lable at ScienceDirect

Social Science & Medicine 108 (2014) 120e127

Contents lists avai

Social Science & Medicine

journal homepage: www.elsevier .com/locate/socscimed

External factors affecting decision-making and use of evidencein an Australian public health policy environment

Pauline Zardo a,b,*, Alex Collie a,b, Charles Livingstone c

aMonash University, Department of Epidemiology and Preventive Medicine, Australiab Institute for Safety, Compensation and Recovery Research, AustraliacMonash University, Department of Health Social Sciences, Australia

a r t i c l e i n f o

Article history:Received 8 February 2013Received in revised form28 January 2014Accepted 28 February 2014Available online 1 March 2014

Keywords:PolicyDecision-makingResearchUseAustraliaTranslationPublic health

* Corresponding author. Level 11, 499 St Kilda Rd.,E-mail address: pauline.zardo@monash.edu (P. Za

http://dx.doi.org/10.1016/j.socscimed.2014.02.0460277-9536/� 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

This study examined external factors affecting policy and program decision-making in a specific publichealth policy context: injury prevention and rehabilitation compensation in the Australian state ofVictoria. The aim was twofold: identify external factors that affect policy and program decision-makingin this specific context; use this evidence to inform targeting of interventions aimed at increasingresearch use in this context. Qualitative interviews were undertaken from June 2011 to January 2012 with33 employees from two state government agencies. Key factors identified were stakeholder feedback andaction, government and ministerial input, legal feedback and action, injured persons and the media. Theidentified external factors were able to significantly influence policy and program decision-makingprocesses: acting as both barriers and facilitators, depending on the particular issue at hand. The fac-tors with the most influence were the Minister and government, lawyers, and agency stakeholders,particularly health providers, trade unions and employer groups. This research revealed that in-terventions aimed at increasing use of research in this context must target and harness the influence ofthese groups. This research provides critical insights for researchers seeking to design interventions toincrease use of research in policy environments and influence decision-making in Victorian injury pre-vention and rehabilitation compensation.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Internationally there is growing interest in increasing researchevidence-informed decision-making in public health policy andpractice (Nutbeam and Boxall, 2008; Rychetnik et al., 2012). Publichealth research is expected to contribute to improved public healthoutcomes by providing research evidence that can enhancedecision-makers understanding of public health issues, potentialsolutions, their costs and benefits and the likelihood of effective-ness (Brownson et al., 2009; Killoran and Kelly, 2010). However it isestimated that only 8e15% of efforts to translate research intohealth policy and practice are effective (Best and Holmes, 2010). It isalso argued that research relevant to public health policy issues islacking (Carter, 2010; Green et al., 2009). Many factors can act asbarriers to use of research in public health policy decision-making(Mitton et al., 2007; Orton et al., 2011). It is recognised that there isa need to build capacity in both public health policy and academic

Australia.rdo).

research environments to increase and support research evidenceinformed decision-making in public health (Haynes et al., 2012;Martin et al., 2011).

While there have been many interventions, tools and strategiesemployed to increase use of research in public health policydecision-making; evidence of their effectiveness is limited(Grimshaw et al., 2012; Orton et al., 2011). To increase the likeli-hood of achieving intended outcomes and to effectively contributeto the evidence base, interventions aimed at increasing use ofresearch must be informed by theory guided context-specificresearch (Dobrow et al., 2006; Greenhalgh et al., 2004). Suchresearch evidence can inform intervention design and imple-mentation, ensuring that particular factors affecting use of evi-dence in a specific public health policy context are addressed(Glasgow and Emmons, 2007; Michie et al., 2011).

Public health systems are diverse and complex. Public healthpolicy decision-makers are faced with the challenging task ofdeveloping and implementing policies and programs that areeffective at the health system, or population level (Brownson et al.,2009; Rychetnik et al., 2012). Public health policy development isaffected by individual level factors, organisational level factors and

P. Zardo et al. / Social Science & Medicine 108 (2014) 120e127 121

also by factors external to and not in the control of individualdecision-makers or organisations (Damschroder et al., 2009; Laviset al., 2012). There is a substantial body of knowledge around fac-tors that affect use of research in health policy contexts, the ma-jority of which is focused at the individual and organisational level(Mitton et al., 2007). However, more research is needed to informresearch translation interventions in specific public health policyenvironments (Grimshaw et al., 2012; Orton et al., 2011).

Workplace and transport injury and illness prevention andrehabilitation are significant public health policy issues interna-tionally and in Australia. Public injury and illness compensationschemes run by Australian state governments have played a centralrole in supporting the recovery of those injured, ill and disabled dueto a workplace or transport incident. In 2009 in Australia aloneapproximately 638,400 residents suffered a work-related injury orillness, estimated to cost in excess of $60 billion (SafeWork Australia,2012a, 2012b). The cost of transport injuries in Australia is estimatedto be $27 billion per annum (Australian Government Department ofInfrastructure and Transport, 2011). In 2008/09 there were 53,406persons hospitalised due to land transport injury in Australia(Australian Institute of Health andWelfare, 2012). Whilst there havebeen great achievements in reducing rates of workplace and trans-port fatality, injury and illness and improving rehabilitation and re-covery outcomes in Australia, many challenges remain. Workplaceand transport injuries that were the main focus of prevention andcompensation policy and programs in the past have decreased and/or plateaued. Complex health issues such as stress andmental illness,that lack evidence of effective system level solutions, are emerging(International Labour Organization, 2010).

Research has demonstrated that people receiving compensationfor injury and illness can experience worse health outcomes thanthose not receiving compensation (Gabbe et al., 2007; Murgatroydet al., 2011). Other research has shown that compensation policychange at themacro, or legislative level, can lead to improved healthoutcomes (Cameron et al., 2008; Cassidy et al., 2000). Thesefindingssuggest that there is a critical need to explore the factors that affectpolicy and program decision-making in the injury prevention andrehabilitation compensation (IP&RC) policy context; as the policydecisions made can have unintended consequences, affectinghealth outcomes both positively and negatively. To facilitate andsupport increased use of research inworkplace and transport IP&RCpolicy and program decision-making, there is first a need to un-derstand the factors that affect decision-making in this context.

This study sought to: identify external factors that affect policyand program decision-making in workplace and transport IP&RC inVictoria, Australia; and describe how these factors could be tar-geted and tailored for in the design and implementation of in-terventions aimed at increasing research use in this context. Thefocus of this study was on factors that affect decision-makinggenerally; in contrast to factors that affect use of research evi-dence specifically. Understanding how external factors affect‘everyday’ decision-making in this context is necessary to under-standing how use of research evidence can be contextualisedwithin broader decision-making processes in this context.

This study forms part of a larger study also examining organ-isational factors that affect decision-making generally and organ-isational and individual factors that affect use of research evidencespecifically in the Victorianworkplace and transport IP&RC context.

2. Methods

2.1. Context

The Victorian WorkCover Authority and the Transport AccidentCommission (referred to here as Agency 1 and Agency 2,

respectively) are state government statutory authorities respon-sible for workplace and transport IP&RC in the Australian state ofVictoria (Australia’s system of government is a federation with sixstates; elected federal and state governments and their ministersmanage portfolios and related departments and authorities). Assuch, they play a significant role in the public health system;managing the public insurance schemes that pay compensation forthe costs of workplace or transport injury treatment and rehabili-tation (Personal Injury Education Foundation, 2011).

These Agencies have the specific mandate of administering theTransport Accident Act, 1986 (VIC), the Occupational Health andSafety Act, 2004 (VIC) and the Accident Compensation Act, 1985 (VIC)and other related pieces of legislation. The Agencies are involved inthe development and revision of such legislated Acts and relatedRegulations, which are then subject to the Victorian parliamentaryprocess to be ratified. They also develop policies and programs thatinform, guide and support the interpretation and implementationof these Acts and Regulations, which are approved within theAgencies. For example, this includes but is not limited to: thedevelopment of policies regarding which types of injury andrehabilitation treatments and services will be compensated; pol-icies regarding best practice workplace safety approaches; and thedevelopment of targeted programs to improve health, safety andrehabilitation outcomes for particular groups of workers andinjured persons.

These agencies have both a public health and financial imper-ative. They must balance the challenging task of preventing injuryand compensating individuals for the costs of workplace andtransport injury and illness treatment and rehabilitation, whilst atthe same timemaintaining and sustaining a viable public insurancescheme (Transport Accident Commission, 2012; WorkSafe Victoria,2012).

The two agencies also have important differences. Agency 1 isprimarily focused on workplace injury and illness prevention andreturn to work after injury or illness. Agency 1 is responsible forenforcement of the OHS Act and Regulations, through workplaceinspectors, legal review and public prosecution. Agency 2 is focusedon effective rehabilitation for those injured in a transport accident.Injury prevention policy and program development and enforce-ment for Agency 2 are undertaken in partnership with the VictorianPolice, VicRoads (the transport licensing and registration stategovernment authority) and the Department of Justice (PersonalInjury Education Foundation, 2011).

The way that compensation claims are managed by theseAgencies also differs. Agency 1 contracts out their insurance claimsmanagement to private insurers. Agency 2 undertakes claimsmanagement in-house (Personal Injury Education Foundation,2011).

2.2. Sampling and recruitment

Monash University Human Research Ethics Committeeapproved this study. Potential participants were identified throughreview of organisational charts. Employees from business unitswhose work actively involved the development, implementationand/or evaluation of strategy, policy, programs and projects wereincluded.

Representation was sought from senior managers, managersand non-managers. A minimum of four participants from each ofthese role levels, in each Agency, were sought from the potentialparticipant pool as it has been argued that qualitative data satu-ration can be achieved with approximately 12 participants (Guestet al., 2006).

Invitations to participate were sent from internal HR represen-tatives via email to a sample of eight individuals at each role level,

Table 1Agency 1 ¼ Victorian WorkCover Authority; Agency 2 ¼ Transport Accident Com-mission; Cert ¼ Certificate; Under-grad ¼ Undergraduate Degree (including Hon-ours); Post-grad ¼ Post Graduate Degree (Masters, PhD, etc).

Allparticipants

Seniormanagers

Managers Non-managers

Total participants 33 17 9 7Agency (1, 2) 14, 19 7, 10 6, 3 6, 1Age yrs

(median, range)42, 27e66 45, 37e65 42, 27e66 36, 27e56

Sex (female, male) 15, 18 7, 10 3, 6 5, 2Education level

(Cert, Under-grad,post-grad)

2, 13, 18 0, 8, 9 0, 3, 6 2, 2, 3

Employment inagency (yrs)(median, range)

5, 1.5e26 10, 1.5e26 4, 1.6e22 4, 2e13

Employment incurrent role (yrs)(median, range)

2.5, 0.02e25 2.66, 0.02e25 2, 0.33e6 3, 0.5e9

Employment ingovernmentroles (yrs)(median, range)

8.5, 1.5e32 11, 1.5e32 6.5, 1.66e22 8, 2.5e20

P. Zardo et al. / Social Science & Medicine 108 (2014) 120e127122

beginning May 2011. The invitation included: Participant Informa-tion Sheet; Consent Form; and ‘donot reply to this email’ instructionto ensure the Agencies remained unaware of participation. Partici-pation was self-selected. The research team waited two weeks forreplies from potential participants. The process then started overagain for the role level group where the quota was not filled.

In August and September 2011, the HR representatives of bothAgencies were unable to continue supporting recruitment. Tominimise the burden on Agency staff, project key contacts sent theemail invitation, in one round, to all senior managers in eachAgency, and to all staff in particular teams from selected areas thathad large numbers of managers and non-managers. The alternativerecruitment method also received Ethics approval. The finalrecruitment email round was sent in October 2011. Participantscontinued to self-select until January 2012.

2.3. Interviews and analysis

In total 33 in-depth, semi-structured qualitative interviewswere undertaken. Interview schedule development was informedby a broad reading of the literature around use of research evidencein policy; and in particular by Michie et al.’s (2005) domainframework. Michie et al.’s (2005) framework is based on systematicreview of psychological and social-psychological theory relevant tobehaviour change regarding use of evidence. This framework wasused to identify domains relevant to the research question thatwould be important to explore in the interviews. Whilst thisframework was designed to address factors affecting clinicians’ useof evidence, the domains represent concepts found to affect policymaker use of evidence, including skills, environmental context,resources etc. (Mitton et al., 2007; Orton et al., 2011).

The interview schedule was piloted with two Managers fromAgency 1. The semi-structured interviews were conducted over thephone and electronically recorded; most often lasting between40 min and 1 h. Interviews were transcribed by a professionaltranscription service and checked by the lead author.

The results presented in this paperarebasedon thematic analysesof participants’ responses to the interview question: ‘what externalfactors affect decision-making within your organisation? The inter-view data generated from this question were thematically analysed.The lead author undertook the analysis. Interviews were openlycoded to represent concepts found in the data. Coded datawere thenanalysed inductively to identify emerging themes (i.e. categories andsub categories of coded concepts) (Patton, 2002). Concepts thatsignificantly overlappedweremerged. Data saturationwas achievedat approximately 15 interviews.Despite data saturation at 15, allwhoself-selected to be part of the study were interviewed so as not toexclude interested decision-makers from participation. NVivo 9 wasused to organise the data. Following review by the research team;codes and themes were discussed and debated. The key themes andtheir related concepts are described below.

3. Results

3.1. Participant demographics

Detailed participant demographics are provided in Table 1.Participants were asked to describe their role and identify the

main focus of their role. Five participants indicated that theyworked on policy development most of the time, 26 indicated theyworked on developing programs and projects most of the time andthree indicated they worked equally on both policy and programdevelopment. None of the 33 participants, in describing their roleor their work, used the terms ‘policy maker’ or ‘policy making’.Terms used included officer, manager, or director in ‘x’ policy or

legislative development team. The term ‘policy development’ wasused by participants; ‘policy making’ was not.

The vast majority of strategic and day-to-day decision-makingfor participants and in these Agencies generally, as described in theinterviews and evidenced in organisational charts, was involvedwith development, implementation, management, monitoring andevaluation of programs and projects. ‘Program’will be used to referto programs and projects, as programs are comprised of projects. Asa result of these findings it was decided that the term ‘policy andprogram decision-maker’ would be used rather than the morecommon ‘policy-maker’ to better reflect the way that the partici-pants perceived themselves and their work.

Educational background varied widely, including: economics,law, psychology, business, sciences, health and public health, man-agement, politics, personal injury, physiotherapy, actuarial studies,accounting, occupational health and safety, statistics, criminology,and marketing. Professional backgrounds also varied widely. Themost common included: actuarial and financial services; insuranceand injury compensation; lawyers; and physiotherapists.

The majority of participants were Senior Managers who holdauthoritative positions within the Agencies. The results presentedbelow should be read with this in mind.

3.2. Themes

Five key themes emerged from the analyses of data on externalfactors that affect decision-making, as identifiedbyparticipants. Thesethemes represent theexternal key factors affectingdecision-making inthe workplace and transport IP&RC context in Victoria, Australia.

3.2.1. Stakeholder feedback and actionStakeholder feedback and action was referred to as formal and

informal feedback, action, advice, advocacy etc. from organisedgroups who represent issues and groups related to or affected byworkplace and transport injury, treatment and rehabilitation.Stakeholder feedback and action was the factor most frequentlyidentified as affecting policy and program decision-making; dis-cussed by 26 participants (79%), across all role levels.

Participants from both agencies identified health service pro-viders, injured persons and lawyers as key stakeholder groups.Unions, employer groups and private insurers were discussedmostly by Agency 1 participants. The extent and significance of

P. Zardo et al. / Social Science & Medicine 108 (2014) 120e127 123

discussion around injured persons and lawyers warranted theirseparate categorisation (see below).

When identifying barriers and facilitators, participantsdescribed how: “all the external stakeholders. can be both sup-portive and issues at times” (Manager 5). Participants from Agency 1described how: “Unions can be really helpful if they support aparticular initiative, or project. they have quite a bit of visibility withthe director, and up from there.the Unions, I think, are a hugeinfluencer .when they’re disengaged, then that can have a hugedisruptive effect” (Non-Manager 4); “Employer groups would be themain block” (Non-Manager 6). Manager 8 had positive experiences:“We’re fortunate in that, for example, our Employer stakeholders.they’re sort of a facilitator for us. I think the Unions have beenprobably more a facilitator for us than a barrier”.

Private insurers, referred to as ‘Agents’, were important stake-holders for Agency 1: “Obviously Agents, they’re a big party” (Man-ager 5); “Agent resistance. We’re really heavily reliant on all of themimplementing a particular project in the waywewant them to, in orderfor it to have the outcomes that it needs to have” (Non-Manager 1).

The Agencies also regard each other as stakeholders: “a lot of thethings that we (Agency 2) do will affect them (Agency 1) and viceversa. And although we still have autonomy there may be some in-stances where you know their disagreement may actually (have an)effect” (Manager 2).

Participants gave insight to how stakeholders utilise structuresand processes to influence decision-making: “For example we mightbe implementing a new travel policy.We might then get the taxidirectorate for example, lobbying the Minister to say ‘this new initia-tive’s unreasonable’. So then the Minister might come to us and say ‘Iwant a full review of that, put it on hold’” (Manager 3).

Senior Manager 16 and Manager 3 described how health pro-viders can influence: “Providers, if they're not willing to engage, or ifthey decide they're going to have a campaign against that change.They can lobby the Minister; they can do all sorts of things to block achange; lobby the Board and things like that”; “they all have, youknow, presidents’ talk to presidents and directors talk to directors.and say ‘I don’t like how you’re stopping these.services, keep it goingfor another year’”.

Senior Manager 1 identified health providers that wereconsidered particularly influential: “the medical community,particularly the AMA (Australian Medical Association), the doctors Imean. The paramedical and ORs (Occupational Rehabilitators), etcetera, to a degree”.

Manager 5 explained that these agencies establish formal andon-going as well as issue specific forums to gain stakeholder feedback:“There’s a whole range of forums and all of those are about gatheringfeedback and ideas and issues, consulting about howwemanage things,how we can do things depending on what the issue was”.

Non-Manager 1 described how research evidence acted as animportant support in addressing stakeholder feedback and action:

Whenever we need to influence external stakeholders, so particu-larly Unions or Provider groups, that’s when evidence becomesreally important within the project. Because in those cases we haveto justify why we’re doing them. And if we can justify it throughevidence that’s been gathered that’s believable and credible to anexternal group then we’re more likely to have success in imple-menting the project.

3.2.2. Ministerial and government inputMinisterial and government input was described as any input,

feedback or decisions provided or applied to the Agency by gov-ernment ministers, other government departments and individualdecision-makers within other government departments. Eighteen

(54%) participants across all role levels discussed ministerial andgovernment input, however the majority were senior managers.Senior Manager 1 candidly explained: “If you’re a public servant,then you are at the behest of the government of the day, and our job isto implement the policy of the government of the day”.

Senior Manager 6 described the direct influence governmentand ministerial input can have: “if the government says a decision‘we’re not going to do X, Y, Z0, then so be it, get on with it, don't worryabout it”. Senior Manager 7 identified the “The Minister and Cabinet”as the biggest influence on policy and program developmentprocesses.

Senior Manager 5 and Manager 12 described other governmentinput broadly as a barrier: “certainly there’s government departmentsthat block policy development”; “We've worked closely with Medi-care.whilst we have a great working relationship.there's stillblockers.that's not anybody deliberately blocking, it's just how theyoperate, so that's quite difficult”.

Senior Managers from both Agencies discussed the role of thestate government finance department: “Department of Treasury andFinance (DTF), which oversights (Agency 1), will block policy devel-opment”; “potentially DTF could block. I’ve not observed thathappening but.certainly.they have some control over what wespend and what we spend it on”.

Participants also described large-scale policy initiatives whereministerial and government input affected decision-making. Non-Manager 6 explained how “the OHS National Reform that ishappening at the moment.the Federal Government and the old StateLabour Government were driving it so.we had to do it. Now it’s allfallen apart because the State Government now says ‘well hang on.’”.Senior Manager 17 also identified this particular example as abarrier.

Senior Manager 3 experienced government input as facilitative:“We have a large.project in disability housing and that got legsbecause the government said.‘we want you to use your money tobuild infrastructure in Victoria for disability housing’.that's where itstarted, it was the government”. Even on less visible or critical issues,government input can drive policy and practice: “something likerecords management, until it got into the public domain through theOmbudsman, would not have been regarded as something that was‘drop everything to do’” (Manager 5).

Several seniormanagers howeverdid not experience governmentandministerial input as significant factor affecting decision-making:“We don’t get a lot of interference from government or anyone else. Wecan prettymuch design our own future and aim for it with a big, healthybucket of money, quite frankly” (Senior Manager 13).

3.2.3. Legal feedback and actionLegal feedback and action refers to advice and feedback from

lawyers, legal reviews and decisions, case law and changes to fed-eral or state legislation (not initiated by these Agencies) related toworkplace and transport injury prevention, treatment and reha-bilitation. Twelve participants, across each role level, discussedlegal feedback and action; the majority of which were SeniorManagers.

Senior Manager 3 explained that the Agencies engage withlawyers: “in all sorts of ways at all sorts of levels”. The barriers thatcan be created by legal feedback and action were described: “ourlegal stakeholders, if they don’t like what we’re doing they cancertainly make life quite difficult for us” (Senior Manager 12); “theplaintiff lawyer community, the lawyers that represent clients, occa-sionally they’ll try and do their best to block something we may bewanting to do” (Senior Manager 13).

Manager 5 described how legal feedback and action can affectstrategic policy and program decision-making: “lawyer groups.which probably have the more policy related impact, maybe more high

P. Zardo et al. / Social Science & Medicine 108 (2014) 120e127124

level. They’remore about ‘how should we treat this particular instance,or this particular worker’. And as a result, so when you go to makestrategic changes, they don’t care about the strategic changes. Theycare about: ‘these four cases that I’ve got, they’ve been mistreated’”.

Senior Manager 11 held a pragmatic view of legal feedback andaction:

We could talk about clinical justification and appropriate evidence,having an evidence base and evidence-based service and all thosesorts of things; lawyers and clients don’t speak that way and theydon’t think that way. They (lawyers) think ‘my client wants it, yougive it to them’.I don’t see it as blocking, and a lot of advocacy isnot about blocking. But it certainly doesn’t necessarily take accountof the evidence base in the same way that you or I might, becausethe evidence before them is an individual who says ‘I want it’, andtheir role is to try and get it.

Senior Manager 11 later added that: “I think they’re just doingtheir job, but does that mean at the end of the day you don't getoptimal policy development? Sometimes yes, absolutely”.

Manager 1 described how lawyers can “assist or alert.we havetaken feedback from them on some of our income policies.areas thatneeded some clarifying or tightening up. our interpretation of the Actand just court decisions, that is a big one”.

Senior Manager 3 gave an example of how legal feedback andaction has led to project development and funding:

If someone has a serious injury and they are wheelchair dependentor physically disabled we’ll modify their houses. Very long drawnout, complicated process.Through our regular meetings withlawyers they’ve raised this issue and said ‘you have to do somethingbetter about this’.We (now) have a project to greatly overhaul andimprove our home mods (modifications) process. So having thatlegal stakeholder pressure has helped that project get legs.

3.2.4. Injured personsEight participants from both Agencies and from across each role

level, discussed feedback provided by injured persons, which par-ticipants referred to as ‘clients’. Clients are a special stakeholdergroup. They differ from the stakeholders described above in thatthey provide their input individually, or via their lawyer, on anindividual case basis. They do not advocate for all injured persons,but rather provide input and take action related to their specific,individual injury case.

Senior Manager 11 talked about barriers and facilitators createdby client feedback: “There’s a lot of external stakeholders that eitherrepresent clients or work with our clients who bring value to the tablein terms of what’s right and wrong about the way to treat people”; and

Clients themselves sometimes come from a different perspective.The number of conversations I’ve had with clients where I’ve talkedabout evidence bases. And their conclusion to the whole thing is:“Yes, but I just want it.” And in reality, in the statutory approach,you often make compromises. And so a compromise in itselfsometimes is not optimal policy.

Client feedback was identified as the most influential factoraffecting decision-making for two senior managers: “individualproviders working with individual clients.because they influenceclients and clients ultimately are the ones who influencemost what(Agency 2) does” (Senior Manager 11).

our three pillars.two of those are really client centered.I thinkabout sort of a lot of the projects that have got up and the driver

behind that and it's about ultimately.‘how's this going to benefitour client?’.conversely, if clients are vocal about something thatthey see as really bad that will drive what we do as well. (SeniorManager 12).

3.2.5. Media coverageMedia coverage was identified as actual and perceived risk of

commentary and critique of the Agencies by the media. Five par-ticipants, from across each role level, four of which were fromAgency 1, mentioned the media being an external factor that canshape decision-making. Media coverage was predominately dis-cussed as a barrier to policy and program development processes,however one participant said; “I think that can go either way”(Manager 9). Potential threats of media coverage were described,rather than actual examples of firsthand experience of the mediaacting as a barrier.

Participants described the barrier as not media coverage itself,but political pressure arising from, or driving media coverage: “No,it’s not the media, it’s.fear of bad press, bad publicity or politics,political pressure” (Manager 9). Non-Manager 1 explained that tocreate political pressure, stakeholders “sometimes go to the media toreport their dissatisfaction with an aspect of the scheme”. Politicalpressure to make a decision or support a particular policy or pro-gram due to media focus was seen as risky, detracting from theAgencies’ capacity to do their day to day work and to address theissue through usual process or in due course: “with regards tosolicitors.if they were to find out about us looking in to a particularway of doing things. we run the risk of them going to the media”(Manager 2); “sometimes we have to behave irrationally, to meet theneeds of stakeholders, community, manage perceptions, so we canactually get on and do everything else we need to do” (Senior Man-ager 1).

4. Discussion

Analysis of data from 33 interviews revealed that a range ofexternal factors affect decision-making processes within govern-ment agencies. The external factors identified in this study werevariable; whether and to what extent they acted as barriers or fa-cilitators was dependent on the particular policy or program issueat hand.While external factors could have a significant influence ondecision-making processes, it is important to note that individualand organisational factors also significantly influence governmentdecision-making, and that factors at these different levels can beinterrelated or interdependent (Mitton et al., 2007).

The external factors affecting policy and program decision-making and use of evidence identified in this study are consistentwith findings from previous relevant research and political andsocial theory (Haynes et al., 2012; Lewis, 2005; Orton et al., 2011).This study has highlighted the importance of external factors inshaping policy and program development, and confirmed thecomplex and interdependent nature of policy and programdecision-making and the necessity for consultation and collabora-tion within these processes.

In the specific context of workplace and transport IP&RC inVictoria, Australia, the key external factors affecting policy andprogram decision-making were: government and ministerialinput, stakeholder and legal feedback and action, injured personsand the media. Interestingly, participants in this study did notmention external factors commonly noted in existing frameworksand research, such as financial or economic factors (Green et al.,2009; Lavis et al., 2012). The most commonly identified externalfactor was feedback and action from health providers, employers,

P. Zardo et al. / Social Science & Medicine 108 (2014) 120e127 125

trade unions and lawyers. Union and employer groups were foundto have a greater effect on decision-making in Agency 1 than inAgency 2. Health providers and lawyers affected decision-makingin both Agencies, making them particularly influential stake-holders in the Victorian IP&RC context. These stakeholders: wereoften focused on different issues; involved with different parts ofthe policy and program development process; held different ideasand values regarding which issues should be given priority; andused different forms of capital to influence decision-making.

In this context, policy and program decision-makers oftenconsult, collaboratewith and act on advice and feedback from thesestakeholder groups. Stakeholders were an important source ofvaluable ‘on the ground’ knowledge and experience. These groupscould also independently take action to affect policy and programdevelopment or change in this context. Stakeholder input was oftenexperienced as a barrier when a policy or program was viewed asimpractical for implementation, or unfair or unreasonable in someway. Stakeholder input informed and influenced the policy, pro-gram and project agenda, as well as development and imple-mentation. However, participants also made it clear that therewereinstances where these groups would not necessarily have a sig-nificant influence on decision-making.

These findings are consistent with recent research withAustralian public health policy decision-makers and academics(Haynes et al., 2012). As identified in this study, ministers andpolicy and program decision-makers in the Haynes et al. (2012)study often needed to use research to respond to, address, or in-fluence stakeholders, highlighting the influential role that stake-holders play in the policy process. Also commensurate with Hayneset al. (2012) was the central role of policy and program decision-makers compared to ministers. This study demonstrated that thevast majority of day-to-day decision-making regarding governmentpolicies and programs was done at the agency or departmentallevel. In this context the Minister tended to only intervene oncertain issues and in certain parts of the policy process; there werespheres of decision-making in these Agencies that rarely or neverexperienced ministerial input.

More broadly, Haynes et al. (2012) identified researcher’s abilityto communicate and engage effectively with government andstakeholder groups as critical to use of research evidence. It wasnoted that stakeholders are uniquely positioned to influence use ofresearch and engage researchers in the policy development process(Haynes et al., 2012). The importance of engaging stakeholderswhen seeking to increase use of research evidence in governmentpolicy development was highlighted in this study; and is increas-ingly recognised (Flitcroft et al., 2011; Lavis et al., 2012).

The external factors identified here and the need for collabo-ration and accordance between researchers and government, andgovernment stakeholder groups, have been identified in modelsand research on use of evidence in health policy environments.However, the role and influence of particular stakeholders, andhow to utilise this influence in strategy design has received muchless attention. Much research in the field is focused on increasinginteraction between researchers and policy decision-makers(Armstrong et al., 2011), neglecting to address the fact thatdecision-making in health policy environments is often influencedor mediated by external stakeholder groups, as this study andothers have demonstrated (Flitcroft et al., 2011).

One theory that is useful in examining how stakeholders utilizeand navigate social and political structures and processes to affectdecision-making in the IP&RC context is Bourdieu’s Habitus andField theory. Habitus is described as a set of dispositions, mindsets,or mental structures held in the mind of ‘agents’ (Swartz, 1997).Agents are individuals, institutions or organisations who employ, orhave ‘agency’ when they influence their social worlds. Practices,

institutions and social relations are structured and reproduced byhabitus and in turn practices, institutions and social relationsstructure habitus (Swartz, 1997). As habitus is produced by inter-nalisation of social structures, those who occupy similar socialpositions will ‘tend to have similar habitus’ (Ritzer and Goodman,2004, p. 522). Habitus shapes agents expectations and beliefsabout what is possible and probable, and therefore shapes and canconstrain thought and action.

Field is described as ‘a network of relations among the objectivepositions within it’ (Ritzer and Goodman, 2004, p. 522). In-stitutions, organisations and individual agents occupy positions inthe field. Fields are hierarchical but may be interrelated or distinctfrom other fields, and may transcend class. Fields operate ‘withtheir own specific logics’ and ‘generate among actors a belief aboutthe things that are at stake in a field‘ (Ritzer and Goodman, 2004, p.522). Agents strategically deploy capital (economic, social, cultural,symbolic etc.) to navigate, establish and protect their field posi-tions. The structure of the field constrains agents’ capacity tooccupy particular positions (Swartz, 1997). Field theory in partic-ular provides a useful framework to explain how different fieldpositions occupied by stakeholders, as agents in the IP&RC context,enable and/or constrain the different ways in which these stake-holders can and/or do affect decision-making within the Agenciesstudied.

Whilst participants in this study talked most about stakeholdersand lawyers; examination of findings utilising Bourdieu’s fieldtheory however, would position the government and specificallythe relevant Minister, as those most capable of exercising power inthe political field that structures workplace and transport IP&RCpolicy in Victoria (Swartz, 1997). The theory constructs thedeployment of power within political fields vertically and hori-zontally: political influence can be exercised from above, forexample from the Minister’s office down to government agencies,and also between or across, from one agency to another. The po-litical field permeates all other fields because political power oc-cupies a dominant, hierarchical position. The Minister andgovernment are extremely influential in affecting the habitus ofthose who conduct the practical operations of the Agencies. Theidentified stakeholder groups are important for researchers tounderstand and work with in this context, primarily because theyinfluence the Minister and Agency policy and program decision-makers: the most powerful players in this arena, as ultimate orfinal decision-making power rests with them. The fields affectingVictorian IP&RC are depicted in Fig. 1.

Ministers and governments utilise political, governmental andlegal structures and processes to exercise their power. The leg-islative process, existing legislation and structure of governmentestablish and delineate the role of ministers and governments,and provide broad direction for possible policy and programresponses within particular jurisdictions. So whilst the politicalfield, through legislation and regulatory systems, positions thegovernment and ministers as powerful agents in IP&RC it alsoconstrains the issues they can respond to and how. Ministerialand government power can also be derived from expectations ofthe Minister’s perspective or preferences that policy and programdecision-makers hold and react to, based on past experiences andassumptions. Interestingly, a network analysis of influence inpublic health policy in Victoria (Lewis, 2009) similarly identifiedthe relevant Minister and government as the main influence.However, unlike Lewis’ (2009) research, no participants in thisstudy identified academics as influential. This has emphasisedthat factors affecting research use can differ across similar con-texts, highlighting the need for further studies examiningresearch use within and between specific decision-makingcontexts.

Fig. 1. IP&RC Fields. It depicts the fields influencing IP&RC in Victoria, Australia. The unbroken lines represent legal or contractual relationships between groups. The broken linesrepresent relationships that are not formalised through legislation or other legal or contractual relationships. The heavier or thicker lines represent a more direct or strongerinfluence between groups, and the arrows depict the direction of influence. The size of the circles represents the weight of influence on IP&RC policy and program development.

P. Zardo et al. / Social Science & Medicine 108 (2014) 120e127126

Through the legislative process, the political field has estab-lished rights and responsibilities for different stakeholder groups inthis context. For example, trade unions and employer groups rightto consultation with Agency 1 is enshrined in legislation (Maxwell,2004). Lawyers are also afforded a prominent position in the fielddue to the legal right to seek compensation for workplace ortransport injury through common law processes and legal repre-sentation. In practice however, the analysis revealed that such legalstructures and process do not always, or necessarily, afford greaterinfluence. For example, health providers can be more influentialthan union or employer groups on specific issues, exercising agencythrough refusal to implement, as they are often not legally obligedto cooperate. Further, the political field and habitus influencinghealth providers is distinct from that of IP&RC.

Injured persons, from the perspective of field theory, have theleast power to influence decision-making in this field. Whilst IP&RCpolicy and programs are centred on injured persons and those atrisk of injury, individual injured persons must work through theagencies’ feedback systems, and structures and processes such asthe legal system or health providers to influence decision-making.Injured persons, most often, lack experience of the compensationsystem and process, therefore their expectation of the process andpotential outcomes are much less established than institutionalisedgroups, such as the stakeholders described above. Therefore,working through established structures and processes affordsinjured persons increased power within the field, subject to theinfluence of their individual habitus.

The stakeholders in this context were able to drive improve-ments in policy and programs on particular issues, but were alsofound to block evidence informed policy and program develop-ment. These findings highlight the importance of strategies that

bring researchers and policy makers plus key interested andinfluential individuals and groups together to interact and collab-orate on the development and implementation of evidenceinformed policy and programs. Leading research translation ex-perts and organisations have developed strategies to supportevidence-informed exchange and to facilitate demand, or ‘pull’ forresearch (Lavis, 2006) by government decision-makers and rele-vant stakeholders in public health contexts, see (AustralianNational University, 2010; Institute of Work and Health, 2011;McMaster Health Forum, 2012).

Despite implementation of such strategies, most are yet to havepublished evidence of their effectiveness. There is little evidencedemonstrating how external agents, structures and processes canbe effectively harnessed or addressed in strategies aimed atincreasing use of evidence in health policy environments(Grimshaw et al., 2012; Orton et al., 2011). To enhance the likeli-hood of evidence playing a substantial role in informing and sup-porting policy and program decision-making in IP&RC in Victoria;strategies aimed at increasing use of evidence would need to targetthe stakeholder groups described here, the Minister and govern-ment, and Agency policy and program decision-makers.

Researchers’ seeking to influence and support decision-makingin transport and workplace IP&RC in Victoria need to have a thor-ough understanding of the position, ideology and practices of therelevant government and minister, the key stakeholder groups, aswell as the nature of relationships between stakeholders, thegovernment and the Agencies. Researchers will also need to beskilled in effective communication and relationship and networkbuilding to engage and collaboratewith Agency policy and programdecision-makers and their stakeholders in this context (Hayneset al., 2012; Orton et al., 2011).

P. Zardo et al. / Social Science & Medicine 108 (2014) 120e127 127

5. Conclusions

In the Victorian IP&RC policy context, government decision-making is affected by factors operating outside of governmentpolicy and program development agencies. Governments, govern-ment agencies and departments, ministers, health providers,unions, employer associations, lawyers, the media and individualinjured persons affected decision-making in the two governmentcompensation agencies studied. These groups have a shared in-terest in improving the outcomes of IP&RC policy, programs andpractice, but have different perspectives and approaches regardinghow to achieve that aim. Research translation interventions, toolsand strategies aimed at increasing evidence informed decision-making in IP&RC policy and practice should consider howexternal agents, structures and processes that affect decision-making can be harnessed to enhance effectiveness.

Acknowledgements

I would like to thank the Institute for Safety, Compensation andRecovery Research for funding this research and the policy andprogram decision-makers who took the time to participate.

References

Armstrong, R., Waters, E., Dobbins, M., Lavis, J.N., Petticrew, M., Christensen, R.,2011. Knowledge translation strategies for facilitating evidence informed publichealth decision making among managers and policy-makers (Protocol). TheCochrane Library 6.

Australian Government Department of Infrastructure and Transport, 2011. RoadSafety. Retrieved December 14, 2011, from: http://www.infrastructure.gov.au/roads/safety/.

Australian Institute of Health and Welfare, 2012. Serious Injury Due to LandTransport Accidents, Australia 2008e09. Commonwealth Government,Canberra.

Australian National University, 2010. Australian Primary Health Care ResearchInstitute. Retrieved 16 July 2013, from: http://aphcri.anu.edu.au/about-us.

Best, A., Holmes, B., 2010. Systems thinking, knowledge and action: towards bettermodels and methods. Evidence & Policy: A Journal of Research, Debate andPractice 6 (2), 145e159. http://dx.doi.org/10.1332/174426410x502284.

Brownson, R.C., Fielding, J.E., Maylahn, C.M., 2009. Evidence-based public health: afundamental concept for public health practice. Annual Review of Public Health30 (1), 175e201. http://dx.doi.org/10.1146/annurev.publhealth.031308.100134.

Cameron, I.D., Rebbeck, T., Sindhusake, D., Rubin, G., Feyer, A.M., Walsh, J.,Schofield, W.N., 2008. Legislative change is associated with improved healthstatus in people with whiplash. Spine 33 (3).

Carter, B.J., 2010. Evidence-based decision-making: practical issues in the appraisalof evidence to inform policy and practice. Australian Health Review 34 (4),435e440.

Cassidy, J., Carroll, L., Cote, P., Lemstra, M., Berglund, A., Nygren, A., 2000. Effect ofeliminating compensation for pain and suffering on the outcome of insuranceclaims for whiplash injury. New England Journal of Medicine 342, 1179e1186.

Damschroder, L.J., Aron, D.C., Keith, R.E., Kirsch, S.R., Alexander, J.A., Lowery, J.C.,2009. Fostering implementation of health services research findings intopractice: a consolidated framework for advancing implementation science.Implementation Science 50 (4).

Dobrow, M.J., Goel, V., Lemieux-Charles, L., Black, N.A., 2006. The impact of contexton evidence utilization: a framework for expert groups developing healthpolicy recommendations. Social Science & Medicine 63 (7), 1811e1824. http://dx.doi.org/10.1016/j.socscimed.2006.04.020 pii:S0277-9536(06)00230-9.

Flitcroft, K., Gillespie, J., Salkeld, G., Carter, S., Trevena, L., 2011. Getting evidence intopolicy: the need for deliberative strategies? Social Science & Medicine 72 (7),1039e1046.

Gabbe, B.J., Cameron, P.A., Williamson, O.D., Edwards, E.R., 2007. The relationshipbetween compensable status and long-term patient outcomes following or-thopaedic trauma. Medical Journal of Australia 187 (1).

Glasgow, R.E., Emmons, K.M., 2007. How can we increase translation of researchinto practice? Types of evidence needed. Annual Review of Public Health 28,413e433.

Green, L.W., Ottoson, J.M., García, C., Hiatt, R.A., 2009. Diffusion theory andknowledge dissemination, utilization, and integration in public health. AnnualReview of Public Health 30 (1), 151e174. http://dx.doi.org/10.1146/annurev.publhealth.031308.100049.

Greenhalgh, T., Robert, G., Macfarlance, F., Bate, P., Kyriakidou, O., 2004. Diffusion ofinnovations in service organisations: systematic review and recommendations.The Millbank Quarterly 82 (4), 581e629.

Grimshaw, J., Eccles, M., Lavis, J., Hill, S., Squires, J., 2012. Knowledge translation ofresearch findings. Implementation Science 7 (1), 50.

Guest, G., Bunce, A., Johnson, L., 2006. How many interviews are enough?: anexperiment with data saturation and variability. Field Methods 18 (1), 59e82.http://dx.doi.org/10.1177/1525822x05279903.

Haynes, A.S., Derrick, G.E., Redman, S., Hall, W.D., Gillespie, J.A., Chapman, S.,Sturk, H., 2012. Identifying trustworthy experts: how do policymakers find andassess public health researchers worth consulting or collaborating with? PLoSONE 7 (3), e32665 http://dx.doi.org/10.1371/journal.pone.0032665.

Institute of Work and Health, 2011. Clinical Networks. Retrieved 16 November, 2011,from: http://iwh.on.ca/clinical-networks.

International Labour Organization, 2010. Emerging Risks and New Patterns ofPrevention in a Changing World of Work. International Labour Organization.

Killoran, A., Kelly, M.P., 2010. Evidence-based Public Health: Effectiveness and Ef-ficiency. Oxford University Press, Oxford ; New York.

Lavis, 2006. Research, public policymaking, and knowledge-translation processes:Canadian efforts to build bridges. The Journal of continuing education in thehealth professions 26 (1), 37e45. http://dx.doi.org/10.1002/chp.49.

Lavis, Røttingen, Bosch-Capblanch, Atun, El-Jardali, Gilson, Haines, 2012. Guidancefor evidence-informed policies about health systems: linking guidance devel-opment to policy development. PLoS Med 9 (3), e1001186. http://dx.doi.org/10.1371/journal.pmed.1001186.

Lewis, J.M., 2005. Health Policy and Politics: Networks, Ideas and Power. IP Com-munications, East Hawthorn.

Lewis, J.M., 2009. Understanding policy influence and the public health agenda.New South Wales Public Health Bulletin 7-8, 125e129.

Martin, G.P., Currie, G., Lockett, A., 2011. Prospects for knowledge exchange in healthpolicy and management: institutional and epistemic boundaries. Journal ofHealth Services Research & Policy 16 (4), 211e217.

Maxwell, C., 2004. Occupational Health and Safety Act Review (State of Victoria).McMaster Health Forum, 2012. Stakeholder Dialogues. Retrieved 16 October

2012, from: http://www.mcmasterhealthforum.org/index.php/stakeholders/stakeholder-dialogues.

Michie, van Stralen, West, 2011. The behaviour change wheel: a new method forcharacterising and designing behaviour change interventions. ImplementationScience 6, 42.

Michie, S., Johnston, M., Abraham, C., Lawton, R., Parker, D., Walker, A., 2005.Making psychological theory useful for implementing evidence based practice:a consensus approach. Quality and Safety in Health Care 14, 26e33.

Mitton, C., Adair, C.E., McKenzie, E., Patten, S.B., Waye Perry, B., 2007. Knowledgetransfer and exchange: review and synthesis of the literature. Milbank Quar-terly 85 (4), 729e768 doi: MILQ506, [pii]10.1111/j.1468-0009.2007.00506.x.

Murgatroyd, D., Cameron, I., Harris, I., 2011. Understanding the effect of compen-sation on recovery from severe motor vehicle crash injuries: a qualitative study.Injury Prevention 17, 222e227.

Nutbeam, D., Boxall, A.M., 2008. What influences the transfer of research intohealth policy and practice? observations from England and Australia. PublicHealth 122 (8), 747e753. http://dx.doi.org/10.1016/j.puhe.2008.04.020.

Orton, Lloyd-Williams, Taylor-Robinson, O’Flaherty, Capewell, 2011. The use ofresearch evidence in public health decision making processes: systematic re-view. PLoS ONE 6 (7).

Patton, M.Q., 2002. Qualitative Research and Evaluation Methods, third ed. Sage,Thousand Oaks.

Personal Injury Education Foundation, 2011. Understanding Accident Compensationin Australia and New Zealand. PIEF, Melbourne.

Ritzer, G., Goodman, D.J., 2004. Agency-structure integration. In: Ritzer, G.,Goodman, D.J. (Eds.), Sociological Theory, sixth ed. McGraw Hill, New York.

Rychetnik, L., Bauman, A., Laws, R., King, L., Rissel, C., Nutbeam, D., Caterson, I., 2012.Translating research for evidence-based public health: key concepts and futuredirections. Journal of Epidemiology and Community Health. http://dx.doi.org/10.1136/jech-2011-200038.

Safe Work Australia, 2012a. Australian Work-related Injury Experience by Sex andAge: 2009e10. Commonwealth Government, Canberra.

Safe Work Australia, 2012b. Cost of Work-related Injury and Disease for AustralianEmployers, Workers and the Community: 2008e09. Commonwealth Govern-ment, Canberra.

Swartz, D., 1997. Culture & Power: The Sociology of Pierre Bourdieu. University ofChicago Press, Chicago.

Transport Accident Commission, 2012. Annual Report 2012 (Victoria).WorkSafe Victoria, 2012. WorkSafe Victoria Annual Report (Melbourne, Victoria).