Redwood, S., Brangan, E., Leach, V. A., Horwood, J., & Donovan, J.(2016). Integration of research and practice to improve public healthand healthcare delivery through a collaborative 'Health IntegrationTeam' model - a qualitative investigation. BMC Health ServicesResearch, 16, [201]. https://doi.org/10.1186/s12913-016-1445-z

Publisher's PDF, also known as Version of recordLicense (if available):CC BYLink to published version (if available):10.1186/s12913-016-1445-z

Link to publication record in Explore Bristol ResearchPDF-document

This is the final published version of the article (version of record). It first appeared online via BioMed Central athttp://dx.doi.org/ 10.1186/s12913-016-1445-z. Please refer to any applicable terms of use of the publisher.

University of Bristol - Explore Bristol ResearchGeneral rights

This document is made available in accordance with publisher policies. Please cite only thepublished version using the reference above. Full terms of use are available:http://www.bristol.ac.uk/red/research-policy/pure/user-guides/ebr-terms/

RESEARCH ARTICLE Open Access

Integration of research and practice toimprove public health and healthcaredelivery through a collaborative 'HealthIntegration Team' model - a qualitativeinvestigationSabi Redwood1,2* , Emer Brangan1,2, Verity Leach1,2, Jeremy Horwood1,2 and Jenny L. Donovan1,2

Abstract

Background: Economic considerations and the requirement to ensure the quality, safety and integration of researchwith health and social care provision have given rise to local developments of collaborative organisational forms andstrategies to span the translational gaps. One such model – the Health Integration Team (HIT) model in Bristol in theUnited Kingdom (UK) - brings together National Health Service (NHS) organisations, universities, local authorities,patients and the public to facilitate the systematic application of evidence to promote integration across healthcarepathways. This study aimed to (1) provide empirical evidence documenting the evolution of the model; (2) to identifythe social and organisational processes and theory of change underlying healthcare knowledge and practice; and(3) elucidate the key aspects of the HIT model for future development and translation to other localities.

Methods: Contemporaneous documents were analysed, using procedures associated with Framework Analysis toproduce summarised data for descriptive accounts. In-depth interviews were undertaken with key informants andanalysed thematically. Comparative methods were applied to further analyse the two data sets.

Results: One hundred forty documents were analysed and 10 interviews conducted with individuals in leadershippositions in the universities, NHS commissioning and provider organisations involved in the design andimplementation of the HIT model. Data coalesced around four overarching themes: ‘Whole system’ engagement,requiring the active recruitment of all those who have a stake in the area of practice being considered, and‘collaboration’ to enable coproduction were identified as ‘process’ themes. System-level integration and innovationwere identified as potential ‘outcomes’ with far-reaching impacts on population health and service delivery.

Conclusion: The HIT model emerged as a particular response to the perceived need for integration of research andpractice to improve public health and healthcare delivery at a time of considerable organisational turmoil and financialconstraints. The concept gained momentum and will likely be of interest to those involved in setting up similararrangements, and researchers in the social and implementation sciences with an interest in their evaluation.

Keywords: Research and healthcare collaborations, Coproduction of healthcare knowledge and practice, Integratedknowledge translation

* Correspondence: sabi.redwood@bristol.ac.uk1School of Social & Community Medicine, University of Bristol, Canynge HallBristol39 Whatley Road, Bristol BS8 2PS, UK2National Institute for Health Research, Collaborations for Leadership inApplied Health Research and Care West (NIHR CLAHRC West), 9th floorWhitefriars, Lewins Mead, Bristol BS1 2NT, UK

© 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Redwood et al. BMC Health Services Research (2016) 16:201 DOI 10.1186/s12913-016-1445-z

IntroductionThe imperative to close the translational gaps betweenscientific research evidence and routine practice in thedelivery of healthcare is particularly pressing in finan-cially straitened times.In advanced healthcare economies, the drive to exploit

the potential of scientific innovation to improve qualitythrough new approaches to prevention, diagnosis andtreatment of diseases, and decrease costs by integra-tion between services has given rise to a range ofnew organisational partnerships [1–3]. Such collaborativepartnerships bring together the producers (e.g. academicsand researchers) and users (e.g. service leads, policymakers, healthcare professionals) of research findings. Inthe United Kingdom (UK), the systematic integration ofresearch into clinical practice and organisational routinesis promoted so knowledge can be coproduced [4]. Thisapproach is based on the accumulating evidence that re-search translation and implementation are contextuallysituated and complex, rely on multiple professionals andorganisations, and involve multifaceted, iterative, andoften unpredictable processes [5–7]. Coproduction ofknowledge can be facilitated through meso-level (organ-isational) partnerships across professional and organisa-tional boundaries, using new organisational forms [8].While there is an increasing number of descriptions ofthese partnerships and their variants, a recent inter-national scoping review [8] has highlighted a major gap inknowledge about the social and organisational processesunderpinning their workings, and a lack of empiricallygrounded theoretical development.The aim of this article is to describe the development

and establishment of micro-level ‘operating units’, orHealth Integration Teams (HITs), of a locally evolvedstructural partnership of National Health Service (NHS)organisations, local authorities, patients and the public,and universities, to foster collaboration across itsstakeholders and generate health improvements for localpopulations through integrated working. Based on datafrom contemporaneous documents and reflections ofparticipating key informants, we document the develop-ment of the HIT model and set it within the context ofthe coproduction of healthcare knowledge in the Englishhealth and care economy, and of the emerging organisa-tional models to facilitate it.

BackgroundThe HIT model was developed following the formation ofa collaborative partnership, the ‘Bristol Health Partners’(BHP) in the West of England in 2011–12. This partner-ship, BHP, was a locally driven initiative by leaders of NHSorganisations, a city council (local government withresponsibility for a range of public services) and twouniversities who decided to build a partnership following

an unsuccessful attempt to win competitive funding to es-tablish a Collaboration for Leadership in Applied HealthResearch and Care (CLAHRC). CLAHRCs were estab-lished in 2008 in several regions in England as networksof research partnerships for applied health research andthe translation of research findings into improved patientoutcomes, supported by central government funding. Thefailure to attract funding led to a re-evaluation of thecurrent arrangements for health research and knowledgetranslation, resulting in the BHP partnership and thesubsequent development of the HIT model. The originalplan was to emulate the Academic Health SciencesCentres (AHSCs) which were well established in NorthAmerica and Europe including London, Cambridge andManchester [9]. These centres combine the delivery ofservices to patients with high levels of research and teach-ing through sophisticated collaborative mechanismsbetween universities, hospitals and, in some cases, primarycare based organisations. Their focus was on biomedicalresearch in specific clinical disciplines, hospital-basedservices and early stage translation. However, the localcontext with research expertise particularly in populationhealth linked to commissioning and public health, was apoor fit with the AHSC model. This article describes howthe dominant biomedical discourse about knowledgetranslation was revised to formulate a new strategy with afocus on patient-centred integration across health andcare pathways. The HITs which were set up from 2012 on-wards and numbered 19 by October 2015, clearly reflectthis re-visioning of knowledge translation, operating inareas related to public health, long-term conditions andcross-sectorial working (see Table 1). For a detaileddescription of the individual HITs see Bristol HealthPartners, 2016 [10].Politically, the period of time during which these de-

velopments took place was characterised by acute uncer-tainty following a new coalition government’s proposalsfor a top-down, highly complex NHS re-organisationwhich passed into law in March 2012. This coincidedwith the establishment of BHP in April 2012 and the is-suing of the first invitations to become accredited HITs.The re-organisation presented those trying to build newpartnerships with particular challenges since establishedNHS bodies such as the Strategic Health Authoritieswho provided leadership at system level were abolished,and new bodies such as the Clinical CommissioningGroups (CCGs) were established as membership orga-nisations led by primary care doctors or GeneralPractitioners (GPs) [11]. The term ‘commissioning’ isspecific to England and refers the strategic planningand purchasing of health care services for the localpopulation. In England, where there is significant cen-tralised direction and performance management fromthe Department of Health [12] commissioning also

Redwood et al. BMC Health Services Research (2016) 16:201 Page 2 of 13

Table 1 Health Integration Teams, their aims and accreditation

Month of accreditation HIT title HIT aims

July 2012 Bristol bones and joints • Harness evidence-based practice and associatedresearch to fill knowledge gaps

• Improve care pathways and outcomes for patientswith musculoskeletal conditions

Avoiding hospital admissions • Reduce complexity in the local urgent care system• Optimise the productivity and efficiency of existingand new interventions

December 2012 Sexual health improvement • Transform services to improve sexual health for thepeople of Bristol and the South West

Improving care in self-harm • Examine the care pathway, and utilise knowledge,expertise and resources to achieve the highest qualityevidence based patient care and treatment for peoplewho harm themselves

Dementia • Deliver dementia-friendly communities and servicesbased on the highest quality evidence

• Conduct world-class research to achieve the bestquality of life for people and families living withdementia

Supporting healthier and inclusive neighbourhoodenvironments

• Use science, community voices and innovation toestablish Bristol as a healthy city

• Reduce health inequities• Closely align city development with health, well-being, social inclusion and green city aspirations

Respiratory infections • Improve the management of patients at every stageof their illness and care

• Use NHS resources as efficiently as possible

Retinal outreach, integration and research • Implement research-driven service delivery• Engage the patient voice, staff and commissioners indeveloping those services

July 2013 Child injury • Help Bristol set the national standard for integratingprevention, care and rehabilitation across children'strauma services

Parkinson's and other movement disorders • Develop whole system partnership working formovement disorders across the Bristol, NorthSomerset and South Gloucestershire region

• Develop a high quality, high impact, internationally-recognised system for Parkinson's and othermovement disorders

December 2013 Chronic Kidney Disease • Improve outcomes for patients with kidney disease inthe Bristol area through: prevention, patient care,education & research

Bristol network for equality in early years health andwellbeing

• Focus on antenatal care to children aged seven• Achieve improvements in oral health, nutrition, andsocial and emotional wellbeing

July 2014 Active people: promoting healthy life expectancy • Encourage the adoption of physical activity and otherhealth behaviours among older age groups in orderto improve their overall health during their later years

Addictions • Maximise the use of the resources already available toreduce substance-related harm

Integrated pain management • Improvements in performance, productivity andefficiency by ensuring that our research programmesand expertise in the management of chronic pain areintegrated into care

Bristol immunisation group • Develop an outstanding immunisation service• Lead research on immunisation development andprovision, responsive to vaccine preventable infectiousdisease outbreaks

Redwood et al. BMC Health Services Research (2016) 16:201 Page 3 of 13

involves the implementation of national policy withinthe context of local needs and resources. Anotherchange resulting from the reforms was the transfer ofpublic health functions from the NHS to local governmentand the formation of a new body, the Health and Well-being Boards to link GP commissioners to local govern-ment and to provide a forum better to link commissioningplans for health and social care services.The implications of the reforms, as Ham et al. [11]

point out, was a leadership vacuum at the system levelas the leadership previously provided by the Departmentof Health centrally and the Strategic Health Authoritiesregionally was distributed between several organisations,each overseeing a part of the NHS without havingsystem level oversight or responsibility. Similarly, at locallevel where system leadership between commissioners,providers, local authorities and other partners was vitalat a time when new integrated models of care were re-quired to meet growing demands on health and socialservices, it was missing, in part due to current seniorleaders having responsibility for individual organisations,rather than the system. The reforms also resulted in thefragmentation of commissioning and a loss of populationbased commissioning because the responsibility for pro-curing services was split between CCGs, NHS Englandand local government. The drive to develop and imple-ment the HIT model was a response to these conse-quences and to mitigate the effects of lack of systemleadership and the risk of fragmentation. It was also aninitiative to promote evidence-based practice in commis-sioning and service delivery and a forum for integration.Drawing on the principles of theory-based evaluation

[13–15], this study sought to provide empirical evidenceabout how the HIT model emerged; identify the theory ofchange and social and organisational processes underlyingthe coproduction of healthcare knowledge and practice;and elucidate the key aspects of the HIT model for futuredevelopment and translation to other localities.

MethodsUsing a qualitative design, the research team accessed,collated and analysed contemporaneous documents, and

conducted interviews with individuals who were leadingthe development of the HIT model. Methods andfindings are described below, adhering to criteria forreporting qualitative research [16].

Documentary analysisThe research team was given permission to access docu-ments belonging to BHP, the organisation accreditingHITs, produced between September 2010 and August2014, describing the development of the HIT model andsetting up of individual HITs. All documents wereanonymised before being imported into a qualitativedata management software package (QRS-NVivo 10) forcoding and analysis. Procedures associated with Frame-work Analysis [17, 18] were used to produce sum-marised data for descriptive accounts. These accountstraced the processes though which the partner organisa-tions collaborated and established the structures andfacilitating conditions for the model, and were used tocomplement the data generated through interviews withkey informants.Documentary materials related to the establishment of

the HIT model and individual HITs, and were collectedfrom the point at which expressions of interest weresubmitted, the application process and later set-upperiod, and so varied depending on how long the HIThad been in existence. They included discussion papers,application forms, feedback documents, progress reportsand meeting notes. An initial coding-frame was drawnup to reflect this. Documents tracking the strategy wereorganised chronologically and coded inductively.

Interviews with individuals who were leading thedevelopment of the HIT modelInterview data were collected face-to-face or via tele-phone from individuals in strategic leadership roles whoformed the BHP steering group and were involved in thedesign and implementation of the HIT model. Partici-pants volunteered to take part following an explanationof the rationale for the study at a meeting and subse-quent distribution of written study materials. Interviewswere digitally recorded and transcribed in full. Interview

Table 1 Health Integration Teams, their aims and accreditation (Continued)

• Use innovative technology to enable people to bebetter informed regarding their own (and theirchildren's) vaccinations

Psychological therapies in primary care • Improve uptake of, access to, and outcomes foreffective psychological therapies

Improving perinatal mental health • Improve the identification and subsequent care ofparents with poor mental health before and followingthe birth of their child

July 2015 (Expressions ofinterest)

Cancer, Chronic eye conditions, Psychosis, Eating disorders

Redwood et al. BMC Health Services Research (2016) 16:201 Page 4 of 13

questions were open and broad, seeking information onparticipants’ roles in the development of the HIT con-cept and their recollections, followed by each partici-pant’s perspective and understanding of how the modelwas supposed to work and produce beneficial outcomes.Fieldwork notes, debriefing notes and analytical memoswere kept to assist in the analysis which was based onthe principles of the constant comparative method [19].This involved initial coding, the forming and refining ofcategories, searching for negative evidence and compari-son across each stage of the analytic development ofexplanatory concepts which were complemented by theanalysis of the documentary data. Data management,coding and categorisation were supported by QRSNVivo software.

ResultsIn total, 140 documents were included in the analysis,with 10 interviews with individuals in senior leadershippositions at two universities, one NHS commissioningorganisation, and four NHS provider organisations. Theyincluded academics and clinical-academics who went onto become HIT directors, and senior managers. VL, EBand SR collated, coded and analysed the documentarymaterials; SR collected and analysed the interview data.Despite the heterogeneity of the data, four overarchingconcepts emerged. These permitted the development ofan initial implicit theory of change and clarified theunderlying logic of improvement while making explicitsome of the mechanisms that were considered crucial inproducing desired outcomes. These concepts were (1)‘whole system’ engagement, (2) collaboration, (3) inte-gration and (4) innovation. First we will describe thegenesis of the model and its context, and then move onto elucidate each concept.

HIT model genesisThe model’s antecedents go back to 2010 when the BHPpartnership was formed across NHS organisations, citycouncil and universities to promote research andinnovation in the area because of set-backs in attractingresearch funds and the perception that local organisa-tions were less well connected than in other areas. Asone participant said, ‘we decided that we would get ouract together in terms of research and other relationships’(Participant 1, academic). Some progress was made but‘[the steering group] really wasn’t moving the agendaforward’ (Participant 2, clinical academic). Consequently,an independent consultant was commissioned to explorepossible options for a model closely aligned to thegroup’s ambitions and to provide an in-depth analysis ofthe potential strengths, opportunities, political and eco-nomic risks, and organisational and governance models.

The model that had gained currency at the time wasthe Academic Health Sciences Centre (AHSC) as out-lined previously, which was focused on hospitals andmedical schools, biomedical research and clinicaldisciplines. The participating organisations were closelyconnected through partnership agreements, commonstandards and goals, and an overarching identity. TheUK Department of Health had accredited five EnglishAHSCs in 2009 to integrate research, teaching andservice strategically and operationally to deliver ‘a wholewhich is greater than the sum of the parts’, and to forma recognised elite attracting investment research andinnovation. This was the direction of travel the consultantwas recommending for BHP:

‘He was very much of the mind-set of an AHSC whereyou had pre-clinical biomedical sciences with earlystage translation and that was the model he was usedto. He based some of his early documents around theKing’s model and even called them Clinical AcademicGroups.’ (Participant 2, clinical academic)

The Clinical Academic Groups [20] were ‘the keybuilding blocks of all English AHSCs [and] are specialty-level, cross-organisational groups’ (Development period,2011), and clearly focused on biomedical research inspecific clinical disciplines, hospital-based services andearly stage translation. These groups were seen as theleaders for research, translation and innovation:

‘[The consultant] was quite insistent that the mostimportant things were strong leadership, and soclinical academics were seen to be strong leaders, andthat it should be around translation. Everybody alwaystalked about translation as if it was just Type 1[early stage basic research] translation. There was noreal consideration for Type 2 [development of newapproaches and technologies] really.’(Participant 1, academic)

The AHSC model initially had strong support becauseof the emphasis on partnership working. However, therewas a growing awareness that there was a lack ofalignment with the more broad-based aspirations forwider collaboration and integration across the healthand care landscape in the local area. The following twoexcerpts illustrate this mismatch:

‘It felt too clinical. It felt too hospital orientated. Itdidn’t seem to draw on primary care, the preventionagenda, public health, the kind of things that weresearch in Bristol, but it also didn’t representthe ethos around what we were trying to do.’(Participant 6, academic)

Redwood et al. BMC Health Services Research (2016) 16:201 Page 5 of 13

‘[The report] does not currently fully represent orengage with Bristol’s key strengths in population healthsciences research, clinical translation, social andcommunity services and public health, or the CityCouncil’s priorities. These issues provide an impetusfor a new vision to drive partnership across the city.’(Document from development period, 2011)

An alternative model was developed from within BHPthat addressed ‘key public health issues through the inte-gration of research and practice’ (Development period,2011). Its aims were to build on existing collaborationsand develop new partnerships between commissioners,primary care and public health professionals, alliedhealth professionals, specialist clinicians, basic scientistsand health service researchers with active involvementof patients and the public. This model was more attract-ive locally because it suggested ways of working acrossthe whole health and care economy, involving a muchwider range of stakeholders, rather than privilegingparticular clinical disciplines. Furthermore, it foundresonance among a wide range of local academics andresearchers who had developed programmes ofpopulation-based research which was more closelyaligned to the new model than the clinical, oftenlaboratory based research carried out in AHSCs. As thealternative model gained currency, ideas about howthese aspirations could be delivered began to emergeand the first references appear to what became themodels’ ‘operating units’, the HITs. Their purpose wasdefined as:

‘[integrating] research and clinical delivery in agreedways between the (…) partners, for example byestablishing new or more efficient care pathways,introducing or evaluating commissioning around theprimary/secondary interface, and more effective,efficient and acceptable ways of treating patients.There would need to be detailed negotiations betweenTrusts [any NHS organisation such as hospitals orprimary care organisations] and Universities toensure integration.’

The BHP steering group then designed a process ofaccreditation, encouraging key individuals and their organi-sations to identify common goals and formulate jointaction. Thus the group created a common structure forcollaborative working across organisations and sectors, and- through the accreditation methods - reinforced the behav-iours and processes they wished to encourage. There wereno prescribed themes or areas of practice around whichHITs were expected to form. Instead each HIT was able todevelop its own collective identity, mobilise commitmentand align with HIT members’ values and aspirations.

‘Whole system’ engagementThe first of the overarching concepts emerging from thedata refers to the process of actively seeking to engageorganisations and people who play a part in the healthand care economy related to the long-term conditionsor public health issue at the centre of the HITs beingformed. Those engaged included provider organisations,commissioning organisations, professionals/staff in theseorganisations, public(s) and service users and carers,other public sector organisations (for example, localcouncils and health and wellbeing boards), and othernon-NHS/social care organisations (for example, thirdsector organisations, industry). Achieving this kind ofwhole system engagement is challenging because indi-vidual organisations are accountable for their own per-formance rather than for their performance in relationto the system. In other words, each organisation is fo-cused on providing the service or function it has beencommissioned to provide with little incentive to considerthe entirety of the provision as experienced by patientsor service users. The proliferation of providers and thesubsequent fragmentation of patients’ journey throughthe health and care system has been exacerbated by theNHS reforms in 2012, particularly because leadershipand commissioning arrangements have been furtherbroken up and senior officers continue to be accountablefor the performance of their own organisation ratherthan that of the whole system.As HITs were charged with the task of integration across

a number of different boundaries within the whole healthand care system, their membership proved to be a decisivefactor in achieving some leverage on the very challengingproblems being tackled. Applicants were required to ex-plain who needed to be involved in the HIT and in whatspecific ways to achieve specific outcomes or goals.This process-related theme of ‘whole system’ engage-

ment provided a range of examples which highlightedthe importance of creating a legitimate space where allthose who had a stake in the public health or servicedelivery issue being addressed could voice their perspec-tives and the accountability demands that are often asso-ciated with these perspectives. Despite a longstandingcommitment to engagement, involvement and network-ing across sectorial, professional and organisationboundaries to tackle shared problems, it would appearthat facilitating such dialogue and interaction throughstructural partnerships made it easier for people to es-tablish personal contacts and coalitions, and promotedjoint action. Table 2 summarises the main features ofthis theme and provides illustrative data extracts.

CollaborationThe second process theme of collaboration refers to thedevelopment and embedding of methods that enable

Redwood et al. BMC Health Services Research (2016) 16:201 Page 6 of 13

partnership working and co-production across HITspecific structures and procedures.Sponsorship at the most senior level in one of the

organisations forming a HIT, usually the employing in-stitution of the HIT director, was seen as vital not justas a sign of organisational commitment to the HIT byproviding time and resources to support its work, butalso to proactively seek out opportunities for collabor-ation and integration (see Table 3). This suggests thatconnecting HITs to local resources was seen to be animportant role of senior leaders in the health and careeconomy to facilitate new relationships and partnerships.While these relationships could be pre-existing or initi-ated through the HIT application and setup process,their quality and fruitfulness were contingent on factorssuch as ‘a common ideology’ of seamless and integrated

Table 2 ‘Whole system’ engagement

Concept type: Process related

Short definition: Identifying and actively seeking toengage organisations and peoplewho play a part in the health andsocial care economy related to thelong-term conditions or publichealth issue at the centre of the HITbeing formed.

Function Data extracts

Dealing with challengingissues which crossorganisational and sectorboundaries.

‘[Those coming together in thenascent HITs] realised they weregrappling with different ends of thesame issue and actually theyhad a lot they could share’(Participant 1, academic)

I think there’s great strength in havingthe whole system involved becausethat is how people’s lives work andthat’s how we’ll get out of the messwe’re in.’ (Participant 7, seniorcommissioner)

Making research morerelevant and deliverable.

‘I bought into the notion that if youhave service providers andcommissioners and researchers alltogether, and then you designresearch which is going to meet theneeds of the commissioners and theservice, you’re more able then todeliver the findings of the research …So I just bought the notion that itwould be more relevant and moredeliverable. That’s why I supported it.’(Participant 7, senior commissioner)

Key aspects

Facilitated by the structured,iterative application process.

‘We were delighted to hear that thework involved in putting theapplication together has alreadyfacilitated new connections and newconversations: this is exactly what wewant to achieve.’ (Feedbackdocument to HIT, 2014)

Creating a new space inwhich ideas could bediscussed and exploredwithout concerns aboutinvading other organisations’territory or individuals’agendas.

‘I think they have createdconversations across the city andlegitimised conversations betweendifferent organisations, not justacademics and the service, but alsobetween different parts of the service’.(Participant 6, academic)

‘Would we have succeeded without theHIT? The answer is I don’t know in allhonesty. I would have thought wewould have got to a certain levelwithout question because of the energyand the drive that was being created.What the HIT’s done is mandated whatwe were doing and it has opened doorsas a results of just having, not just ofhaving a label, but of having anopportunity that has been mandatedby the wider [health community]. Thatwas really important and really was astep change in our speed ofdevelopment.’ (Participant 5, HITdirector, clinical academic)

Table 2 ‘Whole system’ engagement (Continued)

Changing norms about whoshould be included.

‘If I think of the 360 degree segments, Iwas probably covering around 90degrees; maybe 100 degrees if we’rebeing optimistic, of the key peoplethat we really needed to include insomething that was going to look atsuch a challenging area. … Now Ithink I wouldn’t even conceive ofdoing something like this withoutincluding all the players, and we’reconstantly thinking of other people weshould include.’ (Participant 10, HITdirector, clinical academic)

‘Our ambition is that this policyshould underpin a culture shift suchthat PPI [Patient and PublicInvolvement] is embedded at all levels,including commissioning, decisionmaking and policy rather than beinglimited to the logistics of servicedelivery and questions of patientsatisfaction’. (HIT applicationdocument, 2012)

Involvement ofcommissioners – goingbeyond previouscollaborations betweenacademics and largeproviders.

‘What I think is really strong about theHIT model is that involvescommissioners and when you go toother places and you look at whatthey’re doing, they tend to be lessstrong about having the whole systemin the room. So typically you tend tosee a lot of evidence of working withacute, but not necessarily mentalhealth, not local authority and notcommissioners.…. So I think that’s anincredibly important point that ifyou’ve got commissioners at the heartof it and in some of these leadershiproles, it looks very different.’(Participant 7, senior commissioner)

Meaningful and timelyinvolvement andengagement of patients andpublic(s).

‘So how do we infiltrate the realdecision making spaces within theorganisation so public voices areactually being heard where decisionsare really being made?’ (Participant 3,academic)

Redwood et al. BMC Health Services Research (2016) 16:201 Page 7 of 13

care pathways and services. In order to create and sup-port such a common ideology to serve the requirementsfor collaborative working, an infrastructure for sustainedinteraction, communication and exchanging informationand knowledge between and across HITs and theircollaborating organisations and individuals had to bedeveloped.

Several HITs had identified skills gaps within theirteams and had taken steps to address these. Forexample, HITs had enlisted input and advice from thirdsector or public sector organisations whose remitincluded supporting engagement with commissioners,industry, or patients and the public. A review groupcomprised of public members was created with the aimthat every HIT would ‘have genuine PPI (Patient andPublic Involvement) within their structures’ (Participant3, academic). There were varying degrees of previousPPI experience by HIT applicants and additional supportwas put in place:

‘So the thing that came out of that was that werecognise the need to try and offer more practicalsupport to the HITs. And that led to the creation of thePPI facilitator post (…).’ (Participant 3, academic)

As early HITs began to form and new collaborationsemerged, it was important for mechanisms to beestablished which would create and foster conditions forcollaboration across HITs to facilitate shared learningand to build capacity to address complex problems:

‘All HITs are expected to coordinate their activitieswith other relevant HIT teams and this is particularlyimportant in areas where a number of new HITs aredeveloping.’ (Feedback to HIT, 2014)

Indeed, many HIT applicants made reference to plansfor such coordination by specifying and delineating theirparticular sphere of influence while emphasising the roleand added value of ‘collaboration of collaborations’.Some HIT applicants also described strategies to fa-cilitate improvements in collaboration which requiredan internal focus on the part of all partner organisa-tions in order to “establish confidence and capabilityto share data between organisations and agencieswhere this is in the patient’s interest (e.g. to evaluateneed and outcomes of service provision)” (HIT applica-tion, 2013) at an early stage. In other words, for orga-nisations to be able to collaborate within a HIT oracross other HITs, information needed to flow acrossorganisational boundaries, requiring HIT managementgroup members from different organisations to facili-tate progress in this sensitive area, being responsiblefor engaging key individuals from their own organisa-tions in the process:

‘Management group members will liaise withinformation governance leads in their ownorganisations to facilitate collaboration and ensure aneffective data sharing agreement can be established.’(HIT application, 2013)

Table 3 Collaboration

Concept type: Process related

Short definition: Development and embedding ofmethods that enable partnershipworking and co-production, involvingHIT specific structures and procedures

Function Data extracts

Breaking downorganisational, sectorial andprofessional boundaries andsilos by enabling effectivecommunication aboutcommon goals

‘…relationships are the most importantthing in order to move anything togetherin a partnership to me. So the aspect ofone area that has been successful wasthat the relationship with thecommissioners was strong in the sense ofcommon ideology. And mutual respect ofeach other’s tensions to deliver this, sowhen you understand that you canactually find a common goal that wouldachieve. So you’re getting commissioners,the spread of medics and allied healthprofessionals together in order to justsimply put the patient [in the centre] asdoing the right thing for the right reasonsand respecting each other’s tensions inorder to deliver that.’ (Participant 5, HITdirector, clinical academic)

Key aspects

Requirement for sponsorshipat the most senior level inone of the organisationsforming a HIT.

‘The HITs are all supposed to haveexecutive sponsors from the organisationand it’s making sure that’s that a liverelationship because those executives Ithink can really help the HITs but, youknow, they’ve got to keep them in theirmind all the time. Whenever anythingcomes across their desk they have to think“Oh, yeah, this might help myHIT.”’(Participant 7, senior commissioner)

Building infrastructure forsustained interaction,communication andexchange of information andknowledge between andacross HITs and theircollaborating organisationsand individuals.

‘The working groups (…) will each be ledby two individuals, encompassingacademic and service leadership. Thesewill both report to and work with the twothematic groups’. (HIT applicationdocument, 2013)

‘This HIT will be embedded manageriallywithin the Division of [relevant clinicalsector at acute trust]. The Head of Divisionand the Divisional Manager will sit on thesteering group. The steering group willmeet monthly and report to the trust-wide[relevant clinical conditions] steeringgroup, which has representation fromacross the trust, the commissioners and[regional provider]. The agenda of the HITand the [clinical conditions] steering groupcorrelate strongly and support clinicalservice delivery. (HIT applicationdocument, 2012)

Redwood et al. BMC Health Services Research (2016) 16:201 Page 8 of 13

One HIT described a “key strategic advantage” of their13 member executive board as being that “these mem-bers in turn link with additional specific networks”. TheHIT planned to formalise the processes by which thiscollaboration mechanism would be implemented, anduse their executive team as knowledge channels to/fromother organisations. Apart from these more traditionalchannels for collaboration, several HITs also gave detailsof other structural mechanisms which could serve tosupport collaborative work. These included requiringdual leadership for each working group by ‘two individ-uals, encompassing academic and service leadership’(HIT application, 2013); designating individuals who haddual NHS/academic appointments as theme leaders; em-bedding HIT structures within the managerial and ser-vice delivery structures with which the HIT sought toengage; and tying specific work-stream/theme groupsstructurally to cross-cutting themes such as evidencesynthesis, inequalities, and PPI.In summary, the second process-related theme pro-

vided some evidence of the procedures and structuralconditions that were seen to support collaboration. Theorientation of the strategies being employed wasoutward-looking to involve a wide range of stakeholders,and to gain visibility and legitimacy, to facilitate growthand attract the required expertise to develop relevantprojects or work streams. Internally orientated strategiesto build and strengthen the team responsible for thedelivery of the work and to consolidate internal ways ofcollaborating were less evident in the data.

IntegrationThis theme refers to a range of activities that are ex-pected to lead to joint working across a range of spheres:service delivery, biomedical and population based/com-munity research, data linkage and data intelligence.While these activities are process-related, integrationwas also seen as an outcome in terms of the integrationof services across pathways in a unified system of healthand social care, supported by interdisciplinary researchbringing robust and high quality evidence to the carepathway from prevention to treatment and chroniccondition management to palliative/end-of-life care.Notwithstanding existing accountability relationshipsthat militate against fully integrated governance arrange-ments, the excerpt below addresses the importance ofsystem-thinking:

So …moving on beyond the language of collaborationand actually starting to see how that will work inpractice and moving towards a position whereorganisations are operating in a framework which willencourage them to make decisions on the basis ofsystem benefit and not organisational benefit. That’s a

journey we’re still on … and the HITs are important inthat regard because they are a material demonstrationof how that will work in practice. (Participant 8, seniorNHS manager)

There are a range of examples in the documentarydata that illustrate the integrative potential of HITs atvarious levels, from plans to develop specific patient-centred ‘one-stop-shops’ and integrated assessmentclinics, to streamlining strategic oversight in their areaof health and social care. In one long-term conditionarea changes which had occurred since the HIT’s forma-tion, driven in part by the political context, had led tofragmented accountability, and ‘a significant amount ofduplication and overlap across the different meetings’.Stakeholders sought a more appropriate way of ensuringwork in the relevant chronic condition area continued tobe developed and overseen. This was ‘an increasing pri-ority for commissioners in the light of staffing reductionswithin the local authority and within CCGs’. The HITwas central to a proposed new local structure for work inthe area, with various existing structures being subsumedor transformed. The objective was to reduce duplicationwhile continuing to meet statutory requirements andthose of the various stakeholders. The central role for theHIT was expected to both ‘streamline current meetings’and ‘add value to the system’ (Restructure proposal fromCCG, 2014).Many HITs sought to drive integration by working with

commissioners – both by facilitating a combined approachwith coordinated priorities in areas where commissioningpriorities were split, and by developing commissioningstrategies to support delivery of integrated care.Integration of information and data was another area

of focus for HITs, with several teams envisaging theirwork informing the Joint Strategic Needs Assessments,and some had already started bringing together data setsto inform service design and delivery. One HIT wasworking with a local CCG to develop a system of inte-grated information support and advice to GPs, to facili-tate implementation of the integrated pathway they weredeveloping. Another HIT director saw the integration ofservices as a vehicle for geographical equality of access:

‘The community we cover is quite widely spread, we’renot reaching out to our whole community with thesame excellence that we reach within this building[specialist hospital]. There was a massive disparity ofcare and [access to research]. So what (…) the HITshave enabled us to do is to create a culture … thatwill engage commissioners, will engage the public, willengage the vertical integration of primary andsecondary and tertiary care.’ (Participant 5, HITdirector, clinical academic)

Redwood et al. BMC Health Services Research (2016) 16:201 Page 9 of 13

Another type of integration was related to the flow ofinformation and feedback across the system, bringing to-gether knowledge, experience and views from differentgroups of stakeholders, including those of patients andthe public:

‘…support three-way communication between commu-nities, professionals and public health interests,through capacity building, providing an evidence baseand knowledge translation. This will enable commis-sioners to be better informed by communities aboutthe impact of their actions and to better support localneed.’ (HIT application, 2012)

A senior manager involved in the genesis of the HITmodel saw their inclusivity and longer time horizons askey factors in their integrative potential:

‘So this is a fantastically sort of diverse range of teamsworking at different levels and different scales on awhole range of different issues, but unique in the sensethat unlike the vast majority of other healthpartnerships that have kind of either explicitly orunofficially adopted a sort of clinical academic grouptype model, this one is much more inclusive. And Ithink it’s also in the longer term so much morepowerful as an engine of driving integration.(Participant 8, senior NHS manager)

The theme of integration contains elements ofprocess-related aspects, but is predominantly outcomeorientated in as much as the motivating force behind thework related to this theme was generating system-levelchange, especially in service delivery, and linking patientjourneys from prevention to the provision of secondary/tertiary care services.

InnovationThe second outcome-related theme refers to thesuccessful exploitation of new ideas and novel ways ofdelivering services or interventions through the intro-duction or application of new approaches, usually toimprove quality and decrease costs. This may includebiotechnology products and IT solutions; collaborationwith industry; new roles and/or service delivery models;and new insights through data linkage and dataintelligence. Such data linkage and insights, which wouldbe facilitated by the engagement, collaboration andintegration elements of the HIT model, could have thepotential to generate step-changes in the system:

‘The hope is that they will potentially open up newways to think about the sorts of interventions that willhave real impact. So spotting an unknown correlation

might unlock a whole different sort of area of policyinterventions that we just don’t focus on at themoment.’ (Participant 8, senior NHS manager)

Expansion of relationships with industrial partners viaHITs was another element of the model envisaged assupporting innovative work. HIT documents gave an in-dication of some of the specific innovations in develop-ment with industry partners, including novel ‘nearpatient’ testing technologies where investigations arecarried out at the time of the consultation with rapidavailability of results [21] for use where a swift diagnosishas important implications for treatment decisions, thedevelopment of technologies to facilitate access to ser-vices, and redesigning existing products to enhancesafety. Partnerships with industry were also being drawnon to access resources for innovative research whichwould not otherwise be available: funding, directly fromindustrial partners, or via applications to commercial/in-dustrial orientated grants; industrial partners’ specialisttechnology and facilities for research and development;and pools of staff who could fill capacity or expertise gaps.Engagement and collaboration with other types of or-

ganisation were also seen as potentially supportinginnovation, for example, partnering with third sector or-ganisations already known locally for their innovativepractice, or making new academic connections with aview to developing innovative interdisciplinary research.Many HIT applications included plans to pilot newinterventions and services. These involved developing/tailoring decision support tools for primary care to pro-vide local context-specific guidance, and promote agreedlocal strategies and pathways. A CCG collaborating withone HIT aimed to give ‘greater flexibility to Primary, Sec-ondary and Tertiary Care to work interactively and in-novatively to develop new ways of working’ (HITapplication, 2014) in the HIT’s area. A ‘hub and spoke’model to provide access to specialised services, whichwas developed drawing on a local centre of excellence,was actively being considered by the relevant ClinicalReference Group for early adoption as one of the re-gional pilots. Another HIT had used data collected aspart of an innovative surveillance scheme to identify andaddress service provision shortfalls.Highlighting the importance of collaboration between

providers, commissioners and researchers in generatingnovel models of care, the following participant alludes tothe complexity of implementing innovative change andthe need for context-specific and sensitive evaluation:

‘We are doing a whole bunch of real life experimentshere, driven by the need to innovate our way out ofdifficulties. Let the researchers loose on answering thequestions of ‘is this model working’? It is more subtle

Redwood et al. BMC Health Services Research (2016) 16:201 Page 10 of 13

than ‘what works and what doesn’t work’, but what arethe layers of decision-making and service change andinstitutions that are involved? How do they piece to-gether to make a final product as it were? And how isthat final product being perceived? Is it value formoney, is it good quality care?’ (Participant 4, CCGsenior manager)

HITs also developed innovative plans for educationand training – for professionals, but also for patients,carers and the public. These plans included new ways ofdelivering training for front-line staff to improve both qual-ity and access; facilitating evidence-based continued profes-sional development, and inter- and cross-professionallearning; and developing and disseminating materials to abroad range of members of the public and non-health pro-fessionals involved in caring for others, to support decisionmaking about when and where to seek health care. Plansfor innovation also included patients and the public,enabling them actively to contribute to novel products andprocesses through new media, knowledge cafes and work-shops designed for children and young people.The second outcome related theme offers some in-

sights into how conditions for innovation were beingfostered to generate locally appropriate solutions tohealth or healthcare delivery problems.

DiscussionThe aim of this study was to provide empirical evidenceabout the development of a locally-evolved model incross-sector, cross-organisational and inter-professionalworking, and to develop a theory of change for the socialand organisational processes underpinning its functions.Stripped of its specific local context, the theory ofchange we identified was that developing the specificprocesses of ‘whole system’ engagement (identifying thekey organisations and individuals who need to be in-volved in meeting joint objectives - in the case of theHITs this was patient-centred integration across healthand care pathways) and collaboration (developing themost productive ways of working together across profes-sional, disciplinary, sectorial and organisational boundar-ies) created the right conditions to produce the desiredoutcomes of integration (where organisations and indi-viduals came together to produce a good or service thatwas joined up and fit-for-purpose) and innovation(where something new was created or new connectionswere developed with something that already existed).We have provided some specific examples of how theseprocesses and outcomes were anticipated to work in thelocal Bristol context.The HIT model was developed by a partnership of

NHS organisations – providers as well as commis-sioners, two universities, a local council and including

patients and the public to foster the co-production ofknowledge and facilitate collaboration to generate healthimprovements for the local population. It shares manyof the characteristics of the ‘communities of practice’model, originally developed by Lave and Wenger [22] todescribe practice-based professional learning, and lateradapted by Wenger and colleagues [23] to the wider or-ganisational setting to help conceptualise the manage-ment and mobilisation of knowledge by diverse groupsfor particular goals. The HITs are a new organisationalform of structural partnership, designed to achieve link-ages between knowledge requirements and knowledgeproduction and as such can be described as ‘boundaryorganisations’, [24, 25] designed to facilitate collaborationand information flow between research and practicecommunities. The aim is to ensure that research isrelevant to the local health and care economy, and thatthe use of evidence-based practices and local findings isincreased. The collaborative processes described herethat to bring together knowledge producers, knowledgeusers and the wider health and care community includ-ing patients and service users are closely related to whathas been described as ‘integrated knowledge translation,broadly described as collaborations between researchersand decision-makers' [26] which builds on all partici-pants’ capacity to value each other’s very different per-spectives, providing a lever for system-wide approachesto sustainable change [27].Although these types of partnerships are not new, the

inclusion of commissioning bodies, and local government- which since 2013 has a wider remit for tackling the so-cial and economic determinants of health – is unlike otherknowledge mobilising organisations in the healthcareeconomy such as AHSCs. These typically involve largeteaching hospitals and their associated clinical services,leading research intensive universities and their medicalschools, and tend to focus on discovery and early phasetranslational work. Their operational units are typicallyorganised around clinical academics in areas such as car-diovascular services, diabetes care or psychiatry. Althoughthe AHSC model had initially found favour with a numberof key players in the local partnership, the biomedical andclinical services focus was replaced by a broader perspec-tive of population health sciences which underpinned astrategy of integration of care pathways including commu-nity services, the primary/secondary interface, publichealth and hospital care. Furthermore, the organisationallyand professionally broad partnership was set up purpose-fully to gain leverage to effect population health gains bothat pace and scale.

LimitationsThe mechanisms by which this new structural partner-ship model were thought to generate desired outcomes

Redwood et al. BMC Health Services Research (2016) 16:201 Page 11 of 13

had to be identified retrospectively, based on documen-tary evidence and the recollections of key individuals in-volved in the design of the model. Given the time lag ofup to four years since the initial meetings took place,some of the recollections may have been inaccurate andoverlaid with more recent experience and insights. How-ever, the interviewees’ accounts broadly concurred interms of timelines and the content of key documents.Documentary evidence in the form of meeting notes andminutes was available and used to compare the verbalaccounts, and again there were no significant disagree-ments on how the HIT model developed. However, thedocuments were not produced for the specific purposeof tracing the development of the model, and the inter-viewees’ accounts were inevitably partial, reflecting theirperspectives and commitments, hopes and aspirationsfor the model they were putting in place. While overlypositive presentations may have been problematic for aformal evaluation of the model, it was possible in theanalysis to identify the ‘active ingredients’ of the modelin a balanced way. Furthermore, participants were can-did and open about areas where opinions differed amongsteering group members when the HIT model was firstdeveloped and the focus shifted from clinical and bio-medical research to population-based research with apublic health and service commissioning orientation.The documents relating to the specific HITs themselvessuch as application forms were highly structured andformal, and the content was presented in ways to appealto the accrediting panel. It may have been useful to tri-angulate the documentary analysis with interviews withHIT applicants in the same way that interviews wereconducted with the designers of the model itself. How-ever, time and resource constraints militated againstsuch an approach. Furthermore, given the interest in theHITs, many of the HIT directors and members have beenaccessed several times for research and evaluation pur-poses and may have found another research invitationburdensome. While this study is not an evaluation of theHITs’ performance, the data generated give some indica-tion of where a formal evaluation of such structures mayneed to be targeted.

ConclusionsBy tracing the process- and outcomes-related themesunderpinning the development of the HIT model, wehave been able to demonstrate how a local broad-basedalliance between NHS providers and commissioners,universities and local government, not part of an eliteAHSC, has been able to design a flexible structural part-nership to support a growing number of interprofes-sional, cross sectorial and cross-organisational teamsoperating in public health and service delivery for long-term conditions. ‘Whole system’ engagement, requiring

the active recruitment of all those who have a stake inthe area of practice being considered, and collaboration,understood as a range of internally and externally orien-tated strategies to enable coproduction were identifiedas the process themes. System-level integration andinnovation were identified as outcomes with far reachingimpacts on population health and service delivery.Detailed descriptions of emerging organisational modelsof structural partnerships - an example of which is pro-vided in this article - and the identification of theirunderlying social and organisational processes are likelyto be valuable to policy makers and senior leaders in theUK health and care economy, those involved in settingup similar partnerships and teams, and health servicesresearchers in the social and implementation scienceswith an interest in their evaluation. The HIT model ispresented here to facilitate its translation to otherlocalities.

AbbreviationsAHSC, Academic Health Sciences Centre; BHP, Bristol Health Partners; CCG,Clinical Commission Group; CLAHRC, Collaborations for Leadership inApplied Health Research and Care; HIT, Health Integration Team; NHS,National Health Service

AcknowledgementsWe would like to give our thanks to the individuals who gave their time tobe interviewed. This research was funded by the National Institute for HealthResearch (NIHR) Collaboration for Leadership in Applied Health Research andCare West at the University of Bristol NHS Foundation Trust. The viewsexpressed are those of the authors and not necessarily those of the NHS, theNIHR or the Department of Health. The funding body had no role in thedesign, collection, analysis, or interpretation of data in the writing of themanuscript, or in the decision to submit the manuscript for publication.Documentary data were provided by Bristol Health Partners, but theorganisation had no role in the design of the research, and the viewsexpressed here are not necessarily those of Bristol Health Partners. We thankDr Lesley Wye for her thoughtful comments on earlier versions of this article.

Availability of data and materialsData from this study will not be made available for the following reasons:although the documentary data have been anonymised, given the contextualinformation that has to be retained in order to make the data intelligible, theidentity of individuals concerned could be ascertained. In relation to theinterview data, because of the small sample and the seniority ofthe individuals,it would also not be possible to protect individual’s anonymity.

Author contributionsSR, JD and JH conceived and designed the study. SR, EB and VL collectedand analysed the data. SR produced the first draft of the manuscript and allauthors contributed to the development of the final manuscript which hasbeen approved by all authors.

Author’s informationEB and VL are joint second authors.

Competing interestsThe authors declare that they have no competing interests.

Consent for publicationParticipants whose interview excerpts are used in this article have given theirconsent to publish.

Redwood et al. BMC Health Services Research (2016) 16:201 Page 12 of 13

Ethics approval and consent to participateApproval for the study was given by the Faculty of Medicine and DentistryCommittee for Ethics at the University of Bristol (reference number 131451/10681).

Received: 28 October 2015 Accepted: 6 June 2016

References1. Lockett A, El Enany N, Currie G, Oborn E, Barrett M, Racko G, et al. A

formative evaluation of Collaboration for Leadership in Applied HealthResearch and Care (CLAHRC): institutional entrepreneurship for serviceinnovation. Health Serv Deliv Res. 2014;2(31). http://www.ncbi.nlm.nih.gov/books/NBK260030/. doi:10.3310/hsdr02310.

2. Dzau VJ, Ackerly DC, Sutton-Wallace P, Merson MH, Williams RS, Krishnan KR,et al. The role of academic health science systems in the transformation ofmedicine. Lancet. 2010;375:949–53. doi:10.1016/S0140-6736(09)61082-5.

3. Graham I, Tetroe J. Learning from the U.S. department of veterans affairsquality enhancement research initiative: QUERI Series. Implement Sci.2009;4:1–6. doi:10.1186/1748-5908-4-13.

4. Walshe K, Davies HTO. Health research, development and innovation inEngland from 1988 to 2013: from research production to knowledgemobilization. J Health Serv Res Policy. 2013;18(3 suppl):1–12.

5. Rycroft-Malone J, Wilkinson J, Burton CR, Harvey G, McCormack B, Graham I,Staniszewska S. Collaborative action around implementation inCollaborations for Leadership in Applied Health Research and Care: towardsa programme theory. J Health Serv Res Policy. 2013;18(3 suppl):13–26.

6. Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou O. Diffusion ofinnovations in service organizations: Systematic review andrecommendations. Milbank Q. 2004;82:581–629. doi:10.1111/j.0887-378X.2004.00325.

7. Lomas J. The in-between world of knowledge brokering. BMJ. 2007;334:129–32. doi:10.1136/bmj.39038.593380.

8. French CE, Ferlie E, Fulop NJ. The international spread of Academic HealthScience Centres: A scoping review and the case of policy transfer toEngland. Health Policy. 2014;117(3):382–91.

9. Wehrens R, Bekker M, Bal R. Within The Netherlands, Dutch AcademicCollaborative Centres for Public Health: Development through time – Issues,dilemmas and coping strategies. Evidence Policy. 2012;8:149–70.doi:10.1332/174426412X640063.

10. Bristol Health Partners. http://www.bristolhealthpartners.org.uk/health-integration-teams/. Accessed 23 May 2016.

11. Ham C, Baird B, Gregory S, Jabbal J, Alderwick H. The NHS under thecoalition government - Part one: NHS reform. London: The King’s Fund,2015. http://www.kingsfund.org.uk/publications/nhs-under-coalition-government. Accessed 23/09/2015)

12. Gridley K, Spiers G, Aspinal F, et al. Can general practitioner commissioningdeliver equity and excellence? Evidence from two studies of serviceimprovement in the English NHS. J Health Serv Res Policy. 2012;17(2):87–93.http://dx.doi.org/10.1258/jhsrp.2011.010176.

13. Weiss CH. Evaluation Research: Methods of Assessing Program Effectiveness.New Jersey: Prentice-Hall; 1972.

14. Weiss C. Nothing as Practical as a Good Theory: Exploring Theory-BasedEvaluation for Comprehensive Community Initiatives for Children andFamilies. In: Connell J, Kuchisch A, Schorr LB, Weiss C, editors. NewApproaches to Evaluating Community Initiatives: Concepts, Methods andContexts. New York: Aspen Institute; 1995. p. 65–92.

15. Davidoff F, Dixon-Woods M, Leviton L, Michie S: Demystifying theory and itsuse in improvement. BMJ Qual Saf 2015. doi:10.1136/bmjqs-2014-003627

16. Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitativeresearch (COREQ): a 32-item checklist for interviews and focus groups.Int J Qual Health Care. 2007;19(6):349.

17. Ritchie J, Spencer L. Qualitative data analysis for applied policy research. In:Bryman A, Burgess R, editors. Analysing qualitative data. London: Routledge;1993. p. 173–94.

18. Gale N, Heath G, Cameron E, Rashid S, Redwood S: Using the frameworkmethod for the analysis of qualitative data in multi-disciplinary healthresearch, BMC Med Res Methodol. 2013:117; doi:10.1186/1471-2288-13-117

19. Glaser BG, Strauss AL. The discovery of grounded theory. Chicago: Aldine;1967.

20. Fischer MD, Ferlie E, French C, Fulop N, Wolfe C: The creation and survivalof an academic health science organization: counter-colonization through a

new organizational form?. University of Oxford, Saïd Business School; 2013.http://www.sbs.ox.ac.uk/sites/default/files/SBS_working_papers/creationandsurvival.pdf. Accessed 23/09/2015

21. Khunti K. Near-patient testing in primary care. Br J Gen Pract.2010;60(572):157–8. doi:10.3399/bjgp10X483454.

22. Wenger, Etienne, Jean L. Situated Learning: Legitimate PeripheralParticipation. Cambridge: Cambridge University Press; 1991.

23. Wenger E, McDermott RA, Snyder W. Cultivating communities of practice: Aguide to managing knowledge. Boston, MA: Harvard Business Press; 2002.

24. Guston, David H. Boundary organizations in environmental policy andscience: an introduction. Sci Technol Hum Val. 2001;26(4):399–408.

25. Parker J, Crona B. On being all things to all people: Boundary organizationsand the contemporary research university. Soc Stud Sci. 2012;42(2):262–89.

26. Gagliardi AR, Berta W, Kothari A, Boyko J, Urquhart R. Integrated knowledgetranslation (IKT) in health care: a scoping review. Impl Sci. 2016;11(1):1.

27. Kothari A, Wathen CN. A critical second look at integrated knowledgetranslation. Health Policy. 2013;109(2):187–91.

• We accept pre-submission inquiries

• Our selector tool helps you to find the most relevant journal

• We provide round the clock customer support

• Convenient online submission

• Thorough peer review

• Inclusion in PubMed and all major indexing services

• Maximum visibility for your research

Submit your manuscript atwww.biomedcentral.com/submit

Submit your next manuscript to BioMed Central and we will help you at every step:

Redwood et al. BMC Health Services Research (2016) 16:201 Page 13 of 13