AMERICAN JOURNAL OF INDUSTRIAL MEDICINE 56:424–438 (2013)

Review Article

Effectiveness of Joint Health and

Safety Committees: A Realist Review

Annalee Yassi, MD, MSc, FRCPC,1
Karen Lockhart, MA,1 Mona Sykes, Cert. in OHS,2
Brad Buck, BCRSP,3 Bjorn Stime, MPH candidate,1 and Jerry M. Spiegel, MA, MSc, PhD1

Background Joint health and safety committees (JHSCs) are widely acknowledged
as important to a healthy and safe work environment. However, it is also generally
believed that having a JHSC is necessary but not sufficient; the JHSC must be
effective.
Methods A systematic review was undertaken to find empirical studies regarding the
effectiveness of JHSCs; realist review methodology was applied to determine context-
mechanism-outcome patterns. Experts from across Canada and from various sectors
and perspectives including government, employers, and unions, were brought together
to inform the synthesis.
Results Thirty-one studies met inclusion criteria. Mechanisms identified as important
determinants of JHSC effectiveness across various jurisdictions include adequate
information, education and training; appropriate committee composition; senior man-
agement commitment to JHSCs; and especially a clear mandate with a broad scope
and corresponding empowerment (through legislation and/or union presence).
Conclusions Consistent empowerment mechanisms emerge as determinants of
successful JHSCs across contexts despite few evidence-based details for best practice
implementation. Intervention research is warranted. Am. J. Ind. Med. 56:424–438,
2013. ß 2012 Wiley Periodicals, Inc.

KEY WORDS: occupational health; workplace safety; joint health and safety
committee; systematic review

INTRODUCTION

Workplace injuries and illness continue to take an

1
Global Health Research Program (GHRP), School of Population and Public Health, Univer- unacceptably high toll on the health and safety of workers
sity of British Columbia,Vancouver, British Columbia, Canada
2
BC Government and Service Employees’ Union (BCGEU), Burnaby, British Columbia,
and their families, and on the productivity of the work-
Canada force internationally [Dembe, 2001; Schulte, 2005; Putz
3
BC Public Service Agency (BCPSA), Government of British Columbia, Vancouver, British Anderson et al., 2010]. Since the nineteenth century,
Columbia, Canada
Contract grant sponsor: WorkSafeBC.
workers have formed trade unions and other organizations
Contract grant sponsor: Dr.Yassi’s Canada Research Chair. to strengthen the improvement of workplace health and
Conflict of interest: The authors declare no conflict of interest. safety and job conditions. In more recent years, joint
*Correspondence to: Dr. Annalee Yassi, Global Health Research Program (GHRP), School
of Population and Public Health, University of British Columbia, Rm 430, 2206 East Mall,
labor–management health and safety committees (JHSCs)
UBC,Vancouver, British ColumbiaV6T1Z3, Canada. E-mail: annalee.yassi@ubc.ca have been enshrined in legislation in numerous countries
to identify potential health and safety problems and bring
Accepted 24 October 2012
DOI10.1002/ajim.22143. Published online 28 November 2012 in Wiley Online Library
them to the employer’s attention [Eaton and Nocerino,
(wileyonlinelibrary.com). 2000]. A JHSC is composed of people who represent the

ß 2012 Wiley Periodicals,Inc.


workers and the employers. Studies have demonstrated
that effective representation, proper committee operation,
and management attention to committee recommendations
are more likely in a unionized working environment
[Shannon et al., 1992; O’Grady, 2000]. We conducted this
study to ascertain the importance of these various features
as determinants of success of a JHSC. Our goal was to
identify gaps in knowledge such that an informed research
agenda can be formulated.
MATERIALS AND METHODS
For the systematic review, we chose a realist review
design, as realistic evaluation is grounded on the belief
that social interventions, including workplace improve-
ment initiatives, are complex, and that the way they bring
about change is influenced by, dependent on, and in turn
influences their contexts [Pawson and Tilley, 1997].
According to realist evaluation, an intervention (I) in a
context (C) triggers a mechanism (M) which generates an
outcome (O). Mechanisms thus explain what happens
when a particular context plays host to an intervention. A
complex social intervention can therefore yield different
outcomes in different settings due to contextual factors at
the level of the individual, institution and wider infrastruc-
tural setting [Pawson et al., 2005]. CMO configuring is a
heuristic used to generate causative explanations pertain-
ing to the evidence. The process consists of aligning the
context, mechanism, and outcome of a particular point of
investigation to uncover the array of factors that may
contribute to process or end results [Jagosh et al., 2011].
Context pertains to the ‘‘backdrop’’ to the programs;
mechanism pertains to what ‘‘turns on’’ in the minds of
program participants and stakeholders that make them
want to participate, or invest in, programs and outcomes
are defined as either intermediate or final [Jagosh et al.,
2011].
Ethics approval from an institutional review board
was deemed unnecessary for this project as it consisted
largely of a literature review. Our review started with a
systematic search in PubMed, Embase, and Google Schol-
ar. Key terms were: JHSC, workplace health committee
and occupational health and safety committee; these key
terms were searched along with descriptors such as activi-
ties, involvement, outcomes, membership, training, educa-
tion, union, management, mandate, and information. A
further selection was made based on the presence (or ab-
sence) of empirical data about JHSCs—excluding studies
that provided insufficient data. See Table I for inclusion/
exclusion criteria. We then reviewed the reference list
from the articles that met the inclusion criteria and added
relevant articles (see Fig. 1). The review criteria were rela-
tively liberal, noting that if strict criteria (i.e., minimum
sample size, random samples, prospective studies, etc.)
Systematic Review of Committees 425
TABLE I. Criteria for Relevance Assessment
Does the article/document focus on JHSCs?
Isitpublished in English or French?
Isthe date ofpublication between1975 and September 2010?
Does the article/documentreport on quantitative orqualitative data?
Does the article/documentmake conclusionsregardingthe effectiveness ofjoint
committees?
Isitpublishedin a peer-reviewedscientificjournalorbeenthesubjectofpeer-review
by an external body?
Yes, inclusion; No, exclusion.
were applied there would be even fewer studies to consid-
er in formulating the need for stronger research design
would be lost. Following review of each article, we then
characterized the CMO to try to discern a pattern explain-
ing the JHSCs’ effectiveness.
Consistent with guidelines for conducting systematic
reviews [WHO, 2008], we involved experts in shaping the
review and interpreting our findings. This component took
on particular importance as the literature was lacking in
details explaining JHSC effectiveness. We therefore hosted
a day-long workshop with 25 key health and safety
experts from across Canada to present our review findings
as well as guide the synthesis of the literature, informed
by expert experience. Invitees were chosen to provide a
diversity of sectors, different provinces plus the Federal
jurisdiction, and reflecting different perspectives (employ-
er, union, government, workers compensation boards). All
had a minimum of 15 years of direct experience working
in health and safety and were national leaders in this field
(defined as having worked or advised on health and safety
in a minimum of two provinces and had been instrumental
in developing policy and/or training materials for JHSCs).
The workshop consisted of a plenary, followed by small
group facilitated discussions, each focusing on a different
theme identified from the literature and collectively re-or-
ganized in the plenary session. The goal of the workshop
FIGURE 1. Context mechanism outcome.
426 Yassi et al.
was to present our results to the experts and to discuss
strategies for organizing findings as they related to our
CMO framework, as well as to ensure nothing had been
overlooked.
RESULTS
The systematic search yielded 280 article references,
of which 28 met the inclusion criteria. Searching the refer-
ence lists of the identified articles yielded three additional
empirical studies resulting in 31 articles meeting our in-
clusion criteria. Of the studies, 80% (n ¼ 25) were cross-
sectional studies, using qualitative and/or quantitative
methods to examine JHSCs. These are summarized in
Table II by their purpose; methods and context; and results
(including mechanism and outcome). The remainder
(n ¼ 6) was review studies (Table III).
Results of Realist Evaluation
Several studies in Table II are particularly noteworthy
for the CMO analysis; the unionization and legislative en-
vironment are crucial contextual factors. We summarize
results by country below as we distill key messages from
some of the studies included in this review. All of the
studies are discussed in detail in Tables II and III but only
some of the key messages are pulled out for discussion in
the text below.
Canada
Since the retail sector in Ontario, Canada, had not ini-
tially been subject to the requirement to establish joint
committees, it was possible to study differences in injury
performance between the manufacturing sector and the re-
tail sector with the implementation of the Occupational
Health and Safety Act, as well as differences in injury
performance related to whether the JHSCs was established
prior to or after the statutory legal requirement [Lewchuk
et al., 1996]. This study found that where workplaces
implemented JHSCs before they were mandated or imme-
diately upon the government indicating they were likely to
be mandated, JHSCs improved the workplace’s health and
safety record, as measured by time-loss injuries. However,
where workplaces created JHSCs only reluctantly, some-
times after a period when they were in contravention of
existing legislation, the formation of a committee had no
clear effect [Lewchuk et al., 1996]. Examining the same
sectors, Shannon et al. [1996] showed that firms with se-
nior management commitment to health and safety, higher
worker participation, and better communication and labor
relations had lower lost time injury rates. In striving to
better understand the relationship between workplace fac-
tors and injury rates, Shannon et al. [1997] found that
lower injury rates were associated with workplaces, which
espoused values of empowerment of the workforce (in
general matters); delegation of safety activities; and an ac-
tive role (in health and safety) of top management.
Another study, conducted in Quebec and Ontario,
Canada, found that JHSCs with equal numbers of union
and management representatives had both lower injury
rates and demonstrated enhanced problem-solving exper-
tise compared with workplaces without such committees.
The study further showed that the capacity of JHSCs to
function effectively was correlated with the amount of
training and information available to JHSC members and
that the frequency of meetings, record maintenance, num-
ber of inspections, depth of management participation, and
formulation of recommendations was inversely correlated
with injuries as well as regulatory compliance orders
[Tuohy and Simard, 1993]. Havlovic and McShane
[1997], studying the attributes of forest product mills of
British Columbia, Canada, contribute to a safer work envi-
ronment and lower injury rates, found that mills where
management received the JHSC meeting minutes had low-
er rates of serious injuries. Geldart et al. [2010] found that
the number of worker (vs. management) members on the
JHSC was higher for lower injury sites and that workpla-
ces with lower lost-time claims had JHSCs with more ex-
ecutive functions and greater worker involvement. The
mechanisms identified as associated with better outcome
are thus consistent with those identified in Canada.
A review study concluded with four lessons gleaned
from the Canadian experience [Parsons, 2001]: govern-
ment-legislated mandatory JHSCs are needed as are a
clear set of rights for JHSCs; unions must bargain over
health and safety matters; strong government enforcement
is needed; and committees must be given direct
responsibility.
United States
Studies of the association between presence of com-
mittees and injury rates in the United States have been
mixed and note both the need for additional details
about the committees and concerns with temporality. For
example, as shown in Table II, Eaton and Nocerino
[2000], comparing elements of JHSCs to injury and illness
reports in the public sector in New Jersey, found that
committee scope and training were related to perceived
committee effectiveness, and that higher worker involve-
ment was associated with fewer reported injuries and
illnesses.
A study of 107 Ohio companies found that manage-
ment leadership and worker involvement were protective
(OR ¼ 0.36) against safety violation [Akbar-Khanzadeh
and Wagner, 2001]. Boden et al. in a survey of 127 Mas-
sachusetts manufacturing companies and in-depth
TABLE II. Cross Sectional Studies
Article Purpose of study
Advisory Council on Occupational Survey to ascertain how well committees were
Health and Occupational functioning and what they needed to improve.
Safety [1986]
Akbar-Khanzadeh and Study to determine characteristics associated
Wagner [2001] with OSHA safety violations.
Boden et al. [1984] Study to determine ifexistence of and
characteristics associated with JHSCs in
manufacturing firms in Massachusetts
affectedthenumberofOHSAseriouscitations.
Bryce and Manga [1985] To clarify through expert opinion how to ade-
quately measure effectiveness of JHSCs and
to use these measures to determine
effectiveness ofa sample of JHSCs in Alberta
and Saskatchewan.
Methods (design, methods and context esp.
reunionization)
Designed to measurethe effectiveness of the Ontario internal
responsibility system after its first 5 years ofoperations,sur-
veyed,through mailed questionnaires,both worker and man-
agement JHSC
members in a random sample ofover 3,000 Ontario
workplaces.
Data were collected from107 routinely collected and archived
case files on 25 indicators of the quality ofhealth and safety
programsinsmall high-hazardcompanies in Ohio union/non-
union is not one of the 25 quality indicators used.
127 Massachusetts manufacturing companies were studied.
Data from14 interviewsin a sample of13 firms was combined
with OSHAinspection data.These are all joint union^man-
agement committees.
Data in Saskatchewan and Alberta,Canada, ofjoint committees Key factors: (1) commitment from employers; (2) the mandate of the
that had been establishedby Ministerial orderover previous
8 months was tabulated into statistical summaries.Whether
or not a worksite is unionized is listed as an external factor
influencingeffectiveness,butnodataonthisisavailableinthe Data presented more just as a statistical summaryofdifferent mea-
datasets summarized.
Results/findings (including mechanism [M] and outcome [O])
Where jointcommittees were mandatory,93% of workplaces had
established them,including 88% ofnon-union sites and 96% of
unionizedestablishments.Foundthatjoint committeeswerefunc-
tioningwellin58%ofworkplaces,adequatelyin30%ofworkplaces
and poorly in12%. Survey result highlighted the difficulties ofjoint
committee members
principally labor members
in obtaining
what they regarded as the necessary information to perform their
tasks; 28% of worker members and 9% ofmanagementmembers
reportednot having adequate information.Overall,it concluded
that ‘‘compliance with specific provisions of the act is poor among
workplaceswithJHSCs:onlysome22%ofworkplaceswithJHSCs
appear to be in full compliance with the act’’ (M: information
provision; O: JHSC considered functioningwell).
Thenumberof serious violations and regulatory violations was
significantly correlated to all 25 indicators.In particular,this study
found that management leadership and worker involvement were
protective(OR ¼ 0.32,0.33)forseriousOSHAsafety violation(M:
managementleadership andworkerinvolvement;O:fewerserious
safety violations).
While the existence ofa JHSC had no effect on either thenumberof
OSHA complaints or workplace hazards,those with a JHSC that
was seen as‘‘effective’’ had fewercomplaints and fewer serious
hazards.Objective JHSC attributes may be less important to its
successthanthecommitmentofmanagementandlabortosolving
workplace safety problems (M: management and laboron JHSCs
being committed to solvingworkplace safety problems; O: fewer
complaints andhazards).
committee
authority and responsibility; (3) committee exis-
tence
mandatory committee vs.volunteer; and (4) committee’s
involvement in the worker’s right to know.
sures of JOHSC functioning.Little to no actual rigorous analysis of
the data is carried out. Authors feel that the data give empirical
Systematic Review of Committees
indication that JHSCs are effective and efficient instruments for
reducingoccupationalinjuriesanddiseases(M:fourkeyfactors;O:
reducedhazards in workplaces).
427
(Continued )
TABLE II. (Continued )
Article Purpose of study
Cooke and Gautschi [1981] Study to determine the association between the
existence of JHSC and time -loss injuries in
Maine.
Eaton and Nocerino [2000] Study to determine relationshipbetween details
ofcommittee structure and perceptions of
committee effectiveness, as well as relation-
ship to injury and illness incidents.
Frick and Walters [1998] Qualitative and quantitative analysis of theregio-
nal safety representative (RSR) system in
small and medium enterprises in Sweden.
There are currently1,500 RSRs in Sweden
covering700,000workers.Thissystemallows
trade unions in workplaces with fewer than 50
employeesto appointsafetyrepsfromoutside
the workplace,even when the workplace has
in-house reps.
Geldart et al. [2005] Study to ascertain workplace level health and
safety policies, practices and attitudes, and to
ascertainrelationshipwithlost-timefrequency
rate (LTFR).
Hall et al. [2006] To explain why and how someworker represen-
tatives have been able to make significant
Methods (design, methods and context esp.
reunionization)
Sixyears ofdatawerecollectedin113manufacturingcompanies Thepresence of a JHSCin theworkplacewas associatedwith a small
in Maine,controllingfor thenumberofemployeesintheplants
andtheimpactsofbusinesscycles.Nomentionismadeabout
unionized vs.non-unionized workplaces.
427NewJerseypublic-sectorworksites’laborandmanagement Consensuswas that injury and illness are not the best measure of
representatives were surveyed andlinked with state-col-
lectedinjuryandillnessincidents.Unionizationofworkplaces
is not mentioned.
Questionnaire in1994 sent to all regional safety reps in the two RSRs enjoyedgoodcooperation between themselves and employers
largesttradeunionconfederationsinSweden.Interviewswith
regional safety reps to gain more information on disagree-
ments between employers and unions and how these were
handled/resolved.
Data were providedby theWorkplace Safety and Insurance
Board (WSIB) in Ontario,Canada, on 435 workplaces from an
original pool of 770 manufacturingworkplaces, supplemen-
ted with an additional 279 workplaces to increase the power
of this study.This was a mail survey,with telephone
interviews.
Investigated forms/styles of worker representation in 27 differ- ‘‘Knowledge activists’’reported a higherdegree ofeffectiveness than
entsmall to medium-sized (50^500 employees) Ontario,
Canada, autoparts plants and four larger assembly plants.
428
Results/findings (including mechanism [M] and outcome [O])
decreaseintime-lossclaimsduetoinjury infirmswith300ormore
employees (M: presence of JHSC [jointly administeredplant-spe-
cific programs] and OSHAcitations; O: reduced injuryrates,
Yassi et al.
decrease in time-loss claims in large firms).
committee effectiveness. Also consensusthat JHSCs alone are
notenough
employersneedtodedicateattentionandresources
to health and safety in general and to committeesin particular.The
study concluded that there shouldbe state sponsored guidelines
about the minimum requirements ofcommittees rather than just
the existence ofcommittees (M: committee functioning, adequate
resourcing ofcommittees and attention to H&S; O: decreased inju-
ries and illnesses,perceptions ofcommittee effectiveness).
in most workplaces.
80% said their main function was‘‘activatinghealth and safety’’

when asked to statewhat they said meant,many said registering
and solvingproblems while another large proportion said it
revolved around education ofin-house safety reps and managers
(M:regionalsafetyrepsadvising,training,andaugmenting/raising
the profile ofhealth and safety; O: more awareness ofhealth and
safety matters).
Lower LTFRs were associated with recording of H&S measures,
greater involvement of workers in decision making, andbetter
managerial attitudes concerning the importance of H&S.The
compositionoftheJHSC,differedaccordingtoinjuryrate:themean
numberof worker (vs. management) members on the JHSC was
higher forlowerinjury sites, and themean numberof worker mem-
bers attending JHSC meetings was also higher for lower injury
worksites.WorkplaceswithlowerLTFRbenefitedbyhavingJHSCs
with more executive functions and greater worker involvement.
Managerial and worker participation in H&S initiatives charac-
terizes safer workplaces associated with lower injuryrates (M:
characteristics of JHSC,workplace health and safety policies,
practices, and attitudes; O: fewer lost time injuries).
other worker representatives in that they were more likely to iden-
tify significant workplace hazards and morelikely to convince
(Continued )
TABLE II. (Continued )

Methods (design, methods and context esp.

Article Purpose of study reunionization) Results/findings (including mechanism [M] and outcome [O])
improvements to working conditions and Open-ended interviews with unionized worker health and management (andworkers) thatparticularproblems were serious
others have been less successful. safety representatives were conducted and summarized. enoughtowarrantexpendituresormajorchangesinworkprocess,
Most were union co-chairs ofthe JHSCintheir workplace and orthatchangeswouldpayoffintermsofproductivity,costsavings,
the designated certifiedrepresentatives. or worker satisfaction (M: approaches/models of worker H&S
representation as depictedby [a] the strength ofpolitical orienta-
tionand[b]theknowledgeactivistcapacity;O:moreactiveefforts
to act on unhealthy/unsafe working conditions,increased scale
and importance ofissues addressedby JHSC,increased focus on
underlying causes,increased willingness topresent managers
with solutions, and more effective implementation ofchange).
Havlovic and Study to ascertain relationship between JHSCs Questionnaires to management (n ¼ 106) and employee LongerserviceonJHSCsbyemployeerepresentativeswaspositively
McShane [1997] andinjuriesintheforestproductmills ofBritish (n ¼ 137) representatives in137 forestry mills in British related to serious injuryrates.
Columbia,Canada. Columbia,Canada. Favorable employee^management JHSC committee relationships
wererelatedtofewerseriousinjuryrates.PostingtheJHSCminutes
in the workplace was related to fewer serious incidents.
Themore effective the JHSC was perceived tobe,the lower the minor
incident rates while JHSCs that serve as decision makers and per-
ceived as more effective wererelated to seriousinjuryrates (M:
characteristics of JHSCs; O: fewer serious injuries).
Kochan et al. [1977] DescriptivestudytoascertainperceptionofJHSC Questionnaires administered to labor and management co- Three quarters ofmanagement members ofcommittees viewed
committee members regardingtheir chairs in 51manufacturingplants in NewYorkstate. themselves as being authorized to make all or most relevant deci-
empowerment. sions on health and safety.
Union counterparts had a more ambiguous perception: only half of
union members believed that the management members ofcom-
mitteescouldmakeallormostrelevantdecisions(M:functioningof
JHSCs; O: better health and safety in the workplaces).
Levesque [1995] Descriptive study focused on thetactics Interviews with labor and management representatives in 71 Most members ofjoint committees,whetheremployeror labor repre-
employed in JHSCs and their perception of the unionizedQuebecmanufacturingestablishments.Thesurvey sentatives,typically use both coercive and persuasive tactics.
tactics usedby the other party. involved direct interviews with labor and management and In only18% ofjointcommittees didboth parties confine themselves to
focused on the tactics; they employ in joint committees and persuasive tactics.
their perception of the tactics usedby the other party.Tactics Overall,64% ofmanagementrespondents and 56% oflabor respon-
were classified as either ‘‘coercive’’or dentsreliedon amixofcoerciveandpersuasivetactics.Workplace
‘‘persuasive.’’ Two questions were posedby health and safety cannot be divorced fromthe broadercontext of
Levesque: what is the incidence ofcoercive vs.persuasive conflicting interests and priorities between labor and manage-
Systematic Review of Committees

tactics and what arethe external correlates ofa propensity to ment.

use coercive tactics. Also found that the propensity to use coercive tactics correlated to
conflict over production goals; the tensions over production objec-
tivesoverlapwithsafety(M:characteristicsofJHSCs,tacticsused
429

by the committee; O: safety vs.production).

(Continued )
TABLE II. (Continued )
Article Purpose of study
Lewchuket al. [1996] Study to compare the safety performance ofindi- Three data sources in Ontario: (1) a study conducted in1991; (2) There were fewer work-related injuries and illnesses after the JHSC
vidual workplaces before and after theintro-
duction of Bill 70 in the late1970s (mandatory records; and (3) survey distributed to manufacturingwork-
committees), andbefore and after thecreation
of a JHSC at a workplace.
Morse et al. [2008] Twenty-ninecommitteesinthreeindustrysectors
(manufacturing,health care, and transporta-
tion) in Connecticut,the USA.
Nichol et al. [2009] Descriptivestudybasedonsurveyofco-chairs at
acute carehospitals in Ontario,Canada.
Pragnell [1994] Descriptive study of 2004 workplaces in Austra-
lia (762 were in New SouthWales) tofind out
about their practices with regard to commit-
tees and health and safety.In New South
Wales,45% of surveyed workplaces had
establishedcommittees.In manufacturingthe
proportion was 59%,while in recreation ser-
vices the proportion was19%.In the wholesale
and retail sector,the proportion was 36%.
Methods (design, methods and context esp. 430
reunionization) Results/findings (including mechanism [M] and outcome [O])
WCB annual data from1976 to1989 on lost-time claims was put in place. As he timing of this variable is different across
workplaces,it is less likely (than in the case of Bill 70,forexample)
places in1994 concerningthe formation and activities of thatthisvariableis simplycapturingan effect atsomepointintime
Yassi et al.
JHSCs required under the1979 legislation. that has othercauses (e.g., a mis-specified trend).
These results suggest a majordrop in injury and illness rates after a
JHSCwasformedatworkplaceswheretheJHSCwasformedprior
to1980,but an insignificant effect forcommittees formed after the
legislation came into effect (M: formation of JHSCs,especially
before beinglegislatively required to do so; O: fewer injuries).
Pilot study using administrative data (including HSC meeting Lower injuryrate companies were found to have a larger JHSC
minutes) of a stratifiedrandom sample
data drawn from a comparedwithsizeoftheemployer,andrevieweda largeraverage
unique program that requires HSCs in companies of 25 or numberof workercomplaints and suggestions.
more employees and inspects these companies to ensure Two-thirds oflow injuryrate companies reviewed safety issues from
compliance.ThusHSCscharacteristics(derivedfrommeeting prior meetings compared with only halfofhigh injury firms.
minutes and admin records) were used to compare between Three variables were found tobe correlated with high/low injury sta-
high injury rate andlow injuryrateworkplaces to better tus: numberof suggestions/complaints discussed,rate ofcom-
understand therole of HSCs. mittee size to employment, and the ratio ofmanagers/employees
oncommittees(M:presenceofcommittee,functionsofcommittee,
follow up ofcommittees; O: injuryrates).
Hospitals were identified from a list providedby OSACH,the safe JHSCs functioning in Ontario hospitals include legislative compliance
workplace association designated for the healthcare sector and availability ofresources and experts.
by theWSIB of Ontario; 378 surveys were sent toworksites Gapsidentified include a lackof JHSCmembereducation beyond cer-
deemed eligible; 220 were returned.Sixpages,41-item ques- tification training and suboptimal JHSC status and visibility within
tionnaire developedbased on provincial and federal govern- healthcare organizations (M: structure,function, andresources
ment approved JHSC education resources (questions on available to JHSCs; O: compliance with legislation).
members,information and reporting,training and education,
etc.). Surveys were sent to both management co-chairs and
employee co-chairs
39% ofpars (i.e.,both management
and workerco-chairs) were returned.
AustralianWorkplace Industrial Relations Survey (AWIRS) con- Largerworkplacesmorelikelytohavecommittees;83%ofcompanies
ducted in1989^1990.Legislation provides for the establish- with 200 or more employees had committees compared with 26%
ment of a jointcommittee ifdirectedby theWork Cover ofcompanies with 20^49 employees.
Authority oron application by a recognized trade union.The Thelikelihood ofa committee being established declined as the
Australian system is intermediate between a voluntarist proportion ofpart-time employees increased.
model and a mandatory model.Only 9% ofnon-union work- Only 9% ofnon-union workplaces had committees (M: presence/
places had committees. absenceofcommittees;O: characteristic ofworkplace/committee
existence).
(Continued )
TABLE II. (Continued )
Article Purpose of study
Reilly et al. [1995] Data generatedby the third Workplace Industrial Analytic study to ascertain relationship between JHSCs and
Relations Survey (1990).WIRS is a systematic
review ofindustrial relations practices in the interest.
UK.Aftersampleattrition,432establishments
were included in the analysis.
Shannon et al. [1992] Survey of1,000 employers in eight sectors,in
Ontario,Canada, supplementedby interviews.
Shannon et al. [1996] Six types ofindustry were chosen: metal articles, Separate questionnaires for management and workers were
plastic articles,grain products,textile manu-
facturing,printing, and automobile manufac-
turing. All were in Canada.
Tuohy and Two separate studies in Canada: the first examin- TheOntariostudyreliedon a surveybased on pooleddata for the
Simard [1993] ingthe Ontario experience and the second the
Quebecexperience.TheOntariostudyreliedon
a survey based on pooled data for the period
1980^1985,conducted for the Advisory
Council on Occupational Health and Occupa-
tional Safety.
Methods (design, methods and context esp.
reunionization)
injuryrates. Study didlook at ‘‘unionization’’as a variable of
Analytic study,using a survey,undertaken for the Ontario Indus-
trial Accident Prevention Association.Each survey involved
four sub-surveys
workerco-chair ofjointcommittee,man-
agement co-chair, senior manager, and human resources
director; 44% of surveyed firms completed all four surveys.
About 30% were unionized workplaces. Surveyrespondents
werecross-tabulatedwiththeiracceptedtimelossfrequency
rate,based onWCB data.Firms were categorized as having
low, medium or high lost-time frequency rates (‘‘low’’as a rate
less than orequal to 50% of the average injuryrate of the rate
group (a group ofcomparable businesses: ‘‘high’’a rate more
than 50% above the rate group average.
constructed.We divided the questionnaire for management
intothree sections. Appropriate respondents were the senior
manager,the HR (personnel) manager, and the management
co-chair of the JHSC.The questionnaires for workers were
directed tothe workerco-chairof the JHSC.Some questions
were identical in the worker and
managementquestionnaires.
period1980^1985,conducted for the Advisory Council on
Occupational Health and Occupational Safety.This survey
examined the functioning ofjoint committees in terms ofindi-
cators such as frequency ofmeetings,record maintenance,
numberofinspections,depth ofmanagement participation
and formulation ofrecommendations.Tuohy correlated these
data with administrative data on accepted injury claims,Min-
istryof Labor inspections and compliance orders. Atotal of
920 complete observations form the basis for the study.The
Results/findings (including mechanism [M] and outcome [O])
Workplaces with JHSCs had,on average, 5.7 few injuries per1,000
employees compared with workplaces without committees.
Found that all forms ofhealth and safety arrangements that encou-
rage workforce participation in safety matters reduce the work-
place injury rate.
A consultative committee with all union appointed safety reps has
5.7% fewer injuries per1,000 employees (M: presence/absence of
committee; O: injury rates).
Surveyconfirmedthatjointcommitteestypicallyengagedinadvisory
and reactive roles and didnot exercise executive authority.
Union structure was also a factorofsome consequence: each
additional stewardper100 members reducedlost time from work
injuries by almost 8%.
This result suggests that unions that push responsibility downwards
are more likely to have a positive impact through their structure
thanunionswhicharemorecentralized(M:presenceofcommittee,
level of function ofcommittee; O: injuryrates).
Lowerlosttimefromworkinjurieswasassociatedwithgreaterworker
participation as well as lowerexpectation that workers simply fol-
low management instructions.
Neither the status of the health and safety coordinator nor the exis-
tence of a written H&S policy was related to lower rates.The activ-
ities of the JHSC were generally unrelated to lost time from work
injuries (M: presence ofcommittee,functioning ofcommittee; O:
injuryrates).
Tuohy found that,by far themostimportant variable explaininglower
relative injury rates wasthe presence of an experienced,stable
workforce.Tuohy didnot find the age of the committee,or the pre-
senceofseniormanagersonthecommittee,insmallandnon-union
workplaces tobe directly related to lower injuryrates.He didpoint
out that thesewerefactorswhich reducedthelikelihoodofinspec-
Systematic Review of Committees
tion.
Thisstudyalsofoundthattherewasageneraltendencyforworkplace
factors to have a greater impact,when compared with committee
431
factors (M: functioning ofcommittees; O: injuryrates).
(Continued )
TABLE II. (Continued )
Article Purpose of study
Walters [1985] Interviewswith14 workers’ health and safety
representatives and 24 company doctors
employedby10 companiesin Ontario,Canada.
Walters and Thesamplewasdrawnfromeightworkplacesin a Interviews with 492 workers.The authors aimed to include
Haines [1988] highly industrialized area in Southern Ontario,
Canada.
Walters et al. [2005] A series ofcase studies in two sectors of the
Canadian economy,this report examines the
role of worker representation and consultation
in improvinghealth and safety performance,
exploringwhich arrangements workbest and
why.
Methods (design, methods and context esp.
reunionization)
second study drewon117 usable surveyreturns correlated to
the above administrative data.The survey was conducted in
1985^1986 and assessed committees in terms ofcertain
performance and capacity indicators.
Descriptive study using interviews.
unionized and non-unionized,large and small,public and pri-
vate sector workplaces.Two of the eight workplaces were
non-unionized and the rest (n ¼ 6) were aimingtobetter
understandlinks between work and health and knowledge, Lackofknowledge and access to information were issues that
attitudes and practices of workers.
Reviews the quantitative and qualitative evidence for the link Found that arrangements for worker representation and consultation
between representative worker participation and effective
health and safety management.
432
Results/findings (including mechanism [M] and outcome [O])
Yassi et al.
The study found that workers’representatives were concerned about
delays in getting jointcommittees to addressproblemsconcretely
or in getting senior management to respond to recommendations.
The management prerogative is presentedby one managerial
interviewee: ‘‘management,emphatically retains its right to make
decisions.The [joint] committees are useful but they are strictly
advisory’’ (M: presence/absence ofcommittee, access to infor-
mation; O: quicker tackling of workplace health and safety
problems).
With respect to information seeking,only 23% ofrespondents
reported asking for information.Those who had asked were more
likely toknow atleastone aspectof thelegislation andtoknow the
identity of their health and safety representative.
emerged:workerscan not ask forinformationiftheydonotknow it
exists or know that they shouldhave access to it (i.e.,environmen-
tal monitoring) (M: knowledgeregarding health and safety; O: bet-
ter workplace health and safety)
are dependent upon a numberofpreconditions for theireffective-
ness.These include a stronglegislative steer,effective external
inspection and control,demonstrable senior management
commitment and capacity towards both health and safety and a
participative approach,competent hazard/riskevaluation and
control,effective autonomous worker representation at the
workplace and external trade union support. Such preconditions
were not present in the majority of the case studies and both they
and the review of the wider literature suggest that changes in the
structure and organization of workmean that achievingthem
presentconsiderable challenges (M: worker representation; O:
engagement/awareness/better health and safety).
(Continued )
 Systematic Review of Committees 433

WorkplaceswithfulltimeH&Srepshadmoreawarenessandfocuson
Unions reported that workers felt that they wouldbenefit fromfurther

(M: health and safety representatives (either full time or part time),
Results/findings (including mechanism [M] and outcome [O])

H&S
these workplaces also had a more critical attitude of H&S

training and information with regard to health and safety and that
interviews with 13 of these firms, found that perceived

12.5^15.7% (P ¼ 0.05).This change was primarily in the non-

management especially needed training and education with
effectiveness of the committee was correlated with a posi-

increased followingthe passage ofmandated committees:

Pre^post study:Extracted Oregon data from1988 to1989 and Found that the overall incidence of the complaint inspections
tive impact on industrial relations, management commit-

union sector (M: workplace inspections; O: complaints).

unionization; O: training,enforcement,knowledge,etc.).
ment to health and safety, level of safety, and the number
of items reviewed by the health and safety committee. Per-
ceived effectiveness of the health and safety committee
was strongly negatively correlated with complaint-generat-
regard torights and responsibilities.

ed Occupational Health and Safety Administration

(OSHA) inspections [Boden et al., 1984]. Weil [1999]
found that mandated committees in Oregon increased the
differences between union and non-union workplaces re-
lated to OSHA enforcement.
Morse et al. [2008] in a study in Connecticut, a state
which requires JHSCs in all workplaces with at least 25
employees, found that lower injury rate companies had a
higher ratio of hourly workers to managers on the JHSC, a
larger committee relative to the size of the company, and
ters. Size of the unions ranged from fewer than 500 members
(8)tolargewithupto23,000members(8).Tenunionshad‘‘full

were white collar.Larger unions were more likely to have this.

conducted in Oregon
1year prior to legislated committees

offered suggestions on a larger average number of worker

unions’participation in occupational health and safety mat-

health and safety officers.Seven were blue collar and three

(i.e., spending more than 70% of their time devoted to H&S)
Descriptive survey,conductedbyphone,ofunion officials on

complaints.
Methods (design, methods and context esp.

UK

In the UK, Reilly et al. [1995] found that workplaces

reunionization)

with JHSCs had, on average, 5.7 fewer injuries per 1,000

1992 to1993 on OSHA inspections

employees compared with workplaces without JHSCs.

Walters et al. [2005], in a mixed methods study, concluded
that worker representation and consultation have a signifi-
cant role to play in improving health and safety, although
and theyear after.

details needed for a CMO analysis were lacking. Beau-

mont and Leopold [1982], in another UK study found that
having the senior manager on a JHSC was necessary for
time’’

committee effectiveness. Other factors deemed important

were commitment of the committee members, good com-
munications, regular meetings, training and general co-op-
Australia) 1991Directory were contacted and
Phone surveyofunion officials in Australia using
an eight-page questionnaire. All unions listed

responses were obtained from all except six

eration between workers and management.

in theTrades and Labor Council (Western

small unions.This left a total of 46 unions

Atotal of 23,536 data records from OSHA

inspections in Oregon were included.

Australia
Purpose of study

Warren-Langford et al. examined the situation in

Australia through a survey with regard to unions and joint
committees. This study found that workers felt that they
interviewed.

would benefit from further training and information with

regard to health and safety and that management especial-
ly needed training and education with regard to rights and
responsibilities [Warren-Langford et al., 1993]. Pragnell
focused specifically on New South Wales, based on
the Australian Workplace Industrial Relations Survey
TABLE II. (Continued )

(AWIRS) which covered 2,004 workplaces with more than

20 employees, of which 762 were in New South Wales,
and found that larger workplaces were more likely to have
Warren-Langford
et al. [1993]

committees [Pragnell, 1994]. Of companies with 200 or

Weil [1999]

more employees, 83% had committees compared with

Article

26% of companies with 20–49 employees. The likelihood

of a committee being established declined as the
434 Yassi et al.

TABLE III. Review Studies

Methods (design, methods, and context esp. re

Article
Beaumont and Leopold [1982]
Burke et al. [2006]
Milgate et al. [2002]
O’Grady [2000]
Parsons [2001]
Shannon et al. [1997]
unionization) Results/findings, including mechanism and outcome
Review Documented details of thefailure of voluntarism to lead to a
satisfactory diffusion of JHSCs in the UK.
Review
to determine therelative effectiveness ofdifferent More engagingtrainingresulted in greater knowledge acquisi-
methods of workersafety andhealthtraining aimedat tion, andreductionswere seen in accidents,illnesses, and
improving safety knowledge andperformance and injuries.
reducingnegative outcomes (accidents,illnesses, and
injuries).
Review Managementcommitment,communication,training and infor-
mation,union involvement,the infrastructure ofan organi-
zation,committee processes and the involvement ofpro-
fessional experts are key to effective committees.
Review Examination of the Canadian context as well as focusing more
generallyonwhatmakescommitteeseffective(groupedby
the Canadian context and then international).
Review Fivekey factorsin achieving effective JHSCs: (1) the need for
government-legislatedmandatory JHSCs; (2) a clear set of
rights for JHSCs; (3) unionsmust bargain overhealth and
safety matters;(4) stronggovernmentenforcement;and (5)
committeesmust be given direct responsibility.
Review
pulled data from10 studies examiningthe relation- Foundthat thevariables associated with lower injuryrates
ship between organizational and workplacefactors and were: empowerment of the workforce (in general matters),
injury rates. delegation of safety activities, and an activerole (in health
and safety) of top management.

proportion of part-time employees increased. Only 9% of
non-union workplaces had committees [Pragnell, 1994].
Milgate et al. [2002], reviewing factors that make
JHSCs effective in the Australian context, noted that most
of the evidence is anecdotal but that the fundamental fac-
tors for effective committees include management com-
mitment, communication, training and information, union
involvement, the infrastructure of an organization, com-
mittee processes, and the involvement of professional
experts.
delegation of safety activities, and top management play-
ing an active role in health and safety. Little has emerged
since then to contradict these findings, as discussed further
below.
Results From the Consultation Process
Through discussion of our findings with the multi-
perspective workplace health and safety experts, we de-
rived 10 themes. These were:

All locations 1. empowerment of committees and the role of worker

The themes that emerged from the literature consisted
of the following: empowerment of committees, training,
access to information, legislation and enforcement, role of
unions in health and safety committees, and management
commitment. Shannon et al. [1997] summarized data from
ten studies and found a statistically significant relationship
between organizational workplace factors and injury rates
[Shannon et al., 1997], noting that JHSC characteristics
(such as representation on the committee, duties and prob-
lem-solving approaches) correlated with injury rates, along
with empowerment of the workforce in general matters,
members with respect to their co-workers (i.e., rights
of the committee members to have time for their
roles and to interact with the constituencies);
2. scope of the committee (i.e., including issues such as
harassment and other mental health issues, not just
safety issues);
3. extending committees to workplaces where they do
not exist;
4. composition of the committee;
5. training and education of committee members;
6. access to information by committees;
7. legislation and enforcement regarding committees;

8. role of unions in health and safety committees;
9. management embracing the value of committees—
management support; and
10. management practices and systems regarding occu-
pational health and safety.
After discussion during the workshop, both in plenary
and in smaller break-out groups, the themes were revised
and re-grouped into three broad overlapping themes, with
the major points emerging presented below.
Role and Scope of Committees
It was noted that JHSCs lack power as they can only
make recommendations and it is in fact the employer
who needs to act to improve workplace conditions. The
group acknowledged that JHSC roles differ depending on
the specific worksite as well as the industry, with the
scope and role of the committee closely inter-related. It
was noted that there is a hierarchy, both within the work-
place but also within the committee. All agreed that terms
of reference, in other words, the mandate and expectations
of the committee, are key and that very little is known
about whether terms of reference are used by committees
and what role they play in committee effectiveness.
The group observed that measuring the worth/value of
committees is difficult as there is a myriad of outcomes of
interest, including ‘‘spin off’’ effects such as improved la-
bor relations and productivity. The discussion on this topic
also stressed that the role of committees can change over
time and is always evolving, along with its scope. Discus-
sion also centered on the ideal size for the committee and
mix of committee members (especially employee to man-
agement). The lack of data available to provide guidance
on this was noted. Some experts noted that many work-
places are simply too small to have effective committees;
in this context, the notion of roving health and safety com-
mittees was suggested as was the issue of sectoral
associations.
Committee Membership Logistics and
Building Member Capacity
Participants discussed the logistics of meetings, noting
that worker representatives sometimes feel less inhibited if
they are able to meet separately; it was noted that the
Ontario Occupational Health and Safety Act requires
the employer to allow worker JHSC members 1 hr of
caucus time. The lack of data from a properly conducted
evaluation of what committee membership structure works
best (e.g., all worker reps with no employer, more worker
reps than employers, equal numbers, etc.) was highlighted.
Also the workplace parties sometimes see the effective-
ness of the committee very differently: management
Systematic Review of Committees 435
may feel they are very supportive of the committee when
the committee does not perceive this to be the case,
noting, for example, inadequate time provided to
members for investigations, follow-up or implementation
of recommendations.
Considerable attention was devoted to capacity build-
ing, with a strong consensus that both education and train-
ing should be a right for committee members. The group
agreed that the extent, content and format of the training
should be specified by legislation, with flexibility depend-
ing on the work setting. For example, safety and health
training should include psycho-social issues.
After viewing an online JHSC training resource devel-
oped by our group [Stime et al., 2012], available at http://
innovation.ghrp.ubc.ca/johsc/, all agreed that online
resources are valuable, but having only online training
materials available would be problematic, as employers
may ask workers to conduct their own training on their
own time. Moreover, while information and communica-
tion technology can be very useful in promoting health
and safety, and enhancing JHSC function, the important
group interaction and discussion amongst the workplace
parties must be retained.
With respect to funding training, it was suggested that
the jurisdiction’s organization or government agency re-
sponsible for workplace health and safety could have a
central fund from which employers could access funding
for education and training of committee members. Having
systems to track training and notify worksites when re-
fresher training is needed was also suggested.
Empowerment
This group began with a debate around the positives
and negatives of the enforcement principle and giving
committees the power to write orders. For example, it was
noted that in the Australian State of Victoria and Queens-
land, worker occupational health and safety representa-
tives have the power to write Provisional Improvement
Orders to which the employer must either comply forth-
with or appeal to the government occupational health and
safety authority. The role of employer associations was
also discussed in this context. There was a consensus
around more research needed on associations and their
value.
A key concern raised was the increased precarious-
ness of employment resulting from extensive economic
restructuring (including increasing use of contactors),
making it difficult to ensure that workers can exercise
their responsibilities without fear of reprisal. There are
now increasing numbers of people working multiple
part-time jobs and in self-employment, and the discussion
focused on who represents these individuals’ rights and
where they can go for help on health and safety issues.
436 Yassi et al.
Overall Context-Mechanism-Outcome
Analysis
As found in previous realist reviews [Greenhalgh
et al., 2007; Jagosh et al., 2011], often the ‘‘outcome’’ in
one study or initiative constitutes the ‘‘mechanism’’ in
another. For example, the outcome of effective problem
solving by JHSCs may be a mechanism to achieve
lower lost-time injuries in another study or context.
Regardless, what is clear is that the legislative context
that is the extent to which JHSCs are mandatory, and
the extent of enforcement, sets the stage for the mecha-
nisms that are adopted to achieve the desired outcome.
The mechanisms identified in the studies to achieve
effectiveness (however, defined), informed further through
the expert consultation, include: adequate information pro-
vided to the committee; proper selection, education and
training of committee members; clear mandate with
the committee empowered to act within its sufficiently
broad mandate; and good problem solving within commit-
tees, fostered by management support. Figure 2 is for a
graphic representation of the framework that supported
this review.
DISCUSSION
Our most striking finding was the paucity of good
quality evidence from which to develop best practice guid-
ance. The literature appears to confirm that merely having
a JHSC is not sufficient—it must be an effective commit-
tee, yet there was no consistency across studies as to the
outcome of interest. The literature, as well as the experts,
supported the value of JHSCs but identified many factors
that determine success, consistent with the findings of
Shannon et al. [1997] 15 years ago. The nature of the leg-
islation seems paramount—although without cross-
FIGURE 2. Conceptual framework for the realist review.
jurisdictional comparisons, the only strong conclusion that
can be made is that JHSCs cannot take the place of regu-
lation and government enforcement due to the nature of
power relations in the workplace.
We note, although not explicitly the focus of our real-
ist review, that there seems to be a relationship between
unionization and the existence of legislation, as well as
other features conducive to health and safety. Post-1960
deviations in unionization rates between the United States
(overall 11.8% in 2011 with only 7% in the private sector,
and 29% in the public sector) versus Canada (29.8% over-
all, with 16% in the private sector and 71% in the public
sector) provide an important contextual factor to take into
account when comparing studies between these two coun-
tries [Warner, 2012]. This not only relates to factors af-
fecting effectiveness in unionized and non-unionized
settings, but the effect that unions can have on the nature
of the legislation that governs JHSCs themselves, while
still ‘‘guard[ing] their role in the education of their mem-
bers’’ [Bernard, 1995].
Issues particularly highlighted at the workshop were
the need for more research focused on obtaining better
clarity on variations in outcome depending on: committee
mandate; optimum size and composition, including man-
ner of selection of members and role of unions; training
and education (e.g., whether bi-partite training is superior
to training by an external agency—including unions,
length of training programs, issues beyond the basic
roles and responsibilities of the committee to be covered;
information and resources for the committees, the role
of information and communication technology); and
management support for the committee (not just for gener-
ally promoting a healthy workplace). These mechanisms
employed to address these workplace factors, under
the backdrop of the legislative context, appear to deter-
mine a committee’s effectiveness. Most importantly, it was

stressed that the outcome of interest may be different
among the committee members, and defining outcome of
interest is essential to evaluating the effectiveness of
committees.
Limitation of the Study
The role of bipartite sectoral associations has been
discussed in the literature [Yassi et al., 2005, 2009], but
was not included in the present review which focused on
workplace-level JHSCs. Bipartite sectoral associations
may have an important role to play in improving JHSC
effectiveness; this also needs to be explored. Also, the
search strategy in this study was based on English lan-
guage terms; this limitation precluded insights published
only in other languages [e.g., Demarais, 2004].
In light of the consistency of themes, yet paucity of
details needed for implementing best practices, we recom-
mend conducting a cluster-randomized controlled trial—
perhaps using paired randomization—of a model JHSC
program—implemented and evaluated against a set of rel-
evant outcome measures—sufficiently broad to encompass
the perspective of the different workplace parties. This
model JHSC program that would be developed should
consist of a top-notch program of education and training,
provision of high quality information, clear mandate and
empowerment through detailed terms of reference, policies
and procedures, and measures implemented to address
each of the items identified as important (or likely impor-
tant) in this systematic review. Principles of evaluating
complex public health interventions [British Medical Re-
search Council, 2006] should be applied [British Medical
Research Council, 2006].
The results of this study could be used to develop
detailed guidelines or possibly, where feasible, a JHSC
Regulation which specifies minimum standards that must
be met regarding how committee members are selected;
role of the union, where it exists; size and composition of
the committee according to different workplace factors,
mandate and scope, and how this should be enshrined in
terms of reference; content, duration, and manner of deliv-
ery of education and training; and access to other needed
information and resources. These would be developed into
‘‘leading indicators’’ of workplace health and safety—to
supplement the ‘‘trailing indicators’’ commonly monitored
such as rates of injuries and time lost from work. When
workplaces are inspected, these leading indicators would
be assessed and recommendations offered—possibly even
orders written for failure to comply. While more research
is desirable to support the greater detail that would be
optimal, there is a clear enough consensus to allow
some initiatives to move forward to strengthen committees
and benefit the health and safety of employees and
workplaces.
Systematic Review of Committees 437
ACKNOWLEDGMENTS
This research was undertaken through grant funding
provided by WorkSafeBC. We also gratefully acknowl-
edge the funding of Dr. Yassi’s Canada Research Chair.
REFERENCES
Akbar-Khanzadeh F, Wagner O. 2001. Safety and health program
assessment in relation to the number and type of safety and health
violations. Am Ind Hyg Assoc J 62:605–610.
Beaumont PB, Leopold JW. 1982. Joint health and safety commit-
tees: The senior management role. Ind Manag Data Syst 82(3/4):
20–23.
Bernard E. 1995. Canada: Joint committees on occupational safety
and health. In: Rogers J, Streeck W, editors. Works councils:
Consultation, representation, and cooperation in industrial relations.
Chicago, IL: University of Chicago Press. p. 351–374.
Boden LI, Hall JA, Levenstein C, Punnett L. 1984. The impact of
health and safety committees: A study based on survey, interview,
and occupational safety and health administration data. J Occup Med
26:829–834.
British Medical Research Council. 2006. Developing and evaluating
complex interventions: New guidance. London: UK, Medical Re-
search Council available at http://www.mrc.ac.uk/Utilities/Documen-
trecord/index.htm?d¼MRC004871
Bryce GK, Manga P. 1985. The effectiveness of health and safety
committees. Relat Industrielles 40(2):257–283.
Burke MJ, Sarpy SA, Smith-Crowe K, Chan-Serafin S, Salvador RO,
Islam G. 2006. Relative effectiveness of worker safety and health
training methods. Am J Public Health 96:315–324.
Cooke W, Gautschi F. 1981. OSHA, Plant Safety Programs, and In-
jury Reduction. Ind Relat 20:245–247.
Dembe A. 2001. The social consequences of occupational injuries
and illnesses. Am J Ind Med 40:403–417.
Desmarais L. 2004. Évaluation de l’implantation des comités de
santé et de sécurité du travail: une étude de cas multiples réalisée
dans les petites et moyennes entreprises au Québec. Ph.D., UQAM.
Eaton A, Nocerino T. 2000. The effectiveness of health and safety
committees: Results of a survey of public-sector workplaces. Ind
Relat 39:265–290.
Frick K, Walters D. 1998. Worker Representation on Health and
Safety in Small Enterprises: Lessons from a Swedish Approach. Int
Labour Rev 137:367–377.
Geldart S, Shannon HS, Lohfeld L. 2005. Have companies improved
their health and safety approaches over the last decade? A longitudi-
nal study. Am J Ind Med 47:227–236.
Geldart S, Smith C, Shannon H, Lofeld L. 2010. Organizational
practices and workplace health and safety: A cross-sectional study in
manufacturing companies. Saf Sci 48:562–569.
Greenhalgh T, Kristjansson E, Robinson V. 2007. Realist review to
understand the efficacy of school feeding programmes. Br Med J
335:858–861.
Hall A, Forrest A, Sears A, Carlan N. 2006. Making a difference:
Knowledge activism and worker representation in Joint OHS Com-
mittees. Relat Industrielles 61:408–436.
Havlovic S, McShane S. 1997. The effectiveness of joint health and
safety committees and safety training in reducing fatalities and
438 Yassi et al.
injuries in British Columbia forest product mills. Richmond, BC:
Workers Compensation Board of British Columbia.
Jagosh J, Pluye P, Macaulay AC, Salsberg J, Henderson J, Sirett E,
Bush PL, Seller R, Wong G, Greenhalgh T, Cargo M, Herbert CP,
Seifer SD, Green LW. 2011. Assessing the outcomes of participatory
research: Protocol for identifying, selecting, appraising and synthe-
sizing the literature for realist review. Implement Sci 6:24.
Kochan T, Dyer L, Lipsky D. 1977. The effectiveness of union man-
agement safety and health committees. Kalamzoo, MI: Upjohn Insti-
tute for Employment Research.
Levesque C. 1995. State Intervention in Occupational Health and
Safety: Labour-Management Committees Revisited. In: Giles A,
Smith A, Wetzel K, editors. Proceedings of the XXXIst Conference
of the Canadian Industrial Relations Association. Toronto: CIRA,
p. 217–231.
Lewchuk W, Robb L, Walters V. 1996. The effectiveness of bill
70 and joint health and safety committees in reducing injuries
in the workplace: The case of Ontario. Can Public Policy 22:
225–243.
Milgate N, Innes E, O’Loughlin K. 2002. Examining the effective-
ness of health and safety committees and representatives: A review.
Work 19:281–290.
Morse T, Goyzueta J, Curry L, Warren N. 2008. Characteristics of
effective job health and safety committees. J Environ Occup Health
Policy 18:441–457.
Nichol K, Kudla I, Manno M, McCaskell L, Sikorski J, Linn Holness
D. 2009. Form and function of Joint Health and Safety Committees
in Ontario acute care hospitals Healthcare Quarterly, p. 12.
Ontario Advisory Council on Occupational Health and Occupational
Safety. 1986. Eighth annual report, April 1, 1985 to March 31, 1986,
vols. 2, 3, and 7. Toronto: Government of Ontario.
O’Grady J. 2000. Joint health and safety committees: Finding a
balance. In: Sullivan T, editor. Injury and the new world of work
Vancouver, BC: UBC.
Parsons MD. 1989. Worker participation in occupational health and
safety: Lessons from the Canadian experience. Lab Stud J 13(4):
22–32.
Pawson R, Tilley N. 1997. Realistic evaluation. London, UK: Sage.
Pawson R, Greenhalgh T, Harvey G, Walshe K. 2005. Realist re-
view—A new method of systematic review designed for complex
policy interventions. J Health Serv Res Policy 10(Suppl 1):21–34.
Pragnell B. 1994. Occupational & health committees in NSW: An
analysis for the AWIRS data. Sydney: Australian Centre for Industri-
al Relations Research and Teaching.
Putz V, Schulte PA, Sestito J, Linn H, Nguyen L. 2010. Occupational
fatalities, injuries, illnesses, and related economic loss in the whole-
sale and retail trade sector. Am J Ind Med 53:673–685.
Reilly B, Paci P, Holi P. 1995. Unions, safety committees, and work-
place injuries. Br J Ind Relat 33:275–289.
Schulte PA. 2005. Characterizing the burden of occupational injury
and disease. J Occup Environ Med 47:607–622.
Shannon HS, Walters D, Lewchuk W, Richardson R, Verma D,
Haines T, Moran L. 1992. Health and safety approaches in the work-
place. Hamilton, Ontario: McMaster University.
Shannon HS, Walters V, Lwechuk W. 1996. Workplace organization-
al correlates of time-loss accident rates in manufacturing. Am J Ind
Med 29:258–268.
Shannon HS, Mayr J, Haines T. 1997. Overview of the relationship
between organizational and workplace factors and injury rates. Saf
Sci 26:201–217.
Stime B, Sykes M, Buck B, Yassi A, Moir S. 2012. An online re-
source tool for joint health and safety committees: Results of the first
phase evaluation. Canadian Association for Research on Work and
Health (CARWH) conference, Vancouver, BC.
Tuohy C, Simard M. 1993. The impact of joint health and safety
committees in Ontario and Quebec. Prepared for the Canadian Asso-
ciation of Administrators of Labour Law. Toronto: Ontario Ministry
of Labour.
Walters D. 1985. The Politics of Occupational Health and Safety:
Interviews with Workers’ Health and Safety Representatives and
Company Doctors. Can Rev Sociol Anthropol 22:57–79.
Walters D, Nichols T, Connor J, Tasiran AC, Cam S. 2005. The role
and effectiveness of safety representatives in influencing workplace
health and safety. RESEARCH REPORT 363 Cardiff University
for the Health and Safety Executive. Available at: http://www.hse.
gov.uk/research/rrpdf/rr363.pdf
Walters V, Haines T. 1988. Workers’ perceptions, knowledge and
responses regarding occupational health and safety: a report on a
Canadian study. Soc Sci Med 27:1189–1196.
Warner K. 2012. Protecting fundamental labor rights lessons from
Canada for the United States. Washington, DC: Center for Economic
and Policy Research.
Warren-Langford P, Biggins D, Phillips M. 1993. Union participation
in occupational health and safety in Western Australia. J Ind Relat
35:585–606.
Weil D. 1999. Are mandated health and safety committees substi-
tutes for or supplements to labor unions. Ind Lab Relat Rev 52:339–
360.
World Health Organization. 2008. Who handbook for guideline de-
velopment. World Health Organization. Available at http://www.sear-
o.who.int/LinkFiles/RPC_Handbook_Guideline_Development.pdf
Yassi A, Ostry AS, Hatter B, De Boer HM. 2005. Joint health and
safety committee education and the value of bipartite cooperation in
the healthcare sector in British Columbia, Canada Int J Occup Envi-
ron Health 11:305–312.
Yassi A, Bryce E, Spiegel J. 2009. Building capacity to secure
healthier and safer working conditions for healthcare workers: A
South African-Canadian collaboration. Int J Occup Environ Health
15:360–369.