Your brief is to develop innovative strategies which will encourage the widespread, sustainable and systemic adoption of evidence-based practice, with the goal of improving patient care. These strategies must be capable of being tested in different healthcare settings and of involving both private practitioners, including general practitioners, and the public sector. You should also suggest plans for their evaluation, including impact on patient outcomes, cost-effectiveness, and contribution to the knowledge base about change implementation in healthcare.
This was the set of challenges given by the National Institute of Clinical Studies (NICS) to a multidisciplinary group of more than 70 strategic thinkers who attended a 2-day workshop in Hobart in November 2003. Workshop participants were not there to represent their organisations, but to lend to the task their individual expertise, creativity, and pragmatic knowledge of Australian healthcare systems.
Summary of draft strategies developed by the two working groups to encourage the adoption of best evidence into practice in Australia
General practice/community focus
1. Finding and plugging evidence gaps for common problems in general practice. Clinical data on the most common clinical reasons for encounter would be collected from general practitioners who had been recruited through Divisions of General Practice. This information would identify areas in which best evidence could be applied and would form the basis of an intervention to be tested through a randomised controlled trial, with Divisions as the unit of analysis.
2. An “evidence SWAT team”. The team would raise “evidence literacy” and counter the impact of unreliable information by strategic entry into health debates through the media, general practices and pharmacies. The team would act as a media “watchdog” and produce media releases, a website, briefing notes for GPs and pharmacists, and patient information.
3. A national network of dedicated evidence-based general practice sites for information transfer evaluation. Ten exemplary general practice sites would implement evidence-based practice. They would incubate and test new evidence-based tools and training. These sites would also have a teaching and training role and provide education for patients.
Hospital focus
1. Creating and sustaining the evidence-based hospital. An evidence-based practice support unit would promote organisational change. The unit would regularly review the evidence for best practice, support guideline development and implementation, conduct audits and evaluation of practice, and develop close relationships with clinical units.
2. A stepwise approach to changing behaviour. Volunteer hospital units would be recruited to participate in a pilot study before a national implementation plan was rolled out. Units involved would identify what best evidence would be the focus of implementation and use interventions in stages of increasing cost and complexity to change practice.
3. Clinical research implementation networks. These teams would implement best clinical practice and evaluate its use. They would set clear objectives, form a central steering group, identify relevant evidence and priorities for implementation, and identify relevant process and outcome measures.
SWAT = special weapons and tactics.
In preparing for the workshop, the NICS convened two working groups — one to focus on general practice/community care and another on hospital care — to develop potential strategies to encourage adoption of best evidence, which could then be considered and further developed by the workshop participants.
The two working groups were asked to develop generic change strategies to encourage implementation of evidence. It was specified that the strategies should:
be appropriate to the Australian environment;
produce positive outcomes for the greatest possible number of Australians;
include ways of identifying and incorporating new evidence over time;
be feasible and implementable by the NICS and partners from 2004; and
be affordable.
The detailed draft strategies developed by the two working groups are briefly summarised in the Box.
Workshop participants were given an overview of current knowledge by two international authorities in the area of evidence uptake: Professor Jeremy Grimshaw, Director of Clinical Epidemiology at the University of Ottawa, and Professor Martin Eccles, Professor of Clinical Effectiveness at the University of Newcastle upon Tyne.
Professor Grimshaw said there was increasing interest in knowledge translation activities to promote evidence-based practice. However, there were many barriers to this occurring. The evaluation of guideline dissemination and implementation strategies had been flawed, with the result that there was an imperfect evidence base to guide decision-makers. They would have to exercise considerable judgement about how best to use limited resources for implementation.
Professor Eccles said guideline developers traditionally have not taken the crucial extra step of spelling out the implications of the guidelines for practice. He described an intervention that successfully reduced ordering of lumbar spinal x-rays in the routine investigation of back pain in primary care in England and Scotland. It involved identification of local influential people, dissemination of guidelines, education of key stakeholders (including radiologists), and audit and feedback. A key component was instructing radiologists to attach a reminder note about the appropriate ordering of lumbar spinal x-rays when reporting back to GPs. However, it could not be assumed that such an intervention would be effective in other settings — for example, in Australia, the fee-for-service environment might be a barrier.
Workshop participants then considered the strengths and weaknesses of the six draft strategies (Box) and further developed the proposals. Some participants also wished to develop other, new approaches.
Discussions highlighted the complexities surrounding the design and evaluation of systems to promote evidence uptake. They also reflected the challenges posed by this novel strategy-development process in engaging the expertise and perspectives of a variety of disciplines and stakeholders.
In the end, it was not possible within the format of the workshop to address all of the challenges posed to participants at the outset. It is noteworthy that participants strongly valued the process of the workshop as well as its outcomes, judging by the post-workshop evaluation survey completed by participants.
Group work and discussions led to some of the original proposals being built on, modified, rejected and/or amalgamated.
The overall approaches that participants worked on included:
establishment of clinical networks, using a variety of approaches and in a variety of settings;
establishment of a media “watchdog”; and
development and implementation of point-of-care interventions to promote evidence uptake.
The common thread to these approaches was the notion that strategies may be more effective at encouraging evidence uptake if they target communities rather than individuals.
The following is a brief summary of the workshop outcomes.
A network has been described as
a linked group of professionals and organisations from primary, secondary and tertiary care, working in a coordinated way that is not constrained by existing organisational or professional boundaries to ensure equitable provision of high quality, clinically effective care. The emphasis . . . shifts from buildings and organisations towards services and patients.
Another view was that a network is “a group of people with a common interest”, existing because of the need to accomplish a real practical purpose.
Networks could be pre-existing or new, national or local, and could extend beyond the hospital setting into primary care. Existing networks would need to identify evidence uptake as an additional core goal.
The workshop group thought networks would work in most situations and settings, except where there was no evidence available or no organisational willingness to be involved, or where there were other constraints such as geographical limitations. Generally, networks would be patient-, discipline- or problem-based, rather than organisationally based.
Examples of services and clinical areas in which evidence uptake could be a key feature of a network include pregnancy and newborn services; a diabetes network; a network forming around acute stroke management; maternity services; and intensive-care units. In other words, they must have strong service structures with multidisciplinary care and multiple interfaces. Desirable features of networks would include
commitment to implementing evidence-based practice and sharing lessons with the NICS and other bodies;
capacity to measure practice and change in practice;
strong clinical leadership and sustainable infrastructure, such as colleges/hospitals — ie, a strategy for continuation once NICS support is no longer available;
intention to include consumers and other professionals working in the same field;
capacity to engage a majority of healthcare providers;
capacity to be applied across settings;
coverage of an area that is important and has national relevance (eg, chronic disease such as heart failure);
willingness to develop a strategy for national adoption — one that goes beyond the pilot project and beyond the NICS’ specific support;
capacity to access evidence of effective care — there must be evidence available of a gap, of a useful intervention, and of a significant, measurable, modifiable problem;
a supportive body or partnership;
effective governance and project management;
expertise in the behavioural sciences.
Identifying and supporting potential networks. The NICS could publish its criteria for an evidence-based clinical network and call for expressions of interest from networks to undertake a clinical-change program in areas identified as having gaps between best available evidence and current practice. The networks would source reliable evidence, have explicit criteria for selecting priority areas for action, develop an understanding of the situation and the relevant behaviours (including barriers and incentives to change), develop strategies for change, negotiate issues related to the interface between hospitals and other care environments, have a risk-management process, and measure clinical behaviours. The Institute could provide funding and support and could have a strong role in ensuring there was independent evaluation of the process. The Institute could also develop and provide tools for defining problems, identifying and reducing barriers, measuring, changing management, redesign, organisational development, quality improvement and team building.
Plugging the gaps in primary care. This approach would focus not on individual GPs, but on Divisions of General Practice and other primary-care providers. The community would be involved in identifying areas of action and developing intervention strategies. The NICS and other organisations could help the Divisions drive this forward.
Developing a broad-based approach. The NICS could take a “macro” approach, linking in with existing general practice and primary-care groups and increasing the uptake of evidence-based practice in general practice through capacity building of the whole practice team. There would be focus on a protocol-driven approach in chronic-disease management. The use of pooled practice data and feedback after analysis, together with chronic-disease registers to capture data, would help drive evidence use.
Developing a “home base” of expertise. The NICS could convene a taskforce to develop a discussion document outlining a “home base” model for promoting evidence uptake through Divisions of General Practice. The home base would provide expertise in change strategies, biostatistics, systematic reviews, communication and qualitative research. Clinical priorities would be established, evidence sourced, and change strategies developed.
Using an integrated systems approach. The NICS would encourage an integrated general-practice systems approach focusing on both the practice team and patients, using a “plan–do–study–act” cyclical model. This approach might target areas such as cardiovascular disease, diabetes and “SNAP” (smoking, nutrition, alcohol, physical activity). Strategies for the practice team could include incentives, practice development workshops, skills-building and networks. The patient-based intervention would “skill up” patients so that they could stimulate GPs to better adopt evidence-based practice. The role of the NICS could be to act as a catalyst by providing GPs with evidence and guidelines and getting information to patients.
A media “watchdog” initiative would aim to increase individuals’ awareness and ability to be critical about the validity of health-related claims by raising “evidence literacy” and countering the impact of unreliable information, through strategic entry into health debates via the media. The project would be providing information about evidence — not definitive health advice. It would aim to encourage people to be more questioning of health information, to make greater use of reliable information sources, and to reduce the use of non-evidence-based interventions by the public and clinicians. A further aim would be to improve the quality of media reporting.
The brief of the media watchdog would be to respond to information in news media and other forms of media (including advertising, promotions and Internet campaigns), as well as breaking scientific news and information. It would also be proactive, through “horizon scanning”, being prepared, reviewing the past for recurring issues, and reinforcing positive messages.
This initiative would require considerable strategic planning and risk management. As an initial step, the NICS could establish a pilot project.
Evidence provided at point of care can influence practice, and there is a need to tailor evidence for local practices. Different types of clinical environments require different types of point-of-care information. To inform decision-making, evidence must be limited to the essential.
The NICS could support development of a kit that helps identify need, practice/evidence gaps, evidence, barriers and opportunities, appropriate strategies, information needs of decision-makers at point of care, strategies to ensure that information is used (such as checklists), feedback mechanisms for users, and evaluation.
This approach would work well with existing and future clinical networks for specific issues and problems. Risks include overlap/duplication with other information systems, a perceived threat to professional autonomy, and obsolescence.
As an initial step, the NICS could undertake a systematic review of point-of-care interventions and a stocktake of current developments in this area and identify a few priority areas in which lack of immediate information is the cause of the practice/evidence gap.
The author was paid by the National Institute of Clinical Studies to produce this report of the workshop.