LeadershipdevelopmentasaninterventioninoccupationalhealthpsychologyE.KevinKellowaya*andJulianBarlingbaDepartmentofPsychology,SaintMarysUniversity,Halifax,NovaScotia,Canada;bQueensUniversitySchoolofBusiness,Kingston,Ontario,CanadaA growing body of literature suggests that organizational leadership is linked to a wide varietyof employee outcomes, both positive and negative, relevant to occupational health and safety.All organizations have individuals in a leadership role, but few researchers consider leadershiptraining as an effective intervention. This may be because such studies are difficult to conductand because the target, being the employees, is indirect. In this paper for the special edition ofWork&Stress, wereviewstudies linkingleadershiptoindividual well-beingandsafetyinorganizations. These include studies concerning leadership style, abusive supervision andorganizationalfairness.Wehighlightinterventionstudiesthatsuggestthattheselinkagesarecausal and that leadership development, usually in the formof training, is an effectiveinterventioninoccupational healthpsychology. It isproposedthat leadershipdevelopmentshould be a main target for research on interventions in Occupational health psychology. Thecharacteristicsofleadershipdevelopmentinterventionsanddirectionsforfutureresearcharediscussed.Keywords: leadership; training; interventions; supervision; justice; well-being; work-relatedstressIntroductionThe observation that leadership has an effect on individual well-being would come asnosurprise toany working adult (Gilbreath, 2004). What might be surprising,however, is thebreadthof theeffects (Mullen&Kelloway, 2011). Thequalityofleadershiphas beenlinkedtoanarrayof outcomes withinoccupational healthpsychology: positive outcomes such as psychological well-being (e,g., Arnold,Turner, Barling, Kelloway, &McKee, 2007), and organizational safety climate(e.g., Zohar, 2002a) and negative outcomes, including employee stress (e.g.,Offermann&Hellmann, 1996), cardiovasculardisease(e.g., Kivimaki etal., 2005;Wager, Feldman, &Hussey, 2005), workplaceincidentsandinjuries(e.g., Barling,Loughlin, & Kelloway, 2002; Kelloway, Mullen, & Francis, 2006; Mullen &Kelloway,2009)andhealth-relatedbehaviourssuchasalcoholuse(e.g.,Bamberger&Bacharach, 2006). In short, virtually every outcome variable in the field of*Correspondingauthor.Email:Kevin.kelloway@smu.caWork&StressVol.24,No.3,JulySeptember2010,260279ISSN0267-8373print/ISSN1464-5335online# 2010Taylor&FrancisDOI:10.1080/02678373.2010.518441http://www.informaworld.comoccupational healthpsychologyis empiricallyrelatedtoorganizational leadership(Mullen&Kelloway,2011).Moreover,dataarestartingtoemergesuggestingthatimprovingorganizationalleadership results in improved safety outcomes (see, for example, Mullen & Kelloway,2009;Zohar,2002a)andenhancedemployee well-being(McKee&Kelloway,2009).Despite these data, discussionsof occupational health and safety interventions (e.g.,Burke&Sarpy, 2003) onlyrarelyconsiderleadershiptrainingasanoccupationalhealthandsafetyintervention. Inthisarticle, wereviewtheempirical evidenceforthesuggestionthat leadershipdevelopment shouldbeconsideredas aviableandeffectiveprimaryinterventioninoccupationalhealthpsychology.Wethenconsider theuniquenatureof leadershipdevelopment unlikemostinterventions in occupational health psychology, leadership development is notprimarily intended to affect the experience and behaviour of individuals whoparticipateintheintervention rather, theintentionofleadershipdevelopment ismore typically to affect those who do not participate in the intervention (i.e.,subordinates or employees). Indeed, the design and evaluation of leadershipinterventions is further complicatedbythe fact that indirect anddelayedeffectswould be of primary interest. This characteristic poses practical difficulties indesigningandshowingtheeffectivenessofleadershipdevelopmentinterventions.DefiningleadershipEstablishing and discussing the potential for leadership development as anoccupational health psychology intervention requires, in the first instance, acommonlyunderstooddefinitionofthewordleadership.Offeringsuchadefinitionisadauntingtask. AsStogdill (1974, p. 259) notedsome35yearsagotherearealmost as many definitions of leadership as there are persons who have attempted todefine theconcept;synthesizingthisvastliteratureto derive asingledefinitionisataskwellbeyondthescopeofthisreview.Yukls(2006)distinctionbetweenleadershipasaspecializedroleandleadershipas a shared influence process offers a way forward. He notes that the termleadership is usedintwofundamentallydifferent ways. First, all organizationshave formal leadership roles (e.g., mangers, supervisors) that have specialized, non-sharedfunctions; inorganizations, individuals canbe describedas leaders orfollowers. He also notes that leadership is often used to describe a process of socialinfluence in this view anyone in a group can demonstrate leadership. Much of whatis labelled leadership research has focused on identifying the behaviours orstyles that constitute effective social influence (for a review see Barling, Christie, &Hoption,2011).For our purposes, we define leadership as constituting a process of socialinfluencethatisenactedby designatedindividuals whohold formalleadershiprolesin organizations. Although we do not mean to imply that leadership as socialinfluenceislimitedtotheformalleadersofanorganization,we domeantosuggestthat thoseinformal leadershiproleshaveaparticularlystrongpotential toaffectoutcomesrelevanttooccupational healthpsychology. Indeed, wewouldgoasfaras to suggest that the relationship with ones formal leader in an organization is oneWork&Stress 261of the most important workplace relationships with implications for individualwell-being.The relationshipwiththe formal leader is particularly important for severalreasons. First, formal leaders inorganizations serve as models for others intheorganization. Leaders, forexample, model safeorunsafeworkingprocedureswithconsequences for followers willingness toengageinsafeworkpractices (Cree&Kelloway, 1997). Second, as individuals who possess formal power within theorganization, formal leadersareinapositiontorewardorpunishfollowers. Asaresult, thewayinwhichleadersinteractwiththeirsubordinatesassumesagreaterimportanceforfollowerwell-being.Third,andrelatedly,formalleadersoftenmakedecisions that createadditional stressors for their followers. For example, leadersassigntasks toothers andcandosoinawaythat increases or diminishes thefollowers experienceof roleoverload. Inthissenseformal leaderscanbearootcauseoforganizationalstress(Kelloway,Sivanathan,Francis,&Barling,2005).Thus, forthepurposeofthisreviewwefocusontheroleoftheformal leader.Morespecifically, wefocus onthewayinwhichformal leaders interact, treat orbehavetowardstheirfollowers. Aswill benoted, muchof theempirical literatureexamining leaders treatment of followers has emerged in avariety of domainsincludingstudies of leadership style,abusive leadershipand organizationalinjustice.Adoptingabroadperspectiveonleadershipallowsustodrawonrelevantfindingsfromall of these domains tofocus onthe way inwhichformal leaders behavetowardstheirfollowers.This is an appropriate focus in that data linking the quality of leadership to otherindividuals well-being has been available for almost 50 years (e.g., Day & Hamblin,1964), andevidencelinkingpoorleadershiptoimpairedwell-beinginfollowersisparticularly well-established (for a review see Kelloway et al., 2005). These effects arefar-reaching,andincludeeffectsonbothemployeehealthandemployeesafety.LeadershipandemployeepsychologicalhealthThere isconsistentevidencelinkingleadershipin organizationstothepsychologicalwell-beingof employees, includingoutcomesrelatedtobothill-health(e.g., stress,strain) and more positive conceptualizations of health (e.g., positive moods). In theirmeta-analysis of 27empirical studies, Kuoppala, Lamminpaa, Liira, andVainio(2008) reported moderately strong (i.e., Median Risk Ratio of 1.40; Mediansubsamplecorrelationof.26)relationshipsbetweendifferentdimensionsofleader-ship(considerate, supportive andtransformational leadership) andenhancedjobwell-being (e.g., lower anxiety, job stress, and depression) with most of the empiricaldata addressing the relationship between poor leadership and impaired psychologicalwell-being.Leadershipwasalsoshowntobeassociated withmoreobjectiveindicessuchas usage of sickleave (MedianRiskRatioof 0.73) anddisabilitypensions(Median Risk Ratio of 0.46). Although the meta-analysis included both cross-sectional andprospectivestudies, mostofthedataforthestudycamefromcross-sectionalstudies.Intheirsystematicreview oftheliterature,Skakon,Nielsen,Borg,andGuzman(2010)foundsupportfortheirhypothesisthatleaderbehaviours,specificleadershipstylesandtherelationshipbetweenleadersandtheiremployeeswereall associated262 E.K.KellowayandJ.Barlingwithemployee stress andaffective well-being. The available researchalsooffersconsiderableguidanceastothespecificaspectsof leadershipthat might result inimpaired employee psychological well-being. Broadly speaking, most of the empiricalliterature has focused on either some aspect of leadership style (e.g., abusive, passiveorpositiveleadership),or onemployeeperceptionsofleadersfairness.AbusiveleadershipAwell-developedstreamof researchlinks abusive supervisioninparticular withdiminishedjobsatisfactionandincreasedemployeedistress(Tepper,2000).Definedastheemployees perceptionthattheleaderisengaginginasustaineddisplayofhostile verbal and non-verbal behaviours, excluding physical contact (Tepper, 2000,p. 178), abusiveleadershipmanifestsitselfinthepublicridiculingofsubordinates,blaming subordinates for mistakes they did not make (Tepper, Duffy, & Shaw, 2001)andtheuseofderogatorynamesandintimidation(Keashly,1998).Abusivesupervisionhas beenempiricallylinkedtoimpairedwell-beingmani-fested as burnout (Grandey, Kem, & Frone, 2007), feelings of helplessness (Ashforth,1997), diminished levels of self-efficacy (Duffy, Gangster, & Pagon, 2002), self-esteem(Burton & Hoobler, 2006), affective commitment to the organization (Tepper, 2007),andincreasedemployeestrain(Harvey,Stoner,Hochwarter,&Kacmar,2007).There is a striking parallel between accounts of destructive leadership anddescriptionsofworkplacebullying.Indeed,severalauthorshavedrawntheparallel,arguingthat destructiveleaders areworkplacebullies (e.g., Ferris, Zinko, Brouer,Buckley, &Harvey, 2007; Harvey, Buckleyet al., 2007). Hauge, Skogstand, andEinarsen(2007) providedempirical support forthissuggestionintheirstudyof arepresentativesampleof Norwegianworkers. Theirresultsshowedthat tyrannical(and laissez-faire) leadership styles were related to workplace bullying. As a result oftheir analyses, the authors suggest that bullying is more likely in environmentscharacterizedbytyrannical leadership, andis particularlyprevalent whensuper-visorsdonotintervenetopreventandmanagebullying(Haugeetal., 2007). Thislatterfindingmightprovidesomeindicationforeffectiveinterventions.TransformationalleadershipAlthoughtheexistingresearchhasfocusedalmostexclusivelyonnegativeorpoorleadership (for a review see Kelloway et al., 2005), data are now emerging suggestingthat morepositiveformsof leadershiphaveabeneficial effect onindividual well-being. At one level, there is anextensive bodyof researchonthe effects of theamount and quality of support from managers (see for example, Halbesleben, 2006).These data showthat managerial support has been linkedwith lower levels ofperceivedstress, jobstrain, burnout anddepression(e.g., Lee &Ashforth, 1996;Moyle, 1998; Rooney & Gottlieb, 2007; Van Dierendonck, Haynes, Borrill, & Stride,2004).Consistentwiththesefindings,thereisagrowingbody ofliteraturefocusingonthe effects of transformational leadership onindividual well-being. Transforma-tionalleadershiptheoryisthesinglemostwidelystudiedleadershiptheory(Barlinget al., 2011) and there is an extensive body of literature documenting theWork&Stress 263performance-relatedeffectsoftransformational leadership(e.g., Barling, Weber, &Kelloway, 1996; forareviewseeBarlinget al., 2011).Transformational leadership is defined in terms of four particular types ofbehaviours. Idealized influence takes place when leaders do what is proper and ethicalrather thanwhat iseffortless,and when they are guided by theirmoralcommitmentto their followers and go beyond the interests of the organization. Leaders exhibitinginspirationalmotivation inspire their employeestoachieve more thanwhatwasoncethoughtpossiblebysettinghighstandardsandarticulatinga visionofwhatcanbeachieved. Leaderswhomanifestintellectual stimulationhelpemployeestoquestiontheirowncommonlyheldassumptions,reframeproblems,andapproachmattersininnovative ways. Finally, individual considerationoccurs whenleaders payspecialattentiontotheemployees needsforachievement anddevelopment; theyprovideneededempathy,compassionandguidancethatemployeesmayseekfortheirwell-being. Althoughthere is a considerable evidence for the performance effects oftransformationalleadership,researchers have nowbegun toextendtheiranalysestoconsiderthehealth-relatedeffectsoftransformationalleadership.SosikandGodshalk(2000) foundthat transformational leadershipbehaviour(e.g., social support provided through mentoring a formof individualizedconsideration)indirectlypredictedreducedjob-relatedstress.Furthermore,mentor-ing functions receivedby proteges moderatedthe linkbetweentransformationalleadershipandstress, suchthat therelationshipwas stronger for thesamplethatreceivedhighmentoringfunctions.Usinganexperiencesamplingmethodologyandwithin-personanalyses, Bono,Foldes, Vinson, andMuros(2007)examinedeffectsoftransformational leadershiponstressandsatisfactionatwork. Aswouldbeexpected, participantsexperiencedgreater optimism, happiness and enthusiasmwhen their supervisor engaged intransformationalleadershipbehaviours,comparedtoemployeeswhodidnot.Arnold et al. (2000)7) also present evidence that leaders transformationalleadership was associated with employee well-being. In two studies, they showed thatthiseffectwasmediatedbyemployeessenseofmeaningfulwork.Similarly,Nielsenand her colleagues (Nielsen, Yarker, Brenner, Randall, & Borg, 2008) found that thelinkbetweenleadershipandwell-beingwas partiallymediatedbyperceptions ofwork characteristics (i.e., involvement, influence and meaningful work). Importantly,theseresults werereplicatedandclarifiedinasubsequentlongitudinalextensionofthe original study (Nielsen, Randall, Yarker, & Brenner, 2008). The longitudinal datasupported perceived job characteristics as a mediator between transformationalleadershipand well-being.Although the currently available data support perceptions of the job as amediator, there is also some support for other mechanisms. For example,transformational leaders also positively influence employee psychological well-beingbyevokingpositive emotions throughtheir interactions withsubordinates(Bonoet al., 2007).More recently, McKee andcolleagues (McKee, Kelloway, Driscoll, &Kelley,2009) showed that perceptions of transformational leadership predicted bothemployees sense of workplace spirituality and employee well-being. Spiritualitywas defined as the experience of meaningful work, a sense of community inthe workplace and shared values between the individual and the organization.264 E.K.KellowayandJ.BarlingInadditiontobeingapredictorofwell-being, workplacespiritualitymediatedtherelationshipsbetweentransformationalleadershipandindividualwell-being.Importantly, there are some emerging datasuggesting that changes in leadershipresultinchangesinfollowerwell-being.McKeeandKelloway(2009)reportedonafield experiment examining the effects of leadership development. Participant leaderswere randomly assigned to either a development (i.e., workshop and feedback,Barling et al., 1996; Kelloway, Barling, &Helleur, 2000) or a control group.Comparisonof pre-test andpost-test datafromsubordinatesshowedthat (a) theleadership intervention was successful in enhancing subordinate perceptions ofsupervisory transformational leadership style, (b) enhanced perceptions of leadershipwereassociatedwithindividual psychological well-beingand(c) theexperienceofworkplacespiritualitymediatedtherelationshipbetweentransformational leader-shipandindividualwell-being.LeadershipandemployeephysicalhealthAgain, the positive effects of leadership style on well-being extend beyondpsychological measurestoincludeeffectsonphysical health. Forexample, havingasupportivesupervisorwasassociatedwithlowersystolicbloodpressureamongasample of New York City traffic enforcement agents (Karlin, Brondolo, &Schwartz, 2003). Intheir analysis of prospective data fromthe SwedishWOLFstudy, Nybergetal. (2009)presentresultssuggestingthattheeffectsofleadershipstyleonindividual well-beingarealsomanifestedintermsof physical health.Good leadership style (defined as consideration for individual employees,provision of clarity in goals and role expectations, supplying information andfeedback, ability to carry out changes at work successfully, and promotion ofemployeeparticipationandcontrol, Nyberget al., 2009, p. 51) at timeonewasrelated to subsequent ischemic heart disease in employees, such that higher scores onleadershipwereassociatedwithreducedrisk ofheartdisease.Theseresultsheldupevenafter controlling for a host of traditional riskfactors suchas smoking,exercise andbloodpressure. Moreover, the strengthof the relationshipbetweenleadership and heart disease was stronger the longer the individual had worked in thesame workplace (and presumably was exposed to the same type of supervision). TheauthorsarticulatethemainmessageoftheirpaperasThereisaprospectivedose-responserelationshipbetweenconcretemanagerial behaviours andischemicheartdiseaseamongemployees (Nybergetal., 2009, p. 55), andnotethepotential forleadership development to be a primary means of health promotion in the workplace.Leadershipandhealth-relatedbehavioursInadditiontothefocus onhealthoutcomes, researchthat has examinedhealth-related behaviours has also drawn a link with leadership style. For example,supportive leadership enhanced the success of a broad band organizational healthpromotion programme designed to reduce obesity, smoking and alcohol use(Whiteman, Snyder, &Ragland, 2001) aswell asprogrammesfocusedspecificallyonsingle behaviours suchas smoking cessation(Eriksen, 2005). It is becomingWork&Stress 265apparentthatleaders supportofprogrammesmaybeacriticaldeterminantofthesuccessofhealthpromotioninterventionsinorganizations.This conclusionis enhancedby an experimentalstudy aimed at investigationtheinfluence of leaders in increasing vaccination among health care employees(Slaunwhite, Smith, Fleming, Strang, &Lockhart, 2009). Rather thanfocus onformal leaders (i.e., supervisors, managers), the authors identifiedkey members(called champions) in health care units. Champions were selected by anominationprocess inwhichsupervisors were askedtonominate one individualfromeachparticipatingunitwhooperatedinafront-linecapacityandwerewell-liked by coworkers . . . who were viewed as a leader in their department . . . thatcoworkers trusted, whowerecommittedtothe followthroughonthe studyandwilling to promote and encourage co-workers to accept influenzavaccination.(Slaunwhiteetal.,2009).Championsreceivedaone-dayinformationsessionontheimportanceofvaccination.Comparisonofdatafrom23units withchampions withacontrolgroup(23unitswithoutchampions)showedasignificantlyhigherrateofvaccinationinthe experimental group. Moreover, drawing onarchival data, theauthors demonstratedasignificant 10%increase invaccinationrates amongtheunits withchampions, withonlyamodest non-significant change inthe controlgroupunits.Althoughdatasupport thepositiveeffectsof supportiveleadershipondesiredhealth-related behaviours, there are also compelling data suggesting that poorleadership results in more risky behaviours. Bamberger and Bacharach (2006)showeda link betweenabusive supervisory behaviour and subordinate problemdrinking, including the increasedoccurrence of feelings of guilt associatedwithdrinking,feelingsthatdrinkingbehaviourshouldbereduced,havingthefirstdrinkin the morning and feeling annoyed when criticized about drinking behaviour(Bamberger&Bacharach,2006).LeaderinjusticeandphysicalhealthConsistent with these findings, a great deal of recent data has emerged showing thatleaders unfair treatment of employees is associatedwith adverse outcomes foremployees.Intheirmeta-analysis,Colquittetal.(2001)reportedmoderatelystrongrelationships between perceptions of organizational justice and measures of context-specific mental health (Warr, 1987) such as job satisfaction and organizationalcommitment. For both criteria, procedural and distributive injustice was thestrongestpredictors.Wesuggestthatthesefindingshaveimplicationsforleadershipin that it is the organizational leaders who make (distributive) justice and implement(procedural) justice in organizations. More direct evidence linking leaders behaviourandorganizationinjusticecomesfromaseriesofprospectivestudiesconductedbyKivimaki and his colleagues (Kivimaki, Elovainio, Vahtera, & Ferrie, 2003;Kivimaki et al., 2005) that have identified procedural (organizational) and relational(supervisory) injusticeaspredictorsofminorpsychiatricmorbidityaswell assickabsence. Inthewell-knownWhitehall IIstudiesdatahavealsoemergedsuggestingthe importance of supervisory injustice as a predictor of psychiatric morbidity(Ferrie et al., 2006). These findings are consistent withdecades of researchthathighlight the importance of process fairness in organizations. The effects ofsupervisory injustice on well-being are not limited to psychological outcomes.266 E.K.KellowayandJ.BarlingRather, agrowingbodyofliteraturepointstoempirical linksbetweensupervisoryinjustice andawide range of health-relatedoutcomes including heavy drinking(Kuovonenet al., 2009), impairedcardiac regulation(Elovonio, Kivimaki et al.,2006),anduseofsicktime(Kivimaketal.,2003).While social scientists may be sceptical about the magnitude or clinicalimplication of these effects, it is worth noting that studies have consistentlydocumented an association between supervisory injustice and mortality fromcardiovascular mortality (see for example, Kivimaki et al., 2003, 2005). In oneprospective cohort study, employees reporting more favourable experiences of justiceat workhada45%lowerriskof cardiacdeaththanrespondentsreportinglowerlevels of justice (Elovainio, Leino-Arjas, Vahtera, & Kivimaki, 2006). These data areconsistent with a growing literature showing the positive effect on the cardiovascularsystemofsupportiveandfairsocialinteractions(seeHeaphy&Dutton,2008forareview), includingeffectsonbothsystolicanddiastolicbloodpressure(Brondoloetal.,2003;Wageretal.,2003),andstrengthenedimmunesystems(Kiecolt-Glaser,McGuire,Robles,&Glaser,2002).Inaninteresting fieldstudy,SparrandSonnentag(2008)suggestedthatfairnessandstyle of leadershipmaybe relatedtoemployee well-being. Specifically, theyfoundthatthefairnessofperformancefeedbackwasrelatedtoemployee well-being(asmeasuredbydepression, jobsatisfactionandperceptionsof control), but thatsome of these relationships were mediated by the quality of leader-memberexchanges. Leader-member exchange theory (Gerstner &Day, 1997) focuses ontheontherelationshipbetweenleadersandfollowersratherthansolely onleadersbehaviour as do theories of abusive or transformational leadership. Thus, thefindingsofSparrandSonnentag(2008)areconsistentwithamodelwhereinleaderbehaviour(i.e.,thefairnessofperformancefeedback)influencesfollowerwell-beingbyinfluencingthequalityoftherelationshipbetweenleadersandfollowers.LeadershipandoccupationalsafetyAsisthecasewithhealthoutcomes, alargeandconsistent bodyofliteraturehasemerged documenting the relationship between organizational leadership and safetyoutcomes. In particular, the available evidence seems to support a direct link betweenleaders behaviourandpsychological orbehavioural aspectsofsafety. Inturn, thepsychological andbehavioural aspects of safetyseemtomediatetherelationshipbetween leaders behaviour and safety outcomes such as incidents or injuries (see forexample,Barlingetal.,2002).Dataconsistentlysupporttherelationshipbetweentransformational leadershipbehavioursandperceivedsafetyclimatewithinorganizations(Barlingetal., 2002;Hofmann&Morgeson, 1999; Kelloway et al., 2006; Mullen&Kelloway, 2009;Zohar, 1980; Zohar, 2002; Zohar&Tenne-Gazit, 2008). Theseassociationsshouldnot be unexpected in that safety climate has been defined in terms of the perceptionsof leaders behaviourthat is, asthesharedperceptionsof managerial policies,proceduresandpractices(Zohar,2002,p.75)relatingtosafety anditisasmallleaptoassumethatleaderswhoareseenaspromotingsafetywouldalsocreateapositive safety climate among their followers. In their review, Flin, Mearns,Work&Stress 267OConnor, and Bryden (2000) found that perceptions of management were the mostcommondimensionassessedinmeasuresofsafetyclimate.Zohar (2000) found that climate perceptions were related to supervisory practicesas opposed to organizational policies and procedures concerning safety. In asubsequentstudy,hefoundthatthenegativerelationshipbetweentransformationalleadershipandoccupationalinjuries wasmediatedbythreesafetyclimatevariables,includingtheextent towhichsupervisors tookpreventativeaction, theextent towhich supervisors were reactive to safety issues, and finally, the supervisorsprioritizationof safety(Zohar, 2002). Zohar andTenne-Gazit (2008) alsofoundthattransformational leadershipwasassociatedwithperceptionsofsafetyclimate,although the relationship was mediated by the density of the communicationnetwork. Over all, the available literature draws a clear link between leaderstransformationalleadershipandperceptionsofsafetyclimate.Researchershavealsodemonstratedthepositiveeffectsofsupportiveleadershipontask(e.g., safetycompliance) andcontextual (e.g., safetyparticipations) safetyperformance (e.g., Barling et al., 2002; Hofmann & Morgeson, 1999; Kelloway et al.,2006). For example, Mullen (2005) found that employees reported a greaterwillingness to voluntarily raise safety concerns (e.g., safety participation) whensupervisors were perceived as supportive and likely to listen to their concerns.Hofmann, Morgeson, andGerras(2003) foundthat high-qualitysocial exchangesbetween leaders and employees resulted in expanded role definitions (e.g., employeesperceived safety as part of their job responsibilities), which in turn, predictedemployeesafetycitizenshipbehaviour. Thelinkbetweenhigh-qualityleadersocialexchange and employee safety role definitions was moderated by employeeperceptionsofsafetyclimate.MullenandKelloway (2009) reportedona fieldstudy inwhichhealthcaremanagers were randomly assigned to one of three conditions; safety-specifictransformational leadershiptraining, general transformational leadershiptrainingandacontrol group(notraining). Theyshowedthat bothleaders safety-relatedattitudesandbehavioursandemployees safety-relatedattitudesandoutcomes werepositivelyinfluencedbythesafetyspecificmanagementtraining.Specifically,safetyattitudes andoutcomeswereenhancedfor bothleaders andemployeeswhentheleadersweretrainedinsafetyspecifictransformationalleadership.Kelloway et al. (2006) argued, and empirically demonstrated, that managers whoexhibitpassiveformsofsafetyleadership forexample,management-by-exception(passive), andlaissezfaire) adverselyaffect safetyoutcomes. Theysuggest thatpassive leaders who do not talk about safety in effect communicate the message thatsafetyisnotimportant.Inturn,employeesbelievethatsafetyisnotvaluedintheirorganization,resultinginnegativesafetybehaviourandincreasedinjuryrates(e.g.,see Zohar, 2002a; Zohar, 2002b). In the Kelloway et al. (2006) study, passiveleadershipaccountedfor significant incremental varianceinsafetyconsciousness,safety climate, safety-related events, and injuries, beyond that explained bytransformational leadership. More recently, Teed, Kelloway, and Mullen (2008)examinedtheeffectsofinconsistent leadershiponemployeesafetyoutcomes(e.g.,when leaders display both transformational and passive leadership behaviours).Passiveleadershipmoderatedtherelationshipbetweentransformational leadershipand the safety outcomes employee safety citizenship behaviour, and employee safety268 E.K.KellowayandJ.Barlingattitudes. These results suggest that better safety outcomes are achieved when leadersareconsistentchampionsofsafety.Relative to data linkinghealth outcomesto negative leadershipin organizations,muchlessempirical attentionhasfocusedontheeffectsof negativeleadershiponsafetyoutcomes inorganizations. Conceptually, reactance theory(e.g., Brehm&Brehm, 1981) suggests that employees whoperceive leaders as beingbullyingorunsupportiveofindividualsafetymayretaliatebywithholdingvoluntaryextra-rolesafety behaviours, thereby restoring a sense of justice. Empirically, some evidence foralinkbetweenabusivesupervisorybehaviourandsafetymaybesuggestedbydatalinkingroleoverload(e.g., Barlingetal., 2002)orjobinsecurity(Probst, 2002)toadversesafetyoutcomes.More direct evidence emerges from a series of studies conducted by Mullen. First,in her qualitative investigation of why workers engaged in unsafe behaviours, Mullen(2004) identified abusive leadership as a key determinant of employee unsafebehaviour. Specifically, she identified instances in which managers coerced andintimidatedparticipantsintoperformingunsafetasks. Also, bothsupervisorsandcoworkerswouldteaseor mockindividualsdisplayinganexcessive concernforsafety andtherebydecreasedsafebehaviours.Inasubsequentstudy,MullenandFiset (2008) developedandempiricallyvalidatedamodel proposingthat abusivesupervisionnegativelyimpactsonemployeesafetyparticipationandpsychologicalhealth.Moreover,thelinkbetweenabusive supervisionandsafetyparticipationwasfullymediatedbyemployeeperceptionsofsafetyclimate.There are alsosome preliminary indications that supervisory justice may berelated to safety outcomes. Gatien, Fleming, Slaunwhite, and Wentzell (2009)reportedthat the way inwhichsupervisors respondtohealthandsafety issues(framed as distributive, procedural, and interpersonal fairness) emerged as apredictorof employeesafety-relatedperceptions insamples of constructioncraneoperators.Proceduralandinterpersonaljusticealoneaccountedformorethan50%ofthevarianceinperceptionsofsafetyclimate.Althoughmoredataareneededtodocument these associations, this must be cautiously regarded as a promising line ofenquiry.LeadershipdevelopmentasaninterventionAs the foregoing reviewattests, there is a substantial bodyof literature linkingorganizational leadership to occupational health psychology outcomes. Severalauthors (e.g., Nyberg et al., 2009) have noted the potential for leadershipdevelopment toconstituteaworkplacehealthintervention, andwehavereviewedseveralfieldexperiments(e.g.,McKee&Kelloway,2009;Mullen&Kelloway,2009;Slaunwhite et al., 2009) that suggest that this is aviable approachtoenhancingoccupationalhealthandsafetyinorganizations.Drawing on public health terminology, Hurrell (2005) delineated primary,secondary, andtertiarymodes of intervention. Primaryinterventions arefocusedonreducingoreliminatingthestressors (Hurrell, 2005; Quick, Quick, Nelson, &Hurrell,1997).Secondaryinterventionsfocusonchangingtheindividualsreactionstobeingexposedtoworkstressors. Finally, tertiaryinterventionsrepresentahealthe wounded approach in which the focus is to treat individuals who have developedWork&Stress 269strain reactions (Quick et al., 1997). In this context, leadership development appearstobe apromisingmeans of primaryinterventionthat is focuseddirectlyontheimprovement of workplaceconditionsthat leadtooccupational healthandsafetyoutcomes.Generallythoughttobethemosteffectiveapproachtodealingwithworkstress(Kelloway &Day, 2005), primary interventions can be divided into two basiccategories (1) psychosocial interventions; and (2) socio-technical interventions(Parkes &Sparkes, 1998). AsnotedbyHurrell (2005), psychosocial interventionsfocusprimarilyonhumanprocessesandpsychosocial aspectsoftheworksettingandaimtoreducestressbychangingemployeeperceptions of theworkenviron-ment (p. 624). In contrast, socio-technical interventions focus primarily on changestoobjective workconditions(Hurrell,2005,p.625).Asindicatedbyhisemphasis,thesedistinctions canbedifficult tomakeinpracticebecausesomeinterventionsinvolvebothobjectiveandsubjectivechanges.We suggest that leadership development is such an intervention encompassing,as it does, elements of bothpsychosocial andsocio-technical interventions. Theavailable evidence, for example, suggests that transformational leadershipaffectsindividual well-being by changing employees perceptions of their work (e.g., Arnoldet al., 2007; Nielsen et al., 2008), suggesting that enhancing leaders transformationalleadershipbehaviours wouldconstitute apsychosocial intervention. At the sametime, abusiveordestructiveleadershipcan, initself, beastressor(Kellowayetal.,2005), andinterventionsdesignedtoimproveleadershipwouldconstituteasocio-technicalintervention.Theeffectivenessofinterventionsaimedatenhancing leadershipHowever, one might choose to categorize development of leadership as anoccupational health psychology intervention, there is clear and unambiguousevidencethatleadershipdevelopment works. Thatis, theavailabledatasupportthesuggestionthatactivitiesdesignedtoenhanceleadershipinorganizationsdoinfact result inimprovedperceptions of leadership(seeforexample, Barlinget al.,1996; Kelloway et al., 2000; Mullen & Kelloway, 2009). Such activities have typicallyinvolvedtrainingintheformofworkshops(Dvir, Eden, Avolio, &Shamir, 2002),participationincoaching(Kombarakaran, Young, Baker, &Fernandes, 2008) orcombinationsofbothapproaches(e.g.,Barlingetal.,1996;Kellowayetal.,2000).Intheir recent meta-analysis, Avolio, Reichard, Hanna, Walumba, andChan(2009) provided a comprehensive review of the effectiveness of leadershipinterventions. Drawingondatafromover200studiesoveraperiodof morethan50years basedonavarietyof leadershiptheories, Avolioet al. (2009) reportedevidence that leadership interventions do in fact result in enhanced leadership. In 62of the studies considered, the interventionin questionwas the development ortraining of a leader (as opposedtothe assignment of a leader or an actor toportraying aparticular leadershipstyle). The data supporteda slightly strongereffect for developmental, as opposed to training, activities but overall resulted in theconclusion that leadership development was an effective intervention (correctedeffectsizesd.41to.48).270 E.K.KellowayandJ.BarlingThedesignofleadershipinterventionsIt is instructive to consider the nature of interventions designed to enhanceleadership. For the purposes of discussion, we will focus specifically on the relativelyshort-terminterventionsthat typicallycompriseformal trainingactivitieswith, orwithout, coaching or feedback. Longer term developmental activities (e.g., the use ofdevelopmental assignments, DeRue &Wellman, 2009) are excluded fromourdiscussionbecause, at present, thereis insufficient datalinkingsuchactivities tooutcomesrelevanttooccupationalhealthpsychology.Barling, Weber, andKelloway(1996)randomlyassignedninebranchmanagersof aregional banktotheexperimental group; and11managers toawaiting-listcontrol group. Managers all workedwithin the same geographic area but eachworkedinaseparatebranchof thebank. Managersassignedtotheexperimentalgroupparticipatedinaone-dayworkshopontransformational leadership. Adayafterthetraining, theleadersmetwithacoachwhoprovidedindividual feedbackbasedonemployeeratingsoftheleaderstransformationalleadershipstyle.Duringthe sessions, the emphasis was placed on the development of specific goals toimprove the managers transformational leadership. Subsequently, the managers metwiththe coachfor three follow-upsessions (one eachmonthfor three months)duringwhichgoalsandprogresswerereviewed. Managersassignedtothecontrolgroupreceivedneitherthetrainingnortheworkshop.In evaluating theintervention,theauthorsimplementedan assessmentbasedoncomparison of pre-test and post-test (i.e., three months following training) measures.Barlinget al. (1996) demonstratedthat (a) subordinateperceptions of managerstransformational leadership increased in the experimental but not the control group,(b) employee attitudes (i.e., affective commitment to the organization) were enhancedintheexperimental groupbutnotthecontrol groupand(c)measuresoffinancialperformancewereenhancedintheexperimentalbutnotthecontrolgroup.In a subsequent study, Kelloway, Barling, and Helleur (2000) attempted todisentangle the effects of workshopparticipationandfeedback/coaching ontheeffectivenessofleadershipdevelopment.Again usingapre-test,post-testdesign,40healthcaremanagerswererandomlyassignedtoeither aworkshopor atrainingcondition in a 22 factorial design. Results showed that either training or feedback/coachingwasaneffectivemeansofenhancingsubordinateperceptionsoftransfor-mational leadership but that the interaction of training and feedback did notincreasescoresabovethemaineffectsattributabletotheintervention.In their study of infantry soldiers, Dvir et al. (2002) also focused on thedevelopmentoftransformational leadership. Theyhadsevenindividualswhowererandomly assigned to the transformational leadership condition, which included fivedaysof training, includingroleplayingexercises, simulations, videopresentations,andgroup, peer, andtrainerfeedback. Participantsintheexperimental groupalsoparticipatedina3-hourboostersessionafterassignmenttoaleadershipposition.Comparisonwithdatafromacontrol groupsuggestedthat trainingparticipantsincreased both their knowledge of transformational leadership theory constructs andtheirtransformationalleadershipbehavioursasratedbysubordinates.Mullen andKelloway(2009) furtheradaptedthetraining programdevelopedbyBarling and colleagues (Barling et al., 1996; Kelloway et al., 2000) in their evaluationofasafety-specificmanagementtrainingintervention. Theyrandomlyassigned54Work&Stress 271healthcaremanagersfrom21organizationstooneofthreetraininginterventions(generalvs.safety-specific)orcontrolgroup(notraining).Thegeneraltransforma-tional leadershiptraining interventionconsistedof a half-day workshopfor themanagers (Barling, 1996; Kellowayet al., 2000) designedtofamiliarizemanagerswith the theory of transformational leadership and goal setting. Mullen andKellowaybeganbyhavingmangers identifythebehaviourof thebest andworstleaders they encountered. These characteristics were categorizedby the trainingfacilitator as being transformational, negative, or passive leadership behaviours.Managers were provided with an overviewof transformational leadership andfacilitators worked withparticipantstodevelopspecificbehaviouralgoals(Locke&Latham,1984)relatedtotransformationalleadership.The safety-specific training followed a similar format but the focus was on safetyissues throughout the training program. Both the general and safety-specifictransformational leadership training interventions were standardized in format,length, and method of delivery. The only difference between the two types of trainingwastheexperimentalmanipulation(generalvs.safety-specificcontent).Thecontrolgroupwasawaiting-listcontrol, themembersofwhichreceivedthesafety-specifictrainingattheconclusionofthestudy.Inthat study, MullenandKelloway(2009) examineddatafromboththe 54participantleadersand115 matchedrespondents inorderto assesstheeffectivenessof the training. They found that participation in training resulted in improvements inleaders ownsafetyattitudes, intentiontopromote safetyinthe workplace, andsafety-relatedself-efficacy.Datafromemployeesalsoshowedthattheemployeesofleaders in the safety-specific transformational leadership group reported (a)enhancedperceptions of their leaders safety-specific transformational leadership;(b) enhancedperceptions of safetyclimateandsafetyparticipationand(c) fewersafety-relatedeventsandinjuries.Although all of these interventions were effective, consideration of several designelements is instructive for those considering implementing similar designs. Wehighlight three of these for consideration: the intensity of the intervention, the need tospecifyinterveningvariables,andthelogisticaldifficultiesof evaluation.1.Intensityoftheintervention.Withregardtotheintensity ofintervention,wenotethat published intervention studies show considerable variation. Whereas Mullen andKelloway(2009)basedtheirstudyona3-hour(half-day)intervention, Dviretals(2002) intervention comprised five days of training. Barling et al. (1996) implementedcoaching and feedback sessions along with the training whereas Mullen andKelloway (2009) focusedsolelyontraining basedonthe observationthat bothtraining and feedback were effective means of enhancing leadership behaviours(Kelloway et al., 2000). The available data do not allow a clear determination of theoptimal lengthof trainingor thebest configurationof trainingandfeedback.However, itisclearthatwithinthebroadspectrumoforganizational interventions(e.g., Hurrell, 2005; Parkes &Sparkes, 1998), leadershipdevelopment is a cost-effectiveapproachresultinginminimaldisruptiontotheworkplace.2. Needtospecifyinterveningvariables. Second, we note that acharacteristic ofleadershipdevelopment initiatives is that theyare designedtoinduce change inorganizational leaders in order to change the attitudes, behaviours and experiences of272 E.K.KellowayandJ.Barlingemployees. Nielsen, Taris, and Cox (2010), this issue) define organizationalinterventions as science-based actions that target relatively large number ofindividuals.Unlikeotherinterventions(e.g.,safetytrainingandstressmanagement)that primarily target participants, the ultimate focus of leadership training istypicallyonindividualswhodonotparticipateinthetraining(i.e., onemployees).By definition, therefore, the effects of leadership training on employee outcomes areindirect, beingmediatedbyahostofpotential interveningvariables. Avolio(1999)makesthepointthatifaninterventionhasadirecteffectontaskperformanceitisprobably not dealing with leadership leadership effects manifest themselvesthroughinterveningvariablessuchasattitudesandmotivations.In a similar vein, the effects of leadership development on occupational health andsafetyarebynecessityindirect. Thedirect effect of leadershipdevelopment is toenhanceanindividualsleadershipbehaviours. Inthefirst instance, thesechangesmust be perceived by employees. These perceived changes must then influenceemployees attitudes and motivations. In turn these changes would be expected to bemanifestedinbehavioural change and, ultimately, changes inoutcome variables.Consideration of the mechanisms through which leadership development might affectoccupationalhealthandsafety outcomessuggeststheadvisability ofmeasuringandmodellingeachofthehypothesizedchangesinorder toallow aspecificationofhowleadershipdevelopmentaffectsoccupationalhealthand safety outcomes.Indirecteffectssuchasthosedescribedabovealsoimplyatimelagbetweenthetime of intervention and the effects of the intervention. Improvements or changes inhealthandsafetyresultingfromleadershipdevelopmentmaynotbemanifestedintheveryshortterm. Indeedithasbeencommontoassesstheeffectsofleadershipdevelopmentonlyafterthreemonthshaveelapsedpostintervention(Barlingetal.,1996; Mullen &Kelloway, 2009). Researchers are frequently enjoined to assessvariables in a longitudinal design at appropriate time lags (Edwards, 2008).Although our experience has been that a three-month time lag allows the detection ofsignificant effects of leadership development interventions, we have no theoretical orempirical means to determine what the appropriate time lag is for leadershipintervention research. This is a critical limitation in that missing the appropriate timelagmightleadtoinappropriateinferences arealeffectmaybemissedifthepost-test assessment ismadetoosoon(i.e., beforetheeffect hashadtimetomanifestitself). Similarly, a real effect may be missed if the assessment is made too late and theeffect dissipates prior to assessment. Again, the current data do not allowanassessmentofhowlonganyeffectofleadershipinterventionactuallylasts.3. Logistical difficultiesofevaluation. Finally, thecombinationof rigorousexperi-mental designandindirecteffectsresultsinlogisticallydifficultstudiestoconduct.To some extent, leadership development shares this characteristic with other forms ofinterventioninoccupational healthpsychology. Suchinterventionsarenotoriouslydifficulttoevaluate(Hurrell,2005;Kelloway,Day,&Hurrell,2008)andleadershipdevelopment is noexception. Aconsiderationof data reportedby Mullen andKelloway(2009,pp.258259)clearlyillustratesthelogisticaldifficultiesinvolved.Of the 172 participants[leaders] who received surveys,84 participantsresponded (48.8%responserate).Duetolistwisedeletionofmissingdataonthepre-testmeasure,asampleof60leaderswasobtained . . . Ofthe1,822healthcareworkers[employees]whoreceivedWork&Stress 273surveys, 494participants responded . . . At thepost-test, 269participants completedthesurvey . . . due tomatchingparticipant responses at boththe pre-test andpost-test andlistwisedeletion,115responseswereretained.Therequirementtomatchdatafrompre-testtopost-testandfromsubordinatestoleaderscoupledwiththenormal problemsofsurveyresponseandsubjectattritioncanresult ina large amount of data loss and, potentially, aninability toevaluate theintervention. Although these problems are not unique to leadership interventions, they areenhanced by the multiple levels of matching required for a rigorous evaluation of effects.DirectionsforfutureresearchDespitetheproblemsofconductingrigorousevaluationofleadershipdevelopmentas an occupational health psychology intervention, we have not abandoned hope forpotential for future research in this area. Rather, we begin our consideration of futureresearchinitiativeswithacall formoreinterventionstudiesfocusedonleadershipdevelopment. We believe that the datasupportingalinkbetweenorganizationalleadershipandoccupationalhealthpsychology outcomesareunequivocal.Further-more,theavailabledatasuggestthatleadershipdevelopmentprovidesoccupationalhealthpsychologistswithapragmaticandeffectivetool withwhichtoaffecttheseoutcomes. All interventions aredifficult toevaluate, andtheparticular problemsassociated with leadership interventions dictate the use of strong interventions likelyto produce the intended effects, careful modelling of anticipated effects, and arigorous attentiontodatacollection. We stronglysuggest that theavailabledatawarrant the conductingof more, not fewer, leadershipdevelopment initiatives inoccupationalhealthpsychology.Afocus onleadershipinterventions may mitigate some of the difficulties inimplementingorganizational-levelstressorwellnessinterventions.Nielsen,Randall,Holten, andGonzalez(2010), thisissue)notethatoccupational healthpsychologyinterventionsareseenassomethingseparatefromrunningthedailybusinessandensuringhighperformance.Theycallformoreattentiontohowwemightintegrateoccupational health psychology interventions with the normal functioning oforganizations.Leadershipdevelopmentactivitiesprovideonesuchavenue;theyarewell-acceptedinindustryasameansofincreasingperformanceandorganizationaleffectiveness (see for example, Barling et al., 1996) and have clear links to individualhealth and well-beingSecond, wehavenotedthroughout this reviewthat leadershipdevelopment isuniqueinthatthefocusisonchangingtheleadersinordertochangeemployees.However, we also note here the possibility for leadership development to enhance thehealth and safety of the leaders themselves. Mullen and Kelloway (2009), for example,found that leaders in the safety-specific leadership training condition reportedenhanced safety attitudes, self-efficacy, and intent to promote safety. It is conceivable,althoughtheauthorsdidnottest thehypothesis, thatinenhancingleaderssafetyattitudes, MullenandKelloway(2009) enhancedleaders ownsafetybehaviours.Similarly, whenleadershipdevelopmentresultsinleadershavingagreatersenseoftheir own self-efficacy (arguably a dimension of mental health, Warr, 1987) this mayresult in a greater sense of well-being for the leaders themselves. Examining the effects274 E.K.KellowayandJ.Barlingofleadershipdevelopmentonthehealthandsafetyoftheleadersthemselvesisaninteresting andpotentiallyfruitfularea of future enquiry.Finally, one interesting question that arises when considering leadershipinterventionsistheorganizationalleveloftheintervention.Mostofthestudieswehave reviewed have focused on employee perceptions of the behaviour of theirimmediate supervisor. However, it is plausible that interventions aimedat seniormanagementmightalsobeeffective. Avolioetal. (2009)wereabletoaddressthisquestionintheirmeta-analysisandfoundstrongereffectsforleadershipinterven-tions at the level of the direct supervisor level (effect size: d.69; .71) than at middle(ds.46, .51) or high (d.51) levels of management. Although these findingssuggesttheadvisability of focusingattheimmediatesupervisorylevel,thenon-zeroeffects for higher levels of management also suggest the potential for a morecomprehensiveinterventionthatcrossesmultiplelevelsoftheorganizationsleader-ship structure. Flin (2003) has advanced a similar argument with respect tooccupational safety. She argues that supervisors, mid-level managers, andseniormanagersall haveadistincteffectonemployeesafety. Moreover, sherecommendstheregularassessmentofseniormanagerscommitmenttosafety.ConclusionOur reviewof the existing literature suggests that sufficient data have nowaccumulatedtoallowtheunambiguousconclusionthat organizational leadershipisrelatedto,andpredictiveof, healthandsafety-relevantoutcomesinemployees. Moreover, webelieve that asmall but growingbodyof literature supports the effectiveness ofleadership development as a means of positively influencing these outcomes. Pursuingthis suggestionwill, webelieve, substantiallyadvanceour knowledgeof not onlyleadership development but also other areas of occupational health psychology.ReferencesArnold, K.A., Turner, N., Barling, J., Kelloway, E.K., &McKee, M.C. (2007). Transforma-tional leadershipandpsychological well-being: Themediatingroleof meaningful work.JournalofOccupationalHealthPsychology,12(3),Jul.,193203.Aryee, S., Chen, Z.X., Sun, L.Y., &Debrah, Y.A. (2007). Antecedents andoutcomes ofabusivesupervision:Atrickledownmodel.JournalofAppliedPsychology,92,191201.Ashforth, B.E. (1997). Petty tyranny in organizations. A preliminary examination ofantecedantsandconsequences.CanadianJournalofAdministrativeSciences,14,126140.Avolio, B.J. (1999). Full range leadership development: Building the vital forces in organizations.ThousandOaks,CA:SagePublications.Avolio, B.J., Reichard, R.J., Hannah, S.T., Walumba, F.O., &Chan, A., (2009). Ameta-analytic review of leadership impact research: Experimental and quasi-experimental studies.TheLeadershipQuarterly,20,764784.Bamberger, P.A., &Bacharach, S.B. (2006). Abusivesupervisionandsubordinateproblemdrinking: Taking resistance, stress and subordinate personality into account. HumanRelations,59(6),723752.Barling, J., Christie, A., & Hoption, A. (2011). Leadership. In S. Zedeck (Ed.), APA handbookofindustrial andorganizational psychology. Vol 1:Buildinganddevelopingtheorganization(pp.183240).Washington,DC:AmericanPsychologicalAssociation.Work&Stress 275Barling, J., Loughlin, C., &Kelloway, E. (2002). Development andtest of amodel linkingsafety-specic transformational leadership and occupational safety. Journal of AppliedPsychology,87,488496.Barling, A.J., Weber, T., &Kelloway, E.K. (1996). Effects of transformational leadershiptraining onattitudinal andnancial outcomes: Aeldexperiment. Journal of AppliedPsychology,81,827832.Bono, J.E., Foldes, H., Vinson, G., &Muros, J.P. (2007). Workplaceemotions: Theroleofsupervisionandleadership.JournalofAppliedPsychology,92(5),13571367.Brehm, J., & Brehm, S. (1981). Psychological reactance: A theory of freedomand control. NewYork:AcademicPress.Brondolo, E., Rieppi, R., Erickson, S.A., Bagiella, E., Shapiro, P.A., McKinley, R.P. et al.(2003). Hostility, interpersonal interactions, and ambulatory blood pressure. PsychosomaticMedicine,65,10031011.Burke, M.J., & Sarpy, S.A. (2003). Improving worker safety and health through interventions.InD.A. Hofmann&L.E. Tetrick(Eds.), Healthandsafetyinorganizations: Amultilevelperspective(pp.5690).SanFrancisco:Jossey-Bass.Burton,J.,&Hoobler,J.(2006).Subordinateself-esteemandabusivesupervision.JournalofManagerialIssues,18(3),340355.Cree, T., & Kelloway, E.K. (1997). Responses to occupational hazards: Exit and participation.JournalofOccupationalHealthPsychology,2,304311.Colquitt, J.A., Conlon, D.E., Wesson, M.J., Porter, C.O.L.H., & Ng, K.Y. (2001). Justice at themilleninium: a meta-analytic review of 25 years of organizational justice research. Journal ofAppliedPsychology,86,425445.Day,R.C.,&Hamblin,R.L.(1964).Someeffectsofcloseandpunitivestylesofsupervision.TheAmericanJournalofSociology,69,499510.DeRue, D.S., & Wellman, N. (2009). Developing leaders via experience: The role ofdevelopmentalchallenge,learningorientationandfeedbackavailability.JournalofAppliedPsychology,94,859875.Duffy, M.K., Gangster, D., &Pagon, M. (2002). Social undermining in the workplace.AcademyofManagementJournal,45,331351.Dvir, T., Eden, D., Avolio, B.J., & Shamir, B. (2002). Impact of transformational leadership onfollower development and performance: Aeld experiment. Academy of ManagementJournal,45,735744.Edwards, J. (2008). Toprosper, organizational psychologyshould. . . . overcomemethodolo-gicalbarrierstoprogress.JournalofOrganizationalBehavior,29,469491.Elovainio, M., Kivimaki, M., Puttonen, S., Lindholm, H., Pohjonen, T., & Sinervo, T. (2006).Organizational injustice and impaired cardiac regulation among female employees.OccupationalandEnvironmentalMedicine,63,141144.Elovainio, M., Leino-Arjas, P., Vahtera, J., &Kivimaki, M. (2006). Justice at workandcardiovascularmortality: Aprospectivecohort study. Journal of PsychosomaticResearch,61,271274.Eriksen, W. (2005). Work factors as predictors of smoking relapse in nurses aides.InternationalArchivesofOccupationalandEnvironmentalHealth,79(3),244250.Ferrie, J.E., Head, J.A., Shipley, M.J., Vahtera, J., Marmot, M.G., &Kivimaki, M. (2006).Injustice at work and incidence of psychiatric morbidity: The Whitehall II study.OccupationalandEnvironmentalMedicine,63,443450.Ferris, G.R., Zinko, R., Brouer, R.L., Buckley, M.R., &Harvey, M.G. (2007). Strategicbullying as a supplementary, balanced perspective on destructive leadership. The LeadershipQuarterly,18(3),195206.Flin, R. (2003). Danger: Menatwork. HumanFactorsandErgonomicsinManufacturing&ServiceIndustries,13,261268.Flin, R., Mearns, K., OConnor, P., &Bryden, R. (2000). Measuring safety climate:Identifyingthecommonfeatures.SafetyScience,34,177192.276 E.K.KellowayandJ.BarlingGatien, B., Fleming, M., Slaunwhite, J., & Wentzell, N. (2008, November). An investigation intothe relationship between organizational justice & safety climate perceptions. Paper presented attheAnnualEuropeanAcademyofOccupationalHealthPsychology,Valencia,Spain.Gerstner,C.R.,& Day,D.V.(1997). Meta-analyticreview of leader-member exchangetheory:Correlatesandconstructissues.JournalofAppliedPsychology,82,827844.Gilbreath, B. (2004). Creatinghealthyworkplaces: The supervisors role. InC. Cooper &I. Robertson (Eds.), International review of industrial and organizational psychology, volume 19.Chichester,UK:JohnWiley.Grandey, A.A., Kern, J., &Frone, M. (2007). Verbal abusefromoutsidersversusinsiders:Comparingfrequency,impactonemotionalexhaustion,andtheroleofemotionallabour.JournalofOccupationalHealthPsychology,12,6379.Halbesleben, J.R.B. (2006). Sources of social support and burnout: A meta-analytic test of theconservationofresourcesmodel.JournalofAppliedPsychology,91(5),11341145.Harvey,M.G.,Buckley,M.R.,Heames,J.T.,Zinko,R.,Brouer,R.L.,&Ferris,G.R.(2007).A bully as an archetypal destructive leader. Journal of Leadership & Organizational Studies,14(2),117129.Harvey, P., Stoner, J., Hochwarter, W., & Kacmar, C. (2007). Coping with abusive bosses: Theneutralizing effects of ingratiationandpositive affect onnegative employee outcomes.LeadershipQuarterly,18(3),264273.Hauge, L.J., Skogstand, A., &Einarsen, S. (2007). Relationships betweenstressful workenvironmentsandbullying: Resultsof alargerepresentativestudy. Work&Stress, 21(3),220242.Heaphy, E.D., &Dutton, J. (2008). Positive social interactions and the human body atwork: Linkingorganizationsandphysiology. TheAcademyof Management Review, 33(1),137162.Hofmann,D.A.,&Morgeson,F.P.(1999).Safety-relatedbehaviorasasocialexchange:Theroleofperceivedorganizational supportandleader-member-exchange. Journal ofAppliedPsychology,84(2),286296.Hofmann, D.A., Morgeson, F.P., &Gerras, S. (2003). Climate as a moderator of therelationshipbetweenleader-memberexchangeandcontent speciccitizenship. Journal ofAppliedPsychology,88,170178.Hurrell,J.J.(2005).Organizationalstress interventions.InJ.Barling,E.K.Kelloway,&M.R.Frone (Eds.), Handbook of work stress (pp. 623646). Thousand Oaks, CA: SagePublications.Karlin, W., Brondolo, E., &Schwartz, J. (2003). Workplacesocial supportandambulatorycardiovascular activity in NewYork City trafc agents. Psychosomatic Medicine, 65,67176.Keashly, L. (1998). Emotional abuse inthe workplace: Conceptual andempirical issues.JournalofEmotionalAbuse,1,85117.Kelloway,E.K.,Barling,J.,&Helleur,J.(2000).Enhancingtransformationalleadership:Therolesoftrainingandfeedback.TheleadershipandOrganizationalDevelopmentJournal,21,145149.Kelloway, E.K., &Day, A. (2005). Building healthy workplaces. What we knowsofar.CanadianJournalofBehavioralScience,34(7),223235.Kelloway, E.K., Day, A., &Hurrell, J.J. (2005). Workplace interventions for occupationalstress. InK. Naswall, J. Hellegren, &M. Sverke (Eds.), The individual inthe changingworking life.Cambridge:CambridgeUniversityPress.Kelloway, E.K., Mullen, J.E., & Francis, L. (2006). Divergent effects of passive andtransformational leadership on safety outcomes. Journal of Occupational Health Psychology,11,7686.Kelloway, E.K., Sivanathan, N., Francis, L., &Barling, J. (2005). Poor leadership. InJ.Barling.,E.K.Kelloway,&M.Frone(Eds.),Handbook ofworkplacestress(pp.89112).ThousandOaks,CA:SagePublications.Work&Stress 277Kelloway, E.K., Teed, M., & Prosser, M. (2008). Leading to a healthy workplace. In A. Kinder,R. Hughes, &C.L. Cooper (Eds.), Employee well-being support: Aworkplace resource(pp.2538).Chichester:JohnWiley.Kiecolt-Glaser, J.K., McGuire, L., Robles, T.F., &Glaser, R. (2002). Psychoneuroimmunol-ogy: Psychological inuencesonimmunefunctionandhealth. Journal of ConsultingandClinicalPsychology,70,537547.Kivimaki, M., Elovainio, M., Vahtera, J., & Ferrie, J.E. (2003). Organisational justice and healthofemployees:Prospectivecohortstudy.OccupationalandEnvironmentalMedicine,60,2734.Kivimaki, M., Ferrie, J.E., Brunner, E., Head, J., Shipley, M.J., Vahtera, K. et al. (2005).Justice at work and reduced risk of coronary heart disease among employees: The WhitehallIIstudy.ArchivesofInternalMedicine,165,22452251.Kombarakaran, F.A., Young, J.A., Baker, M.N., & Fernandes, P.B. (2008). Executivecoaching:Itworks!ConsultingPsychologyJournal:PracticeandResearch,60,7890.Kuoppala,J.,Lamminpaa,A.,Liira,J.,&Vainio,H.(2008).Leadership,job well-being,andhealth effects: A systematic review and meta-analysis. Journal of Occupational andEnvironmentalMedicine,60(8),904915.Lee, R.T., & Ashforth, B.E. (1996). A meta-analytic examination of the correlates of the threedimensionsofjobburnout.JournalofAppliedPsychology,81,123133.Locke, E.A., &Latham, G.P. (1984). Goal Setting: AMotivational Technique that works.EnglewoodCliffs,NJ:PrenticeHall.McKee, M., &Kelloway, E.K. (2009). Leadingtowellbeing. Paperpresentedat theannualmeetingoftheEuropeanAcademyofWorkandOrganizational Psychology, SantiagodeCompostella,Spain.McKee, M., Kelloway, E.K., Driscoll, C., &Kelley, E. (2009). Workplace spiritualityandindividualwell-being:Anempiricaltest.Manuscriptsubmittedforpublication.Moyle, P. (1998). Longitudinal inuences of managerial support onemployee well-being.WorkandStress,12(1),2949.Mullen, J.E. (2004). Investigatingfactorsthat inuencesafetybehaviorat work. Journal ofSafetyResearch,35,275285.Mullen,J.E.(2005).Testingamodelofemployeewillingnesstoraisesafetyissues.CanadianJournalofBehavioralSciences,37(4),273282.Mullen, J.E., &Fiset, J. (2008). Theeffectsofabusivesuspensiononemployeeoccupationalhealth and safety outcomes. Paper presented at the 9th World Conference an InjuryPreventionandSafetypromotion.Merida,Mexico.Mullen, J.E., & Kelloway, E.K. (2009). Safety leadership: A longitudinal study of the effects oftransformational leadership on safety outcomes. Journal of Occupational and OrganizationalPsychology,82,253272.Mullen, J., & Kelloway, E.K. (2011). Occupational health and safety leadership. InJ. Campbell Quick&L.E. Tetrick(Eds.), Handbookof occupational health psychology(2nded.,pp.357372).Washington,DC:AmericanPsychologicalAssociation.Nielsen, K., Randal, R., Yarker, S., &Brenner, S. (2008). Theeffects of transformationalleadership on followers perceived work characteristics and psychological well-being:Alongitudinalstudy.Work&Stress,22,1632.Nielsen, K., Randall, R., Holton, A., &Gonzalez, E.R. (2010). Conductingorganizational-leveloccupationalhealthinterventions:Whatworks?Work&Stress,24,234259.Nielsen, K., Taris, T.W., &Cox, T. (2010). Thefutureoforganizational-level interventions:Addressingthechallengesoftodaysorganizations.Work&Stress,24,219233.Nielsen, K., Yarker, J., Brenner, S., Randall, R., &Borg, V. (2008). The importance oftransformational leadership style for the well-being of employees working with older people.JournalofAdvancedNursing,63,465475.Nyberg, A.,Alfredsson,L., Theorell,T.,Westerlund,H.,Vahtera, J.,& Kivimaki, M. (2009).Managerial leadershipand ischaemic heart disease among employees: The Swedish WOLFstudy.OccupationalandEnvironmentalMedicine,66,5155.278 E.K.KellowayandJ.BarlingOffermann, L.R., &Hellmann, P.S. (1996). Leadership behavior and subordinate stress:A3608view.JournalofOccupationalHealthPsychology,1,382390.Parkes, K.R., &Sparkes, T.J. (1998). Organizational interventionstoreduceworkstress:Aretheyeffective?Areviewof theliterature. Oxford, UK: Universityof Oxford, HealthandSafetyExecutive,ContractReportNo.193/198.Probst, T.M. (2002). Layoffs and tradeoffs: Production, quality, and safety demands under thethreatofjobloss.JournalofOccupationalHealthPsychology,7,211220.Quick, J.C., Quick, J.D., Nelson, D.L., & Hurrell, Jr., J.J. (1997). Preventive stress managementinorganizations.Washington,DC:APABooks.Rooney, J., &Gottlieb, B. (2007). Development and initial validation of a measure ofsupportiveandunsupportivemanagerial behaviors. Journal ofVocational Behavior, 71(2),186203.Skakon, J., Nielsen, K., Borg, V., & Guzman, J. (2010). Are leaders wellbeing, behaviours andstyle associated with the affective wellbeing of their employees? A systematic review of threedecadesofresearch.Work&Stress,24,147139.Slaunwhite, J.M., Smith, S.M., Fleming, M., Strang, R., &Lockhart, C. (2009). Increasingvaccinationratesamongcareworkers: Usingunitchampions asamotivator. CanadianJournalofInfectionControl,24,159164.Sosik, J., &Godshalk, V. (2000). Leadershipstyles, mentoringfunctionsreceived, andjob-related stress: Aconceptual model and preliminary study. Journal of OrganizationalBehavior,21,365390.Sparr,J.L.,&Sonnentag,S.(2008).Fairnessperceptionsofsupervisorfeedback,LMX,andemployeewell-beingat work. EuropeanJournal of WorkandOrganizational Psychology,17(2),198225.Stogdill, R.M. (1974). Handbook of leadership: A survey of the literature. New York: Free Press.Teed, M., Kelloway, E.K., &Mullen, J.E. (2008). Young workers safety: The impact ofinconsistentleadership.PaperpresentedatWork,Stress,andHealthConference,Washing-ton,DC.Tepper, B.J. (2000). The consequences of abusive supervision. Academy of ManagementJournal,43,178190.Tepper, B.J. (2007). Abusive supervision in work organizations: Review, Synthesis, andresearchagenda.JournalofManagement,33(3),261289.Tepper, B.J., Duffy, M.K., &Shaw, J.D. (2001). Personalitymoderatorsof therelationshipbetweenabusivesupervisionandsubordinates resistance. Journal of AppliedPsychology,86(5),974983.VanDierendonck, D., Haynes, C., Borrill, C., &Stride, C. (2004). Leadershipbehaviorandsubordinatewell-being.JournalofOccupationalHealthPsychology,9(2),165175.Wager, N., Feldman, G., &Hussey, T. (2003). Theeffect onambulatorybloodpressureofworking under favourably and unfavourably perceived supervisors. Occupational andEnvironmentalMedicine,60,468474.Warr,P.B.(1987).Work,unemploymentandmentalhealth.Oxford:ClarendonPress.Whiteman,J.,Snyder,D.,&Ragland,J.(2001).The valueofleadershipinimplementingandmaintaining a successful health promotion program in the Naval Surface Force, US PacicFleet.AmericanJournalofHealthPromotion,15(6),437440.Yukl,G.(2006).Leadershipinorganizations,6thEdition.UpperSaddleRiver,NJ:Pearson.Zohar, D. (1980). Safety climate in industrial organizations: Theoretical and appliedimplications.JournalofAppliedPsychology,65,96102.Zohar, D. (2002a). Modifyingsupervisorypracticestoimprovesubmitsafety: Aleadership-basedinterventionmodel.JournalofAppliedPsychology,87(1),156163.Zohar, D. (2002b). The effects of leadership dimensions, safety climate, and assigned prioritiesonminorinjuriesinworkgroups.JournalofOrganizationalBehavior,23,7592.Zohar, D., &Tenne-Gazit, O. (2008). Transformational leadershipandgroupinteractionasclimate antecedents: Asocial network analysis. Journal of Applied Psychology, 93(4),744757.Work&Stress 279Copyright of Work & Stress is the property of Taylor & Francis Ltd and its content may not be copied oremailed to multiple sites or posted to a listserv without the copyright holder's express written permission.However, users may print, download, or email articles for individual use.