Healthcare Policy

Healthcare Policy 1(2) January 2006 : 49-54.doi:10.12927/hcpol.2006.17875
Perspectives on Evidence, Synthesis and Decision-Making

A Decision-Maker's Perspective on Lavis and Lomas

Rick Roger


Advancement in research synthesis, so well articulated and advocated by Lomas (2005) and Lavis et al. (2005), is a necessary but not sufficient development for the systematized use of research in managerial practice. Although significant progress in the use of evidence-based approaches can be witnessed, enriching efforts need to progress within both the management and research communities. Contextualized by direct experience with harm reduction approaches to population health improvement, this commentary offers propositions concerning the nature of the researcher/decision-maker interchange, arguing for a pronounced strengthening of involvement and activity at all levels in the service delivery system.

[To view the French abstract, please scroll down.]

A conference entitled "Practical Strategies for Cross Sectoral Allocation of Resources to Improve Health," organized by the Milbank Memorial Fund with a supporting grant from the Robert Wood Johnson Foundation, was held June 14-16, 2000, in New York City. Work of the Vancouver/Richmond Health Board in support of socially marginalized residents attracted attention, and I was invited as regional CEO to participate on a panel discussing housing as a health status determinant. The conference, attended by 22 academic and healthcare policy makers from the United States, Canada, the United Kingdom and New Zealand, was dedicated to the well-established but still debated contention that investments in the health sector alone may not be sufficient to improve population health (Robert Wood Johnson Foundation 2002).

Armed with documentation on the Health Board's rationale and decision-making process, I attempted to establish that a positive health impact had been achieved through a multifaceted approach involving not only housing and the purchase and regeneration of derelict hotels, but also policing, direct services provision, increased support of funded community service organizations, a variety of partnership efforts and funded (drug-using) consumer involvement. Somewhere buried in the Milbank archives will be the report of the conference with a paraphrase of my remarks: "Several things were attempted to improve population health in the downtown eastside of Vancouver; something worked; Mr. Roger has no idea what."

Suitably humbling, but also instructive, this account illustrates prevailing policy development dynamics, contextualizing the managerial reaction to the deliberations of Lomas (2005) and Lavis et al. (2005) in the first and current issues of Healthcare Policy. Confronting the gap between the idealized use of research in policy development and current realities, both authors recognize that healthcare managers and decision- makers do not function solely within the simple world of "What works?" The policy making environment is more a function of "What combination of interventions works where, for which sub-populations, in which environmental circumstances, in which combinations, administered at what rate of intensity, over which period of time and in what order?" Complexity of this nature defines the decision-making role in regional health services delivery where the relations between cause and effect are often only retrospectively coherent.

Lomas and Lavis et al. lay out both diagnostic journeys - examining the methods of research development and synthesis - and remedial journeys contemplating improvements that might be "bootstrapped" from existing methodological approaches and established relationships. In both instances, researcher effectiveness is the focus, and the remedial journey is presented from the perspective of the research community. Policy making and managerial contingents are considered rather more as destinations for the research effort than as fellow travellers as knowledge is gained. We learn from these papers that managers and host organizations will certainly function as entry points, signposts and way-stops, but hardly as road engineers or route-masters. And while there is nothing intrinsically wrong with this perspective - a little expertise can be a dangerous thing in the wrong circumstances - the obligations of policy makers are underplayed as part of the solution set that is advanced in both papers.

Colleagues in the forest industry and the biological sciences sector deploy a well-travelled phrase to portray the interchanges at issue: "Gaelic poetry for deaf seagulls," the construct engendered when the precision of research design and expression demanded by peer-reviewed research (and that sanctioned by the systematic review process) conflicts with the functional applicability and degree of generalizability expected of the research product (Larkin and Pallister 1976; Baskerville 1997). From any health services management perspective, the movement towards "user-friendly" and easily retrievable "poetry" so well described in these papers is of unquestionable value. There is a parallel argument, however, that the "hearing" or reception ability of the management community also needs attention. Deaf seagulls are not well positioned to inform the research agenda or to introduce research into practice.

Lavis and his co-authors (2005) record activities recommended for health services and policy researchers interested in shaping the products of their efforts for healthcare managers and policy makers. A counterpart list for managers might also be advanced, including:

  • cultivation of ongoing relationships with established and emerging researchers;
  • disciplined efforts to involve the research community as new initiatives are contemplated, well in advance of the implementation stage and with follow-through as implementation progresses;
  • joining the conversation in areas of interest, recalling the ultimate accountability of researchers, policy makers and journalists to the person "serving coffee in the doughnut shop" (Waddell et al. 2005);
  • opening organizations to scrutiny, accepting that occasional embarrassment can be the source of inspiration and improvement;
  • encouraging developmental efforts inside organizations, learning how to understand and appreciate research;
  • development of staff exchanges and secondments between and among research organizations, delivery organizations and knowledge brokering organizations;
  • managing the opportunity to broker connections and knowledge exchange between researchers in different areas of specialization;
  • involvement in those peer-review activities structured with a "decision-maker" component contributing to research effectiveness, learning how researchers critique one another;
  • encouraging communities of practice within and without organizations, activating opportunities for learning at organizational boundaries;
  • modelling the way for others in the use of research; sparking evidence, challenging the status quo; and
  • following the lead of some of the best-regarded healthcare leaders, writing and recording personal and organizational research and development efforts.

Hearing-assisted "seagulls" will help shape the research agenda towards the shared goal of improved system performance.

What, then, are the lessons to be learned and applied from the Vancouver experience recounted earlier? Four working propositions help frame the thinking stimulated by the comments of Lomas (2005) and Lavis et al. (2005):

  • Proposition 1: Migration or outright changes in the question(s) under review should be expected in the context of the researcher-policy maker interchange. Vancouver's downtown eastside initiative started as a response to escalating HIV infection rates among intravenous drug users, but quickly progressed to a focus on drug overdose deaths. Neither researchers nor managers had the luxury of "fixing the question" as the Vancouver/Richmond Board responded to pressures for encompassing approaches.
  • Proposition 2: The "what are the issues around doing Y" form of question articulated by Lomas (2005: 58) and expanded by Lavis et al. (2005) in the Cochrane context is of prime importance and should not be discounted as researchers address issues of interest to managers. Externalities (anticipated or not) are consequential in the public policy process. The Vancouver/Richmond Board, the Board's predecessor organizations and its successors have all faced significant challenges in the implementation of harm-reducing approaches to population health improvement.
  • Proposition 3: While intriguing and potentially useful in some respects, the macro-level, integrated source of answers to questions contemplated in both papers is unlikely to add much value to policy developers involved in the introduction and management of significant changes in priorities or in delivery arrangements. The need for program evolution does not manifest in discrete, individually measurable steps. Partners involved in the Vancouver initiative could not stage policing measures in a different time or location from the housing or street service measures. Systematic reviews would have assisted in the roll-out of components more than in the shaping of the overall agenda. We had access to information on how best to respond to the AIDS epidemic; we knew something of the merits of outpatient versus inpatient approaches to the treatment of addicted populations; and we had research-informed perspectives on the need for housing. Systematic approaches could have improved our understanding, but no integrating synthesis would or could have been expected to respond fully to the interlaced agenda and the accompanying needs for research guidance.
  • Proposition 4: Researchers and policy makers have moved beyond denial as respective roles are contemplated. The next step is to learn together (perhaps the hard way, as suggested) how best to conduct and disseminate the findings of systematic reviews. In the research context, mistakes were made in Vancouver. Surrounded by well-regarded researchers, Board members and staff did utilize local expertise, but not with the degree of commitment needed for enduring partnership. More could have been learned; rapid-response capacity emulating the "client-contractor" situation set out by Lomas (2005: 60) could have been established; and the CEO could have been more convincing in New York!

In the summer of 1987, the organizers of "Connections 88," a symposium dealing with research and public policy on aging and health, asked for "views from the field" concerning barriers to the use of research. Seeking input from executive-level officials through an interview and survey approach, 15 detailed responses were received from British Columbia through Ontario. While knowing little of developing approaches to "theme analysis" taking shape in research literature at that time, I recorded a significant degree of skepticism among the respondents; there was not much hope for the evolution of research-informed policy development (Roger 1989). Most decision-makers would agree that the role of research in policy has steadily advanced over the intervening decade and a half, with the development of capacity at all levels of the system. Canada may indeed be "leading the charge in exploring new ways of doing synthesis for healthcare managers and policy makers" (Lomas 2005: 56). Lavis and colleagues (2005), while adopting a differentiated perspective, join Lomas in the sensible ordering of ideas needed for further advancement, enabling the effective deployment of resources now in place. Full realization of potential gains will require constructive efforts in both the research and decision-making communities.


Point de vue d'un décideur sur les articles de Lavis et Lomas


Les progrès réalisés dans la synthèse des preuves, si bien articulés et présentés par Lomas (2005) et Lavis et al. (2005), sont un développement nécessaire mais insuffisant pour assurer une utilisation systématique de la recherche dans le travail des gestionnaires. Bien qu'on observe des progrès significatifs dans l'utilisation des méthodes fondées sur les preuves, davantage d'efforts doivent être déployés pour promouvoir cette utilisation au sein des communautés de gestion et de recherche. Contextualisé par une expérience directe dans les méthodes axées sur la réduction des préjudices en ce qui a trait à l'amélioration de la santé de la population, ce commentaire offre des propositions concernant la nature de l'échange entre chercheurs et décideurs et préconise un accroissement prononcé de la participation et des initiatives à tous les paliers du système de prestation de services.

About the Author(s)

Rick Roger, MHSA
Co-Director, Centre for Health Care Management,
University of British Columbia, Vancouver, BC


Correspondence may be directed to: Rick Roger, email:


Baskerville, G.L. 1997. "Gaelic Poetry for Deaf Seagulls; Encore." Forestry Chronicles 70: 562-564.

Larkin, P. and A.E. Pallister. 1976. "Gaelic Poetry for Dead Seagulls: An Essay on Research Funding." Issues in Canadian Science Policy 2: 3-11.

Lavis, J., H. Davies, R. Gruen, K. Walshe and C. Farquhar. 2005. "Working Within and Beyond the Cochrane Collaboration to Make Systematic Reviews More Useful to Healthcare Managers and Policy Makers." Healthcare Policy 1(2): 21-33.

Lomas, J. 2005. "Using Research to Inform Healthcare Managers' and Policy Makers' Questions: From Summative to Interpretive Synthesis." Healthcare Policy 1(1): 55-71.

Robert Wood Johnson Foundation Grant Results Report. 2002. "National Spending on Health Care May Be Too Narrowly Allocated." Retrieved November 21, 2005. <

Roger, R. 1989. "Profession without Practice: An Administrator's Perspective on Barriers Restricting the Use of Research in Public Policy." In S.J. Lewis (ed.), Aging and Health: Linking Research and Public Policy. Chelsea, MI: Lewis Publishers pp. 379-390.

Waddell, C., J. Lomas, J. Lavis, J. Abelson, C.A. Shepherd and T. Bird-Gayson. 2005. "Joining the Conversation: Newspaper Journalists' Views on Working with Researchers." Healthcare Policy 1(1): 123-139.


Be the first to comment on this!

Note: Please enter a display name. Your email address will not be publically displayed