project, a few months to five years depending
upon the study. Formulation of action plans, and implementation plans, are common
features, with dissemination of change throughout either part of the organization or
28 all of it, mediated by a Steering Group of stakeholders including managers.
Collaboration enables stakeholders with different kinds of expertise to solve
problems together – perhaps also with input from external experts, such as
occupational psychologists – and a feature was the amount of time that had to be
committed in the early phase of the study to agree access, share expectations, agree a
timeframe and to formally establish a commitment from the workforce. For example, in
the study of Bond and Bunce (2001), the conduct of the initial baseline assessment of
the workplace situation did not take place for two months after the authors had first
presented the project to managers and union representatives. In terms of the scale of
the project, the most extensive was a study in a manufacturing plant (Heaney et al.,
1993) which conducted its first survey in 1986 and the last in 1991, covering a
workforce of 1,100 people. Direct engagement and participation therefore had to be
restricted, and PAR activities were led by a “Stress and Wellness Committee” of
26 people representative of the management, union and “shop floor” workforce. In
contrast, the study by Boivie et al. (2003) voluntarily involved just six administrators/
VDU operators in all PAR activities. Both of these studies had been selected for
this review because they met the criterion of having demonstrably positive outcomes,
vis-à-vis the study objectives, and so the difference in number of participants therefore
did not appear to be a factor in their positive outcomes. The scale of the study by
Heaney et al. (1993) was considerably larger than that of the other studies within this
review, and in general most projects had up to 100 participants. What appeared to be
more important was the level of uptake of the intervention by employees within
the organization or unit; this has implications for the design of the study in the context
of its specific aims and scope.

Methods employed
The research methods utilized by the sample studies broadly can be categorized into
quantitative surveys and various qualitative tools.
Surveys were used to identify, or to start identifying, the problems to be addressed,
and to provide an evaluation by self-reported assessment of the workplace before,
during and after an intervention. A total of 32 scales were used across the studies,
either selected from those already available (18 different scales across all studies;
see Table III) or designed specifically for the purpose of the study (three studies), while
three studies also referred to staff or management records for data on absenteeism or
turnover. The most commonly applied scale was the Maslach Burnout Inventory (three
studies). Whilst the tools reflected the focus and aims of the particular study, it is clear
that generic tools on job content/involvement and work environment were common.
Their applicability to the “local” issues is not explicit, and the range used in this sample
also probably highlights the choices researchers and participants have of tools that
purport to be valid for risk assessment (see Rick et al., 2001 for review). The number used
per study (minimum ¼ 1; maximum ¼ 7), and the length of time required for completion
(up to 2 h; Mikkelsen and Gundersen, 2003), were variable. The response rate varied
widely (Table III) and, whilst it is difficult to identify features that determined a good
Author Survey tools Application and response rates

Golembiewski Maslach Burnout Inventory Five surveys: day 45, 200, 295, 425 and 575. No information on response rate
et al., 1987 Job Involvement Scale but it appears initially to be 100%. The fifth survey was nine months after the
Work Environment Scales intervention
Turnover rate from staff records
Heaney et al., Self-developed scales on: involvement in project; Two surveys
1993 labour management relations; participation in Baseline: response rate ¼ 66%
decision making; social support Five years after intervention: response rate ¼ 62%
NIMH Center for Epidemiologic Studies
Depression Scale
Mikkelsen et al., Job Stress Questionnaire Three surveys
2000 The Health Inventory Baseline: response rate 73 and 49%, intervention and control group,
Demands-Control Dimensions respectively
Work Apgar Questionnaire One week after intervention: response rate ¼ 96 and 40% intervention and
Role Harmony Scale control group, respectively
Learning Climate Questionnaire One year after baseline: response rate: 20 and 79% intervention and control
Multifactor Leadership Questionnaire group, respectively
Bond and Bunce, Occupational Stress Indicator Two surveys
2001 Job Content Questionnaire Baseline: response rate ¼ 80%
Self-developed rating scale on performance Nine months after intervention. Attrition rate by second survey: 56% PAR
Sickness absence records group and 53% control group

(continued)

and response rates

Survey tools, application
interventions
Workplace stress

Table III.
29
 6,1

30

Table III.
IJWHM

Author Survey tools Application and response rates

Mikkelsen and Job Stress Questionnaire Three surveys, with final survey one year after pre-test
Gundersen, 2003 Subjective Health Complaints Control group. Response rate 92, 65 then 41%
Job Demands and Job Control Intervention group: response rate 92, 87 then 57%
Job Satisfaction Scale
Learning Climate Questionnaire
Lavoie-Tremblay Job Content Questionnaire Two surveys
et al., 2005 Effort-Reward Questionnaire Baseline: response rate ¼ 98%
Psychiatric Symptom Index One year after end of intervention: response rate ¼ 80%
Management records
Munn-Giddings Maslach Burnout Inventory One survey at each location
et al., 2005 Response rate 23.6% in social services;
Response rate ¼ 12% in Health Trust
Halbesleben Maslach Burnout Inventory Two surveys
et al., 2006 Turnover intention Baseline: response rate 87%
Minnesota Satisfaction Questionnaire One year following completion of the project: response rate not given
Self-developed tool on experience of changes

Notes: Eight studies only included. Three either did not apply survey tools (Gold, 1998; Boivie et al., 2003: qualitative methods only), or did not specify
(Fricke, 1983; feedback appeared to be sought through seminars)
response rate, other than some organizational stability (below), responses appear to be Workplace stress
highest where a workforce of 60-100 employees had a high level of awareness of, and interventions
engagement with, the project (Halbesleben et al., 2006; Lavoie-Tremblay et al., 2005).
Qualitative methods were prominent in the participatory process and included the
use of interviews, focus groups and observations to explore the problem area, usually
after the initial survey had been completed. In some studies they were also used to
develop solutions, and two (Gold, 1998; Boivie et al., 2003) applied qualitative methods 31
in the evaluation of outcomes. Group discussions usually provided the source of data
and at the same time a method of developing the project tasks. When observational
methods were used they were helpful in building trusting relationships between
researcher and participant, and increasing the researchers’ understanding of the
problems (Fricke, 1983; Halbesleben et al., 2006).

Evaluation
Studies were longitudinal in design, providing a before-after evaluation of the
intervention. Although they reported significant outcomes, the prospective nature of
the research designs often introduced concerns in providing an appropriate evaluative
process because of the detrimental effects of staff turnover, management change and
organization-wide policy changes. The study by Heaney et al. (1993) illustrates the
impact on the study sample profile of downsizing followed by expansion: 55 per cent of
staff at the first (pre-intervention) survey were no longer employed at the time of the
second (post-intervention) survey, whilst 59 per cent of staff employed at the time of
the second survey were not employees when the first survey was undertaken.
The volatility of staffing levels and/or turnover may have contributed to some
apparent contradictions in the evaluation, illustrated by the studies of Bond and Bunce
(2001) and Lavoie-Tremblay (2004) in which findings included positive outcomes for
reward-effort balance and job performance, yet these were not associated with
significantly improved job satisfaction or reduced job strain.
Five studies endeavoured to strengthen the evaluation process by comparing
outcomes with those of a non-intervention group, but the extent to which these might
be considered as conventional “controls” is questionable. Efforts to ensure a good
match for the groups are especially apparent in the Bond and Bunce (2001) study, but
despite having addressed many of the difficulties inherent in applying randomized
selection strategies the authors note that there may still have been important, unknown
differences between them, for example in relation to roles and responsibilities. Heaney
et al. (1993) argued that the problems of involving a control group in a PAR study are
unavoidable and insurmountable:
. the “control” workforce would require similar needs, values and context, which
can only be discovered as the PAR process unfolds;
. even if a good match were found at the beginning of the study, there is no control
of events over time (e.g. management turnover, layoffs, re-structuring); and
. different units within the same organization are not “sealed” from each other,
and organization-wide interventions introduced alongside those of the research
will affect both.
In two studies (Mikkelsen et al., 2000; Mikkelsen and Gundersen, 2003) the approach
is more pragmatic. For example, practical reasons meant that Mikkelsen et al. (2000)
applied a random allocation procedure for the control group but not for the
IJWHM intervention group. In the studies by Golembiewski et al. (1987) and Heaney et al. (1993)
6,1 little effort was made to match the comparator groups other than to involve staff
with related roles. Studies that sought to provide comparators therefore can at best
be considered as utilizing a “quasi-experimental” design, though they appear to
strengthen confidence in the evaluation.
An evaluation problem is also introduced by the timescale required for impacts
32 to be discernible (Boivie et al., 2003). This seemed less problematic for short-term
pragmatic outcomes, such as a support group and self-management pack for returnees
to work for NHS staff (Munn-Giddings et al., 2005), or concrete improvements such as
noise reduction and improved ergonomic aspects of machines in a factory (Fricke,
1983). However, Bond and Bunce (2001) identify that wider, far-reaching changes may
have to be introduced over a longer term to have an impact. Halbesleben et al. (2006)
considered that 9-12 months after interventions were introduced may have been
too soon for the impact to be fully developed in their study, and for it clearly to
be sustainable, while a duration of more than 12 months was considered by
Lavoie-Tremblay and Colleagues (Lavoie-Tremblay 2004; Lavoie-Tremblay et al., 2005))
as likely to give a more accurate picture of any eventual benefits. To sustain the
intervention, the latter study identified a volunteer committee of “change agents”, set-
up towards the end of the project in order to maintain the action and reflection beyond
the study period, but with lengthy duration there remains a risk that these may still be
lost through staff attrition, with potential effect on the project (e.g. Golembiewski et al.,
1987; Heaney et al., 1993; Lavoie-Tremblay et al., 2005; Munn-Giddings et al., 2005).
In general, most studies lasted at least 18 months.
In principle, the engagement of participants within PAR designs across every
stage of the research process, including design of an evaluation strategy, ought to be
responsive to such issues provided that the participatory process is retained throughout
the duration of the intervention period. The frequent meetings of “action group” and
steering group, whilst also limiting the study ambitions, seem likely to have been factors
in the rigour of the evaluation in the sample studies, and hence in their apparent success.

Collaboration issues
Improved stress outcomes following organizational change are more likely to be
lasting if they engender a positive attitude to collaborative learning among managers
and workers (Lunt et al., 2007), but this appears to require changes in attitude and
commitment by organization leaders. There is evidence that the Halbesleben et al.
(2006) study may have achieved this level of change as some managers recognized that
interventions were only the start of a process and they planned to continue their
(leadership) training after the project came to an end, but for some studies it was
evident that fully engaging management was difficult. Conflict and resistance
generated by hierarchy and an inequality of power are the most difficult obstacles
to overcome if collaborative forums are to be effective (Fricke, 1983). The application
of a democratic approach is a central principle of PAR, and in studies reporting issues
related to employee participation, it is striking that a lack of engagement of managers,
or distrust of management, was often apparent (see Table II). In particular:
. for Munn-Giddings et al. (2005), a resistance by participants to disclosure of
mental health problems, and pre-existing organizational tensions, appears to
have led to the apparent rejection of the PAR group’s right to explore and
evaluate management input;
 . Golembiewski et al. (1987) noted that employees felt they took a risk in Workplace stress
presenting the problems to an unrelenting manager; interventions
. some employees in the study by Lavoie-Tremblay (2004) failed to adopt changes
introduced in the workplace by colleagues because of a distrust of (managerial)
motives for the changes; and
. women in the Gold (1998) study did not feel able to collectively challenge 33
their workplace managers as they felt that their positions might be at risk
as a consequence.
Engaging the workforce at every level, and especially at senior management level, is a
time-intensive process that has to be maintained. The study by Halbesleben et al. (2006)
suggested that employer commitment may be more forthcoming if they recognize a
clear economic or employer-focused benefit, such as impacts on productivity, sickness,
absence rates and accidents. Senior managers not only have to be aware of the
interdependency of workplace and individual factors in employee well-being but also
recognize the potential benefits of stress prevention to employers. PAR can be effective
here, for example, Heaney et al. (1993) noted that adopting this approach meant that
different conceptions of stress and stressors held by unions and management could be
shared prior to deciding on interventions.

Summary and conclusions

There is evidence that PAR designs can have predominantly positive outcomes for
major stress indicators, including job performance and productivity, absenteeism and
mental health, but there is inconsistency with some studies reporting either partial or
little overall impact (Jordan et al., 2003; Lamontagne et al., 2007) or even negative
outcomes when associated with unconnected widespread organizational reforms
(Haukka et al., 2010). This review of studies that report successful interventions sought
to identify what might be learned from them, with the aim of clarifying a PAR design
that might increase the likelihood of improving stress management in the workplace.
The economic sector and nature of business are not important factors when selecting a
site for participative research into workplace stress, but an intervention is more likely
to be successful if the scale of the study is not too ambitious, so that employees are
more fully engaged with the process. Smaller organizations (or units within larger
organizations) with up to 100 participants seem most appropriate for this approach.
Specific but important problems should be targeted, and evaluation of local issues,
rather than presumptive ones, is an essential part of the assessment process.
A recurring theme from this review is the need for the researcher and organization
to invest considerable time, effort and resources to ensure fully participative
collaboration between stakeholder representatives. The skill of the facilitator/
researcher will be a critical factor as far-reaching interventions over a long period of
time especially require commitment by organization leaders, whilst the identification
and support of key agents or similar to sustain the process of change would also seem
to be beneficial.
Factors most likely to impact negatively on outcomes relate to staff turnover, to
management/employee relationships or to the evaluation strategy:
. The timescale for an intervention to have a confirmative impact is difficult to
ascertain. Outcomes may be discernible either in the short or longer term (12 þ
months), but for longer-term aims the design also should take into account
IJWHM potential staff turnover, management change and organizational policy during
6,1 the period. Non-intervention groups ought to be utilized where possible, despite
difficulties in group matching, as they help to strengthen evidence.
. If participation is to be fully successful then it is essential that senior managers
recognize that resolving factors pertaining to employment, rather than
individuals, are the most appropriate means of managing workplace stress,
34 and that they, and unions, should be engaged in the participatory process and be
supportive of initiatives.
. Self-reported surveys featured in 8 of the 11 studies. A striking feature of the
studies reviewed is the breadth of tools utilized, and it is clear that different tools
might be selected, or new tools developed, even though study aims are similar.
The rationale for choice should be made clear as even well-known psychosocial
measures may have unconfirmed validity and reliability across settings. The
tool of choice in the “local” context may be an issue for further research.
Data collection methods will also have to capture evidence despite variable
response rates due to staff turnover, and a wider application of qualitative
methods (used to effect in two of the studies; Gold, 1998; Boivie et al., 2003) might
also be useful, to ascertain viewpoints of ex- and new employees for comparison
with baseline data.
In conclusion, utilizing PAR shows promise as a vehicle for primary, psychosocial
interventions for workplace stress. Designs are complex, but PAR provides the means
of resolving critical issues of stakeholder engagement, including managers, and of
ensuring a reliable evaluation strategy. However, whilst the focus is organizational,
perhaps the greatest lever to engage participants will always be the perceived benefit
to the individual asked to take part, whether that person is an employer, employee or
union representative, and this will be a challenge for researchers while building and
maintaining rapport and trust.

Note
1. The resultant self-assessment tool for employees is available at www.hse.gov.uk/stress/
standards/pdfs/indicatortool.pdf (accessed 21 January 2010).

References
Anderson, S., Allen, P., Peckham, S. and Goodwin, N. (2008), “Asking the right questions: scoping
studies in the commissioning of research on the organization and delivery of health
services”, Health Research and Policy Systems, Vol. 6 No. 7, pp. 1-12.
Arksey, H. and O’Malley, L. (2005), “Scoping studies: towards a methodological framework”,
International Journal of Social Research Methodology, Vol. 8 No. 1, pp. 19-32.
Boivie, I., Blomkvist, S., Persson, J. and Aborg, C. (2003), “Addressing user’s health issues in
software development – an exploratory study”, Behaviour and Information Technology,
Vol. 22 No. 6, pp. 411-20.
Bond, F.W. (2004), “Getting the balance right: the need for a comprehensive approach to
occupational health”, Work and Stress, Vol. 18 No. 2, pp. 146-8.
Bond, F.W. and Bunce, D. (2001), “Job control mediates change in a work reorganization
intervention for stress reduction”, Journal of Occupational Health Psychology, Vol. 6 No. 4,
pp. 290-302.
Briner, R.B. and Reynolds, S. (1999), “The costs, benefits, and limitations of organizational Workplace stress
level stress interventions”, Journal of Organizational Behaviour, Vol. 20 No. 5,
pp. 647-64. interventions
Caulfield, N., Chang, D., Dollard, M.F. and Elshaug, C. (2004), “A review of occupational stress
interventions in Australia”, International Journal of Stress Management, Vol. 11 No. 2,
pp. 149-66.
Coffey, M., Dugdill, L. and Tattersall, A. (2009), “Designing a stress management intervention in 35
social services”, International Journal of Workplace health Management, Vol. 2 No. 2,
pp. 98-114.
Davis, K., Drey, N. and Gould, D. (2009), “What are scoping studies? A review of the nursing
literature”, International Journal of Nursing Studies, Vol. 46 No. 10.
Fricke, W. (1983), “Participatory research and the enhancement of workers’ innovative
qualifications”, Journal of Occupational Behaviour, Vol. 4, pp. 73-87.
Gold, N. (1998), “Using participatory research to help promote the physical and mental
health of female social workers in child welfare”, Child Welfare, Vol. 77 No. 6,
pp. 701-25.
Golembiewski, R.T., Hilles, R. and Daly, R. (1987), “Some effects of multiple OD interventions
on burnout and work site features”, Journal of Applied Behavioural Science, Vol. 23 No. 3,
pp. 295-313.
Halbesleben, J.R.B., Osburn, H.K. and Mumford, M.D. (2006), “Action research as a burnout
intervention. Reducing burnout in the federal fire service”, Journal of Applied Behavioural
Science, Vol. 42 No. 2, pp. 244-66.
Hatinen, M., Kinnunen, U., Pekkonen, M. and Kalimo, R. (2007), “Comparing two burnout
interventions: perceived job control mediates diseases in burnout”, International Journal
of Stress Management, Vol. 14 No. 3, pp. 227-48.
Haukka, E., Pehkonen, I., Leino-Arjas, P., Viikari_Juntura, E., Takala, E.P., Malmivaara, A. and
Hopsu, L. et al. (2010), “Effect of a participatory ergonomics intervention on psychosocial
factors at work in a randomized controlled trial”, Occupational and Environmental Health,
Vol. 67 No. 3, pp. 170-7.
Health and Safety Executive (HSE) (2009a), “Health and Safety statistics 2007/08 Labour Force
Survey”, available at: www.hse.gov.uk/statistics/overall/hssh0708.pdf (accessed 21
January 2010).
Health and Safety Executive (HSE) (2009b), “Work-related stress”, available at: www.hse.gov.uk/
stress/furtheradvice/wrs.htm (accessed 21 January 2010).
Heaney, C.A., Israel, B.A., Schurman, S.J., Baker, E.A., House, J.S. and Hugentobler, M. (1993),
“Industrial relations, worksite stress reduction, and employee well-being: a participatory
action research investigation”, Journal of Organizational Behaviour, Vol. 14 No. 5,
pp. 495-510.
Hurrell, J.J. and Murphy, L.R. (1996), “Occupational stress intervention”, American Journal of
Industrial Medicine, Vol. 29 No. 4, pp. 338-41.
Jordan, J., Gurr, E., Tinlione, G., Giga, S., Faragher, B. and Cooper, C. (2003), Beacons of Excellence
in Stress Prevention. HSE Research Report 133, HSE Books, Sudbury, available at:
www.hse.gov.uk/research/rrpdf/rr133.pdf (accessed 21 January 2010).
Kelloway, E.K., Teed, M. and Kelley, E. (2008), “The psychosocial environment: towards an
agenda for research”, International Journal of Workplace Health Management, Vol. 1 No. 1,
pp. 50-64.
Kompier, M.A.J., Geurts, S.A.E., Grundemann, R.W.M., Vink, P. and Smulders, P.G.W. (1998),
“Cases in stress prevention: the success of a participative stepwise approach”, Stress
Medicine, Vol. 14 No. 3, pp. 155-68.
IJWHM Lamontagne, A.D., Keegel, T., Louie, A.M., Ostry, A. and Landsbergis, P.A. (2007),
“A systematic review of the job-stress intervention evaluation literature, 1990-2005”,
6,1 International Journal of Environmental and Occupational Health, Vol. 13 No. 3,
pp. 268-80.
Lavoie-Tremblay, M. (2004), “Creating a healthy workplace. A participatory organizational
intervention”, Journal of Nursing Administration, Vol. 34 No. 10, pp. 469-74.
36 Lavoie-Tremblay, M., Bourbonnais, R., Viens, C., Vezina, M., Durand, P.J. and Rochette, L. (2005),
“Improving the psychosocial work environment”, Journal of Advanced Nursing, Vol. 49
No. 6, pp. 655-64.
Leiter, M.P. and Maslach, C. (2009), “Nurse turnover: the mediating role of burnout”, Journal of
Nursing Management, Vol. 17 No. 3, pp. 331-9.
Lunt, J., Fox, D., Bowen, J., Higgins, G., Crozier, S. and Carter, L. (2007), “Applying the
biopsychosocial approach to managing risks of contemporary occupational health
conditions: scoping review. HSE HSL/2007/24”, available at: www.hse.gov.uk/research/
hsl_pdf/2007/hsl0724.pdf (accessed 21 January 2010).
McVicar, A.J. (2003), “Workplace stress in nursing: a literature review”, Journal of Advanced
Nursing, Vol. 44 No. 6, pp. 633-42.
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.
Michie, S. and Williams, S. (2003), “Reducing work related psychological ill health and sickness
absence: a systematic literature review”, Occupational and Environmental Medicine, Vol. 60
No. 1, pp. 3-9.
Mikkelsen, A. and Gundersen, M. (2003), “The effect of a participatory organizational
intervention on work environment, job stress, and subjective health complaints”,
International Journal of Stress Management, Vol. 10 No. 2, pp. 91-110.
Mikkelsen, A., Saksvik, P. and Landsbergis, P. (2000), “The impact of a participatory
organizational intervention on job stress in community health care institutions”, Work and
Stress, Vol. 14 No. 2, pp. 156-70.
Munn-Giddings, C., Hart, C. and Ramon, S. (2005), “A participatory approach to the promotion
of wellbeing in the workplace: lessons from empirical research”, International Review of
Psychiatry, Vol. 17 No. 5, pp. 409-17.
Munn-Giddings, C., McVicar, A. and Smith, L. (2008), “Systematic review of the uptake and
design of action research in published nursing research, 2000-2005”, Journal of Research
in Nursing, Vol. 13 No. 6, pp. 465-77.
Murphy, L.R. and Sauter, S. (2003), “The USA perspective: current issues and trends in the
management of work stress”, Australian Psychologist, Vol. 38 No. 2, pp. 151-7.
Nytro, K., Saksvik, P.O., Mikkelsen, A., Bohle, P. and Quinlan, M. (2000), “An appraisal of key
factors in the implementation of occupational stress interventions”, Work and Stress,
Vol. 14 No. 3, pp. 213-25.
Rick, J., Briner, R.B., Daniels, K., Perryman, S. and Guppy, A. (2001), A Critical Review of
Psychosocial Hazard Measures. Contract Research Report for the Health and Safety
Executive: 356/2001, HSE Books, Sudbury, available at: www.hse.gov.uk/research/
crr_pdf/2001/crr01356.pdf (accessed 21 January 2010).
Rick, J., Thomson, L., Briner, R.B., O’Regan, A. and Daniels, K. (2002), Review of Existing
Supporting Scientific Knowledge to Underpin Standards of Good Practice for Key Work-
Related Stressors – Phase 1. Contract Research Report for the Health and Safety Executive:
024, HSE Books, Sudbury, available at: www.hse.gov.uk/research/rrpdf/rr024.pdf
(accessed 21 January 2010).
 Schurman, S.J. (1996), “Making the ‘new American workplace’ safe and healthy: a joint labor- Workplace stress
management-researcher approach”, American Journal of Industrial Medicine, Vol. 29 No. 4,
pp. 373-7. interventions
Seymour, L. and Grove, B. (2005), Workplace Interventions for People with Common Mental
Health Problems: Evidence Review and Recommendations, British Occupational Health
Research Foundation, London.
Van Veldhoven, M., Taris, T.W., de Jonge, J. and Broersen, S. (2005), “The relationship between 37
work characteristics and employee health and well-being: how much complexity do we
really need?”, International Journal of Stress Management, Vol. 12 No. 1, pp. 3-28.
Winter, R. and Munn-Giddings, C. (2001), A Handbook for Action Research in Health and Social
Care, Routledge, London.
Corresponding author
Andrew McVicar can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints

View publication stats