Knowledge Translation into Policies and Programs: Lessons from 18 Years’ Experience

1. Knowledge Translation into Policies and Programs: Lessons from 18 Years’ Experience Professor John Frank Director Scottish Collaboration for Public Health Research and Policy, Chair, Public Health Research and Policy, University of Edinburgh

2. Outline • Background: “What is knowledge transfer” • Transfer to whom? Which knowledge? – a suggested typology • Some observations from the author’s Canadian experience: • Provostial Advisor for Population Health, University of Toronto, 1994-7 • Research Director, Institute for Work and Health, 1991-2008 • Scientific Director, Canadian Institutes of Health Research: “KT” (2000-8) as seen from a PH viewpoint • Chair, Advisory Council, National Collaborating Centres for Public Health -- Public Health Agency of Canada (2004-9) • Progress Report: Scottish Collaboration for Public Health Research and Policy (2008-9) • Final thoughts

3. Background: What is “Knowledge Transfer”? • Virtually every major health research funding agency in the Western world has embraced the need to foster “knowledge transfer”: ”The application of research results to policy, program and practice settings, via the synthesis and translation of findings for use by decision-makers in those settings – i.e. getting knowledge into action.” • The most sophisticated of these efforts – e.g. Canadian Health Services Research Foundation (founded by Jonathan Lomas a decade ago) – emphasize several key principles of successful “KT” in the worlds of health services and public health – as opposed to the world of translational (bio)medicine (i.e. “bench to bedside”):

4. Some Established KT Principles (focussed on HS and PH Research) • Joint researcher and decision-maker planning and execution of research, from the start • Selection and “framing” of the research question(s), to be addressed, so as to speak to decision-maker needs • Pro-active intelligence-gathering, throughout the project, about probable decision-maker reactions to the results, and their changing needs, informing strategic updates of project KT strategy • Passive dissemination of results through traditional academic channels (conferences, publications) is not enough – the media utilized must fit the audience • KT process must be driven by both research-user “pull” and by targeted, inter-active researcher “push” • Timelines can be long – James Lind and citrus for scurvy Reference: Lomas J.Using linkage and exchange to move research into policy at a Canadian foundation. Health Affairs 2000; 19(3):236-240.

5. ASIDE: Why did Lind’s RCT and other scurvy research take > 4 decades to influence Navy policy? • In a thoughtful revaluation of Lind’s work, Michael Bartholomew of the Open University points out*: • Lind himself waffled for decades on his own conclusions, and much of his writing was impenetrable • Powerful alternative scientific viewpoints had strong theoretical reasons to reject the idea of a nutritional deficiency • Many competing hypotheses had equivalent anecdotal support (the evidentiary standard of the time) • The Royal Navy did not have any precedent for science-based decision-making – at least in the health field *Bartholomew, M. James Lind and scurvy: a revaluation. J Maritime Res 2002, available at: http://www.jmr.nmm.ac.uk/server/show/conJmrArticle.3/viewPage/1

6. General Principles are fine, but… • Experience shows that best practices in KT vary greatly across the primary audience of research-users being targeted • Some widely recognized types of audiences, necessitating different KT approaches, include: • The general public, including patient populations with particular conditions • Health professionals in clinical practice • Managers of health services -- both clinical and public health • Higher-level policy-makers (usually public sector, but sometimes “third sector” and private sector as well) ERGO: “TRANSFER TO WHOM?” IS A CRITICAL CONSIDERATION

7. What (sorts of) knowledge? • Some KT audiences have strong preferences for certain sorts of “knowledge”: • The public: often prefer an engaging narrative, with compelling examples – qualitative research has the edge; but the costs of far-reach and long-impact are very high, as all ad-men know! • Clinical professionals: prefer pre-digested knowledge (e.g. relevant studies have been found, critically appraised for quality and crisply summarized) and then automatically “pushed” to them– preferably by a trusted and arm’s length source – in a quickly scan-able format, ideally integrated into their work (e.g. in primary care; screen pop-ups cued by key words in the client’s reason for the encounter, such as “Chief Complaint,” or “Age and Gender” for preventive measures.) • Higher-level program managers and policy-makers: virtually never make major decisions alone; they need the use of evidence integrated into their organizations’ normal groupprocesses of decision making; very concise lay-versions of “the facts,” in order to be read by such busy persons, should come in at least 3 sizes: 1,3,5+ pages. ERGO,“WHICH KNOWLEDGE?” IS ALSO A CRITICAL CONSIDERATION

8. Experience # 1: Making Heath Data Maps: a Community-University Research Collaboration (University of Toronto, 1994-7) • Experience* with researcher cumresearch-user (i.e. decision-maker) consortia -- for planning/designing, executing and using applied research of interest to both -- suggests: • “cultural” differences between these two worlds are significant, and take time to bridge –e.g. their training, underlying assumptions, language, incentives and reward systems at work, competing demands, time-scales, etc. *Buckeridge DB, Mason R, Robertson A, Frank JW, Glazier R, Purdon L, et al. Making health data maps: a case study of a community/university collaboration. Social Science & Medicine 2002; 55(7):1189-1206.

9. Lessons from Researcher/User Consortia (cont’d) • The wide range of research questions that are relevant in such work, especially if it involves new policy or program interventions, often spans: • environmental scans – i.e. “who has done what in this field?”; • syntheses of available evidence, of many kinds; • detailed social science studies of new policy or program “acceptability” among key stakeholders; • novel effectiveness trials to assess both mean effects and subgroup effects (re equitability); • health economic studies; • detailed implementation and scalability studies; and • policy-analytic studies on the facilitators and barriers to adoption of a new intervention.

10. Lessons from Researcher/User Consortia (cont’d) • The mix of methodological approaches therefore necessitated is often very broad, requiring more than one university’s/research-centre’s engagement and a trans-disciplinary approach (sometimes requiring specialized peer-review, and partnered funding by a range of granting agencies and other sources.) • Trans-disciplinary research is a tender plant, requiring fertile soil, and some “deeply silo’d” universities just aren’t suitable for such experiments!

11. Experience # 2:Institute for Work and Health - Toronto (1991 to 2007) • Some fields of applied research have particularly untrusting stakeholders, with long histories of “contended science” – e.g. occupational health • IWH was set up to overcome this distrust, and selected an “edge of campus,” independent-institute model, with several key features: • Seconded professors from >1 university, and many grad students, all well funded in a welcoming setting • Multi-partite BoD, with reps from government and the university, in addition to labour, management, and the insurer (providing the core funding – $5 million p.a.) • Peer-reviewed grants provide complementary funding • Separate peer-review of all scientific activities, initially twice annually, by an international Scientific Advisory Committee (first Chair: Sir Michael Marmot, 1991-1996).

12. KT Lessons from IWH, 1991-2008 (cont’d) • Some topics are too contentious to research in a way that all stakeholders will agree on – e.g. documenting exposure to established toxic substances in the workplace needs strong regulatory/legislative action FIRST [“You can’t do research in a bar-room brawl”]. • Some kinds of research require YEARS of trans-disciplinary integration, and new methods development, before key tools are ready to tackle large empirical projects – e.g. learning to use workers compensation insurance (administrative) data for research on the long-term prognosis of low back pain/RSIs. Reference: IWH website: www.iwh.on.ca

13. KT Lessons from IWH, 1991-2008 (cont’d) • “Keeping your reputation for independence and integrity intact” is the only route to long-term survival: the edge-of-campus model has many advantages in this regard. • The full occupational health KT process, from doing new research to fully engaging the OSH community in its use, takes many years – one must be patient! Reference: IWH website: www.iwh.on.ca

14. Experience # 3: The KT Journey at theCanadian Institutes for Health Research, 2000-2008 • In 2000, the Canadian federal government agreed to greatly increase flagging health research expenditures IF there was a greater commitment, by the funding agency – the old MRC (Canada) – and its grantees, to move research results towards application, of all kinds. • This led to the formation of the Canadian Institutes of Health Research, with 13 separate Institutes, each targeting a major field, but with only 1% of the budget – which has not increased in the ensuing eight years. • The Institute of Population and Public Health found itself as the only national-level funder dedicated to that field – which was very under-developed (no CDC Atlanta, and no Schools of Public Health, in the whole country.) • The pressure to “do something” for KT was relentless, but the funding was not there to do it properly (given competing demands).

15. CIHR- Institute of Population & Public Health Mission Statement The CIHR-IPPH will support: • research into the complex interactions (biological, social, cultural, environmental), which determine the health of individuals, communities, and global populations; and, • the application of that knowledge to improve the health of both populations and individuals, through strategic partnerships with population and public health stakeholders, and innovative research funding programs.

16. Reflections from CIHR’s Institute of Population and Public Health • Before we could address a knowledge translation agenda, we had to strengthen the capacity of the public health infrastructure to effectivelyuse population and public health (PPH) evidence • 2003: “Future of Public Health” Report, signalling the fragility of the Canadian public health system (before SARS and the Naylor Report) • 2005: Major concept paper to guide a knowledge synthesis, transfer and exchange in public health, which is now informing the six National Collaborating Centres for Public Health* • Created individual and structural incentives to meaningfully engage public health policy makers and practitioners in research planning and governance, in order to foster prompt application of knowledge to PPH policy and practice: • Centres for Research Development • Applied Public Health Chairs *Reference: Kiefer L. et al. Can J Publ Hlth 2005; May-June: I1-I40.

17. Lessons from CIHR, 2000-2008 (cont’d) Some things at CIHR had to change: • Previously only universities could administer grants and essentially only professors could hold them – not consortia involving users. • Over-reliance on passive, response-mode granting – best suited to the basic biomedical sciences in the lab – had held back PPH and HS (and even some clinical) research in Canada, due to lack of strategic (needs-led) investments, based on analysis of infrastructural gaps.

18. Lessons from CIHR, 2000-2008 (cont’d) • Decades of unchanged structures and processes for research training had led to a professoriate only familiar with the passive, response-mode granting model – and very protective of it. • They were thus unprepared to lead Training Programs, Centres and Networks that include KT in their mandate (e.g. requiring, for funding eligibility, co-governance with a research user community) – and their institutions likewise. • All of this has been started – but much of it will take a whole generation of researchers to be replaced, before it can be completed: Thomas Kuhn’s “paradigm shift.”

19. Canada’s National Collaborating Centres for Public Health (2004-2009) • As the lack of resources became evident at CIHR to “do KT for public health in Canada,” fate came to the rescue: • A major SARS outbreak in Toronto, in early 2003, led to the formation of Canada’s first semi-autonomous Public Health Agency (PHAC) and the appointment of a Chief Public Health Officer, and a sub-Minister for Public Health • The Agency soon realized that no one was responsible for either KT or professional continuing education/ upgrading for any public health professionals in Canada – there had been total neglect, and for years, buck-passing across federal/provincial/local governments. • The Agency established, by six arm’s length grants, six Centres across Canada to “do KS(=Synthesis)TE (=Exchange)” for public health professionals and decision-makers in six key topic areas. Reference: Frank JW, Di Ruggiero E, Mowat D, Medlar B. Developing knowledge translation capacity in public health: the role of the National Collaborating Centres. Canadian J Public Health 2007;July-August:1-12

20. Six NCCs across Canada • NCC for Environmental Health - British Columbia Centre for Disease Control (BCCDC), Vancouver - Scientific Lead - Dr. Ray Copes • NCC for Aboriginal Health -University of Northern British Columbia, Prince George, BC - Scientific Lead - Dr. Margo Greenwood • NCC for Infectious Diseases - International Centre for Infectious Diseases, Winnipeg - Scientific Lead - Dr. Margaret Fast • NCC for Methods and Tools - McMaster University, Hamilton - Scientific Lead - Dr. Donna Ciliska • NCC for Healthy Public Policy - Institut national de santé publique du Québec (INSPQ), Montréal - Scientific Lead - Mr. Francois Benoit • NCC for the Determinants of Health - St. Francis Xavier University, Antigonish, NS - Scientific Lead - Ms. Hope Beanlands

21. The NCCs from Coast to Coast Prince George, BC Antigonish, NS Vancouver, BC Winnipeg, MB Montreal, QC Hamilton ,ON

22. Some Lessons Learned from the NCCs • Separation of “KSTE” units from research environments, whilst facilitating their freedom to engage with user-communities, comes at a price – skilled research synthesizers are hard to recruit and retain. • Some user audiences – e.g. aboriginal community groups – require creative use of novel summaries of “traditional knowledge.” • Dedicated public-sector funding is key, distinct from research funding per se; but long-term sustainability likely requires co-funding from employers of public health professionals, and through “professional development fees” added to their annual association dues/ professional college memberships.

23. The Scottish Collaboration for Public Health Research and Policy: Background • April 2001: DoH “Pattison” Report: A Research and Development Strategy for Public Health: • Need to “involve users in all parts of the research process” and “improve the current evidence base, including prioritizing new public health research.” • February 2004: “Wanless” Report: Securing Good Health for the Whole Population: • Recommended “strengthening public health research;” highlighted “the need for greater investment in intervention research” and “greater links between academia and practitioners” to achieve research on a “greater scale.” • March 2004: Wellcome Trust “Frankel” Working Group Report: Public Health Sciences: Challenges and Opportunities: • U.K. needs to: “re-establish public health partnerships between universities and the NHS…to bring together public health science, social science, and public health service delivery.” & “develop more evidence-based policies.”

24. SCPHRP Background (cont’d) • 2006-7 (reported June 2008): UK Clinical Research Collaboration (CRC) Public Health Research Strategic Planning Group (Chair: Prof. Ian Diamond) • Recommendations included: • “Multidisciplinary and collaborative working should be encouraged both within the public health research community and between academics, practitioners and policy makers.” • “.. need for… more research evaluating interventions and policies.” • This Report led to the recent (2008) funding of five CRC Centres of Excellence in Public Health, based in Belfast, Cambridge, Cardiff, Newcastle, and Nottingham, each focused on a specific area of PH research (Scientific Advisory Panel chaired by Prof. Sally Macintyre.)

25. Founding of the SCPHRP • 2006: meetings of Scottish public health research and policy/program and practice communities examined the best way to strengthen the field: • Recommended that “pump-priming funds” from the MRC and Chief Scientist Office be used to set up a Scottish Collaboration for Public Health Research and Policy, to move forward on the above recommendations in the Scottish context • Mid-2007: International competitive search led to hiring of Director, who made several introductory visits, taking up the post in July 2008.

26. Mission of the Scottish Collaboration for Public Health Research & Policy • To identify key areas of opportunity for developing novel public health interventions that equitably address major health problems in Scotland, and move those forward. • To foster collaboration between government, researchers and the public health community to develop a national programme of intervention development, large-scale implementation and robust evaluation. • Build capacity within the public health community for collaborative research of the highest quality, with maximum impact on policies, programs and practice.

27. The Process • SCPHRP will convene a series of consensus workshops to prioritise potential interventions for development, and to establish a series of Working Groups organised around key prevention opportunities in the life course: • Early years • Teenage and early adulthood • Early to mid-working life • Later life • Each Working Group will develop a three-year work programme designed to support the development and piloting of a few promising and novel interventions, eventually at the national program and policy level. • SCPHRP will facilitate the work of the Working Groups and provide limited pump-prime funding, as well as direct support. • Depending on the outcome of these preliminary studies, the final outputs from the Working Groups should be large-scale intervention-grant submissions to U.K. and Int’l agencies, by 2012.

28. Aims of Initial SCPHRP Workshop (Edinburgh, Jan. 27-28, 2009) • To initiate four SCPHRP Working Groups: • Early life • Teenage and early adult life • Early to mid-working life • Later life • To identify prioritized opportunities for the further development, and robust testing, of promising (but unproven) public health program and policy interventions that could equitably improve Scotland’s health, for Working Group action

29. Overview of Jan. 27-28 Workshop Results • Superb participation: over 85% of the 70+ invitees attended, representing diverse fields of: • Public health professional practice, program management, and policy-makers – the users of public health research • Researchers from over a dozen different academic disciplines, representing over a half-dozen universities and independent research units across Scotland

30. Partial Summary of Priority Intervention Categories Selected by SCPHRP Workshop Life-course Groups • Early Life: interventions to improve parenting, especially for high-risk families, with special attention to maternal-infant mental health outcomes • Teenage and Early Adult Life: interventions, including high-risk targeting, to facilitate social, cultural and family connectedness, and mentoring, so that young people make sound decisions about health-related behaviours and, manage life transitions successfully • Early to Mid-Working Life: Interventions to tackle our obesogenic environment: socio-cultural and physical; work and health interventions – e.g. to prevent sickness absence • Later Life: Interventions to maintain function and independence as long as possible, so as to reduce unnecessary or premature disability and dependency

31. Subsequent Working Group Development • 58 members of the public health community (43 research; 15 policy/practice) agreed to participate in one or more of 4 Working Groups • Each WG met for a half day meeting in late April • Meeting aims were to: • Consider the priority areas identified at the consensus workshop • Confirm that areas identified are still of interest and refine further • Identify potential contributions of WG members relevant to the areas identified. • Process of and outcomes from WGs differed, with Working Life and Later Life WGs making most progress towards agreeing focus of their future Work Plans • Contributions from policy leads revealed that there was a considerable amount of policy driven activity relevant to the Early Life and Adolescence WGs • This underlined the importance of a continuing dialogue between policy and research and practice: • to ensure a common understanding among all WG members • to identify a particular niche for activity of WGs

32. Building SCPHRP Capacity • 4 SCPHRP Career Development Fellowships (3 years) • Remit to support the activities of the WGs • Excellent response to job advert • 4/6 candidates interviewed offered posts • All have accepted; one started July 1 • Other three in post by September 2009 • SCPHRP Internship for MRC PhD Student (1 year placement) • 1 candidate accepted placement • Specialist Registrar (Public Health) Trainees (6mth placement) • Agreement in principle for SCPHRP to become accredited training placement • Opportunities for SCPHRP placements will be publicised through Scottish StR network

33. Evaluation of SCPHRP: Initial Thoughts • SCPHRP represents a bold experiment in capacity building in public health leading to the development, implementation and evaluation of public health interventions • AND should be subject to independent evaluation (both formative and summative) • An initial step will be to develop a preliminary logic model in conjunction with key stakeholders, which outlines inputs, activities and outcomes of the collaboration and sets them within a timeframe. • The preliminary logic model will provide the basis for a commissioning evaluation which will be by competitive tender (Autumn 2009).

34. To conclude … • Research need not always inform current, or imminent policies, programs or practices – even in public health. • But much research that aspires to achieve this goal could be organized and conducted so as to improve the chances of doing so • Co-planning and –execution of applied public health intervention research is a special opportunity in this regard, one that “deserves a fair trial” • Scotland has particular strengths as a natural laboratory for this kind of work.

35. How to Reach Us John Frankjohn.frank@hgu.mrc.ac.uk Sally Hawsally.haw@hgu.mrc.ac.uk Caroline Reescaroline.rees@hgu.mrc.ac.uk Human Genetics Unit Western General Hospital, Crewe Rd., Edinburgh EH4 2XU Tel 0131-332-2471, ext. 2119