Keynote Address: Midwest Nursing Research Society - March 2001 - "Research, Practice and Policy: Issues in Evidence Based Care"

  • Linda R. Cronenwett, PhD, RN, FAAN
    Linda R. Cronenwett, PhD, RN, FAAN

    Dr. Linda Cronenwett is Dean and Professor of the School of Nursing, University of North Carolina at Chapel Hill. Her professional experience includes fourteen years as a nurse researcher and administrator at Dartmouth-Hitchcock Medical Center, Lebanon, NH. Prior to her appointment as Dean, she was the Sarah Frances Russell Distinguished Professor of Nursing Systems at UNC-Chapel Hill.

    Dr. Cronenwett is a member of the Editorial Advisory Board of the Joint Commission Journal on Quality Improvement. For the past three years, she has been a faculty member of the Dartmouth Summer Symposia on Building Knowledge for the Leadership of Improvement of Health Care.


CRONEN.jpgGood morning. I am honored to be here with you on your 25th anniversary. I worked and studied at the University of Michigan from 1973 to 1984, so I was part of you during the early years of MNRS. In fact, I remember every detail of my first presentation as a doctoral student, which occurred during your 3rd annual meeting. I can tell by the audience that MNRS continues to launch new nurse scientists, and I welcome the doctoral students who are here today. It is obvious that you continue to nurture intellectual growth and achievement and continue to bring faculty and students together to advance science and health care and I congratulate all of you on your 25th anniversary.

MNRS has always challenged me and this speech was no exception. When Dr. Riesch called and asked me to talk to you today about evidence -based practice, she mentioned that many of you would have had an all day workshop on the topic yesterday -- with speakers who, by the way, are the leading experts in the country. The rest of you, I suspect, either teach a course or do continuing education workshops on the topic, or you have experimented with striving to move research into practice in your institutions. The only saving grace for me, I think, is that evidence-based care is something we really haven't figured out yet. There is still a lot of ambiguity about what we are trying to do and why we are trying to do it. And I admit to continuing to enjoy the opportunity to hear someone else to try to make sense of it all, so I'll hope to entertain you in the same way.

My hope is to engage you in reflections on your beliefs about evidence-based care and to challenge you with my own. I won't be talking at the level of the patient or the individual clinician or the unit but from the 30,000 foot view. My remarks will be tentative rather than conclusive, less about giving information than about trying to make some sense out of the ambiguity and confusion about how to improve health care, while, as some might say, "Rome is burning".

The Context

From the 1998 report of the President's advisory commission on consumer protection and quality in the health care industry "Quality First: Better Health Care for All Americans," we have a glimpse of the problems facing us in healthcare today:

Underuse of services. Examples include:

  • Twenty-four percent of children don't receive the appropriate set of immunizations.
  • Only 52% of elders receive annual Influenza vaccines and only 28% receive pneumococcal vaccines.
  • Only 14% of men and 27% of women who are 40 years or older receive a test for occult fecal blood.
  • Only 70% of ideal candidates for thrombolytic therapy after acute myocardial infarction receive it.
  • Only 49% of diabetics receive a dilated eye exam in a year.
  • Only 20-30% percent of depressed patients in general medicine services receive medication for their depression, and 30% of those receive sub-therapeutic doses.

Overuse of services. Examples include:

  • Office visits for URI accounting for 10% of all visits.
  • Antibiotics over-prescribed.
  • Inappropriate CABG procedures in 2-14% of patients or performed in circumstances of questionable value.

Errors (or misuse of services). Examples include:

  • 10 to 30% of laboratory test results are inappropriately classified as normal.
  • 28% of adverse reactions to medications are preventable, and 42% of the life-threatening events are preventable.
  • Approximately 20% of patients in tertiary medical centers experience adverse events.

Shortages of personnel. Examples include:

  • Instances of 100% annual turnover rates of assistive personnel in nursing homes.
  • Moderate to severe shortages of nurses and other types of health professionals.
  • Family members assuming an increased burden of care.

In this context, clinicians legitimately look at the producers of evidence reports and practice guidelines and ask: "If I can't be sure that a medication will be safely administered, if I can't get access to accurate or timely lab results, if my patients are getting pressure sores because position changes are not sufficiently frequent, if my colleagues continue to insist on restricting times when family members can be together during surgery, illness, emergency and death, what on earth do you expect me to do with the latest five evidence-based protocols?" During this period of diminished resources and increased demands, if our leadership capital needs to be expended just to try to insure patient safety, how on earth are we going to insure best practice?

These questions are sobering, they're legitimate, and they affect the issues related to evidence-based care that I have identified in each domain of my remarks today.


I am going to begin with research. After all, you are here today because of your involvement in the research missions of your schools or organizations. Evidence for practice mounts slowly over time, as scientists discover first what works in controlled environments and second what works in daily clinical practice.

Positive gains

Nursing has made some incredible gains during the last 25 years. We have produced graduates who value and understand science. We have enlarged the body of scientific knowledge about nursing practice. We have increased the number of nurse scientists and we have advanced beyond description to testing of interventions. Equally important for the goal of evidence-based care, we have championed research utilization for almost two decades now. We have witnessed incredible achievements of early research utilization pioneers. We transformed curricula to include emphases on the use of research and practice. We have led the development of clinical practice guidelines from local to federal levels.

In the health care fields as a whole, we experienced the creation of AHCPR - now AHRQ, and witnessed the steady increase in funding to that organization. We have seen increased resources devoted to compiling evidence that then becomes available to us, our students, and to clinicians. We have seen increased access to scientific articles on the internet.

We have continued to grow our numbers of scientists, the professional scientific knowledge base, and the body of clinicians who are open to leading and participating in using that science in practice. We have integrated the dissemination of research into our professional meetings and journals. We teach students how to critique scientific studies and assess levels of evidence. We are key players in the movement to address pain management as a national problem'¦and to make pain assessment the fifth vital sign. We have made a difference.

So what boundaries are we pushing now? What are the issues within the research domain that pertain to achieving world-wide goals of evidence-based care?

Clinical practice guidelines

Let's begin with clinical practice guidelines, defined as "systematically developed statements to assist practitioners and patient decisions about appropriate health care for specific clinical circumstances" (IOM, 1992, p.27) Guidelines have been developed to improve clinician knowledge about evidence related to a particular disease or procedure and to change attitudes about what might be considered best practice so that we have the potential to improve patient outcomes. We have many wonderful examples, not the least of which come from the University of Iowa, where Dr. Marita Titler and her staff have disseminated multiple research-based guidelines for care of the elderly. In spite of their success with 21 guidelines, they caution that successful implementation requires tailoring to local settings, interdisciplinary collaboration, change champions, and attention to monitoring both processes and outcomes (Titler, Mentes, Rakel, Abbott, & Baumler, 1999).

In the medical domain, the Institute for Clinical Systems Improvement (ISCI) is another example from the Midwest. ISCI has 17 medical groups who have developed healthcare guidelines and technical systems reports. They have 45 guidelines in use and have been able to document improvements, yet they caution that you need staff expertise in epidemiology and statistics in addition to clinicians to accomplish this work (Greer, Mosser, Logan, & Halaas, 2000).

So what are the issues then, if we are having these successes? First of all, practice guidelines can be so vague that they don't provide useful information for clinicians faced with an individual patient, or they can be so rigid that they are not useful when the patient doesn't fit the norm. Second, we know that amassing the science base takes time and political activity. The early practice guidelines in particular took a lot of time and political activity. When they're done, clinicians legitimately ask whether they're still relevant.

With the proliferation of guidelines coming from multiple organizations, different recommendations are sometimes made even when using the same data. So what are clinicians supposed to do? Lohr, Eleazer, & Mauskopf (1998) argued that, to date, guidelines alone have had regrettably little impact in the absence of concrete efforts to translate them into tools usable in everyday practice.

Evidence reports

Another type of aid to evidence-based practice has been the evidence reports coming out of, among other places, the AHRQ Evidence-Based Practice Center (EPC) Program. The EPC sponsors and disseminates state of the art systematic reviews on important topics that provide the evidence bases for guidelines, quality improvement projects, quality measures and insurance coverage decisions. They sponsor both methodological investigations and publications of systematic reviews.

Twelve evidence-based practice centers in the country produce reports that include the critical appraisal of the literature using explicit grading systems (Lohr & Carey, 1999). The reports tend to be developed by scientists from single disciplines; and clinicians, patients and advocacy groups are rarely involved, in contrast to the original approach to developing clinical practice guidelines. An evidence report includes analyses, evidence tables, references, and search strategies and is available to the public through the AHRQ Clearinghouse. AHRQ is concerned that these evidence reports be used and so they are documenting these outcomes.

But there are some issues with evidence reports as well. The literature databases can be incomplete, there can be publication biases, and the marked emphasis on drawing evidence from randomized controlled trials (RCTs) can skew the information so that it is relevant only to the patient populations included in the studies.

We need more effectiveness data instead of the heavy reliance on efficacy data. Dr. Titler said, "the outcome achieved in a controlled environment, when a researcher is implementing a study protocol with a homogenous group of patients, may not be repeated when the practice is used by multiple caregivers in a natural clinical setting" (Titler et al., 1999). So the questions remain: Can evidence reports be produced frequently enough so that current evidence is always included? Can clinicians be expected to evaluate evidence on their own? And, if they are not involved, will the conclusions drawn be accurate for practice?

Clinical/critical pathways

We have one other aid that I want to mention, one that is familiar to all of you in the audience who are in practice-- the tools where we try to translate evidence reports or practice guidelines into something that guides best practice at the micro-system level. The clinical pathway:

  • defines best practice when practice styles are considered to vary unnecessarily
  • coordinates or decreases the time spent in different steps of the care process for the patient
  • gives all staff a common game plan from which to view their roles in the overall care, and
  • improves patient and family education about the plan of care.

But clinical pathways raise issues associated with autonomy and standardization. Clinicians are reluctant to give up autonomy of decision-making without proof that outcomes improve. In many instances, we don't know for sure that one can actually identify a best practice for all patients with a particular condition. We don't know if one actually improves upon usual practice by setting a standard of evidence-based care. And how many conditions are actually amenable to a critical path? We don't always know.

The amount of evidence we produce

In the midst of the learning that is taking place regarding the scientific community's contributions to evidence-based care, we have some issues that surface from time to time in nursing circles - that inhibit or limit our contributions to the production of the science needed for practice. First, the reality is that we may never have, at least in my lifetime, a critical mass of nurse scientists in all the domains of healthcare from which evidence is needed to support nursing practice. I'm proud of how well we do at the UNC-Chapel Hill School of Nursing in competing for research funds, yet the NIH dollars coming into our School of Nursing in fiscal year 2000 were roughly 4% of the NIH dollars that came into our School of Medicine.

I would argue that our science base must be interdisciplinary if we are going to produce more evidence. To whatever extent we are exposing students only to the science done by nurses, we are failing them. To whatever extent we're holding back from partnerships with scientists across disciplines, we are limiting the amount of evidence we could produce. To whatever extent we are holding back from attempts to interest scientists in Public Health and Medicine in the significant problems of our practice, we are dramatically reducing the likelihood of a future evidence-base that's sufficient for practice.

After just two years in my position, I am beginning to struggle with the question of how to increase school of nursing research outputs. We've reached a level of success where we have some self-limiting boundaries that may prevent us from growing our research missions. I think the following scenario is common in schools of nursing today:

  • Almost all the grants on which faculty are funded are submitted through our Schools (we love the overhead funds that accompany these submissions).
  • Our productive faculty earn 20-30% of their salaries through external funding.
  • The modal pattern is "one grant at a time."

We consider ourselves fortunate to have 2-3 people with the same area of expertise, because in order to meet our teaching mission, we require diversified areas of expertise. Many of our Research I schools of nursing don't need more faculty to teach in their doctoral programs, yet when you are trying to recruit the next scientist, those are the courses they want to teach.

So here we sit at the end of the period of success in our development of research expertise with a natural cap on our potential for science. When you add to this picture the fact that we are initiating doctoral programs that are not aimed at research training and the fact that we have limited student pools for those that are research intensive, we face a future with a continuing relatively insignificant contribution to evidence-based care.

I think it is important to face these facts and to think about what to do. If you agree at all with my assessment, I think one thing we can do, because I still see it happening, is to cease our displays of arrogance about nursing roles and the production of evidence in the larger world. We are doing great things, but we are a small piece of the pie. So let's not complain about the fact the world doesn't tell us how important we are every day. The world of science wants and needs us all, every available body that can do the work. Media want to tell important stories no matter who the scientists are. All of our energy should be channeled into just doing it. If we dedicate ourselves to the study of important problems, the rest will follow. We don't need to whine.

Another idea is to experiment with other models of organizing ourselves for science. We need to brainstorm future scenarios in each of our institutions that will broaden the amount and quality of knowledge we produce. For instance, we may need to reach a state where a larger percentage of our researchers are funded for 60% of their time or more on grants or contracts. One way to increase that potential is if our research faculty members are sought out by investigators from other disciplines to provide some needed portion of their science. With few nurses with the same area of expertise in a school of nursing, we can't expect nearly as much cross-grant participation as we can if our expertise is valued beyond our doors. We need to be building those bridges for our scientists, old and young, and eventually it will become a way of life.

Another option may be to create and stimulate interest in research positions that exist mainly on soft money, as have Schools of Public Health and departments in Schools of Medicine for decades. If we want to produce BSN to PhD graduates who have never practiced and never taught and don't particularly wish to start, we will need these positions. Under North Carolina regulations, a faculty member cannot teach BSN students if they have not had two years of full-time equivalent clinical practice. But we are interviewing emerging scientists who don't meet this requirement, and we may have other faculty members who would prefer to devote their careers solely to the research mission. So I think we need to craft the positions and figure out how to include research scientists as full and welcomed members of our faculties.

Schools that can achieve these and other models will be able to hire more scientists without the attendant morale problems that result from inequity in workload-- and I mean when there aren't enough courses that scientists want to or can teach -- and without the inherent limitations on concentration of expertise. To preserve faculty satisfaction in the face of limited doctoral teaching needs, we need to reduce the collective time for teaching that tenured faculty are available. I think this boundary around our research missions is something that we need to be pushing if we want to become more significant players in moving evidence into practice.

The roles of clinicians in the evidence we produce

The second challenge I see before us is to increase the involvement of our clinical community partners during each phase of the research process. And yes, that means during the identification of the problem, development of the intervention, planning the methods, collecting the data, analyzing and publicizing the results. The research we do is infinitely more likely to be translated into practice if these partnerships are in place.

At Carolina right now, we're involved in a number of partnerships with minority communities, as we struggle to develop science that might assist in the reduction of health disparities, either in the outcomes of healthcare or in the education of minority healthcare professionals. The minority communities are being incredibly clear. "Include us up front. We want to help define the problem. We want you to work on problems that are important to us. We want to develop the hypotheses to be tested. We will help you with access and data collection, but we want to be there when you look at the results and decide what to say about them. We want something to be returned to the community. Long term relationships, social relationships, better health care. We want a real partnership, not something that only goes one way."

I suspect our profession would have benefited greatly if nurses in practice had been equally clear and had an equally powerful voice over the years with nurses who are conducting research. Where we have succeeded in clinician- researcher partnerships, and many of you have, we see scientists working on problems that are important for nursing practice. We see the production of evidence that benefits practice. We see people solving implementation problems as the research is being conducted instead of having to adapt the intervention afterwards for practice with the attendant need to re-evaluate whether the findings continue to be valid. We see clinical partners writing companion articles on how to implement evidence-based interventions. Where clinical partners do not exist, the dissemination and implementation of research are impeded ... often forever.

AHRQ is convinced of this reality and is changing the nature of their request for proposals and applications to insure a new role for researchers. Carolyn Clancy, the Director of the Center for Outcomes and Effectiveness Research at AHRQ, gave the keynote speech for our 2000 Program on Health Outcomes Lecture Series and shared information about AHRQ's TRIP (translating research into practice) agenda. She acknowledged that the production and synthesis of evidence greatly exceeds its use and AHRQ intends for that to change.

AHRQ's agenda is for researchers to ask users what they want to know, partner with users in research design, include in the research all the people who are affected. They're asking researchers to tell user partners what they found, know why the findings matter, and say so. They want scientists to tell funding agencies if anyone is using the research they're producing -- or they might not get funded the next time. Her summary of the roles they want to see researchers play reiterates the partnership agenda.

The ways we teach about evidence-based care

Another challenge exists primarily in the teaching domain. Because most of us are faculty members, we determine how we teach about the translation of research to practice. I commented earlier that we have given this topic greater centrality than it had two decades ago; yet in many ways we continue to focus solely on research utilization without exposure to the broader issues of using evidence for practice. Assignments continue to include a critique of a single study, but isn't a clinician going to be more concerned with the grading of evidence across multiple studies and trying to figure out whether that evidence is strong and what to do with it?

We have stopped the requirement that students compute their own statistics, and, in many schools, students learn about the correct application and interpretation of statistical tests. I suspect few medical students ever write a paper to critique a single study --- and potentially that is a problem. But I can imagine that we, too, may shift away from focusing on a single study to how to find already compiled evidence reports and make sense of them in the context of the care of individual patients and communities. Wherever the right path lies, we have creative work to do if students are to learn the skills and knowledge necessary to be prepared for evidence-based care.

The type of evidence we produce

One final issue and I will move on to practice, although in many ways this issue is a segue to the issues of practice, and that issue is the type of knowledge we produce. I don't know how many of you support the argument that we have two types of science emerging now:

  • classic science that produces professional knowledge, tends to be disciplinary, tends to use the traditional scientific method, and
  • improvement science that is focused on systems, tends to be multidisciplinary and tends to focus on variations in outcomes and whether we are making a difference in the outcomes of care.

If you do, I think we could agree that professional knowledge has really driven most of our past improvements in healthcare. It's what answers the question:

  • Does this process or intervention improve outcomes for patients or communities regardless of the difference in providers and systems?

Systems knowledge, where it exists and very little does, answers the questions:

  • Do these providers perform differently from one another?
  • Do they perform well against an evidence-based standard of care?
  • What are the areas in which these providers should improve?
  • Are they improving the care they deliver?

Healthcare workers are unaccustomed to using prediction and measurement for learning outside of their own individual clinical care. Learning is viewed as something that happens to a person, not something that happens to an organization. And we have this prevailing sense that healthcare work is too urgent and workers too busy for them to plan and evaluate the outcomes of change.

If systems knowledge is not produced, if the methods of producing it are not taught to graduate students, and if we don't produce health professionals who are prepared for the continuous improvement of healthcare, then our knowledge about the effectiveness of using evidence in practice will continue to be inadequate. If these conditions persist, we will not meet the real challenge of evidence-based care.


Let's move on to practice then, which is the place where patients can be helped, where the quality of care can be improved by the use of evidence. What are the gains and challenges that face us in the practice world? Once again, keep in mind the context and pressures of the current environment, the current staffing patterns, and the fragmentation of our systems of care.

John Eisenberg (1999), the head of AHRQ, pointed out the differences between the clinical practice model of yesterday:

  • where knowledge is handed down to students by experts, and
  • quality of care is envisioned as dependent on the combined abilities of individual clinicians.
  • to the clinical practice model of today:
  • which requires the integration of multiple sources of information, including patient preferences, and
  • quality of care is determined both by the abilities of individual healthcare professionals and the systems in which they work.

The moment we move into the world of practice and talk about implementing research, guidelines, evidence reports, critical pathways or any portion thereof, we increase dramatically the complexity of the issues.

Positive gains

We have made steps forward in practice. Many healthcare systems are teaching nurses the concepts, skills and techniques of improvement science. Nursing pioneers are leading hospital-based departments that support clinical improvements. In many studies that have been done about moving research into practice, nurses have been found to be the key change champions in successful improvements in care. In fact, a colleague in the School of Public Health who study organizations recently asked me: what are the nurses doing? I wish could answer her.

Professional autonomy versus standardization to best practice

Because of our systems of health professional education and regulation, we often face the challenge of implementing "best practices" almost literally one health care professional at a time. We teach nurses and physicians to make their own assessments. We emphasize the importance of each individual practitioner's critical thinking. We send them out to practice in a world where it would be impossible for every individual to stay on top of the research literature about each patient problem they will face. Not surprisingly, the results we get are wide variations in practice unrelated to the patient's presenting condition - all in the name of professional autonomy.

We have evidence that some health care micro-systems are able to improve care by identifying the evidence base, translating it into a "best practice" guideline, and monitoring the implementation process and health care outcomes that result. Titler and colleague's (1999) excellent article "From Book to Bedside" talks about exactly this process within nursing; however, they also describe the difficulty of sustaining achievements -- about talking to people, physician by physician, nurse by nurse'”about the need to return to units to reinforce evidence-based practice if change champions leave their roles.

In a January article in the Joint Commission Journal on Quality Improvement, Lichtman and colleagues (2001) at Yale and Cedars-Sinai, Los Angeles, presented findings from a multi-center interventional study designed to answer the question: Can practice guidelines be transported effectively to different settings? Having developed a standardized protocol for low-risk chest pain with proven good results in one setting, they chose five hospitals in different states (teaching and community, urban and rural) where they could attempt to replicate their experience. Only two of the five sites achieved significant increases in protocol adherence from the time of baseline data collection to a period six months following implementation of the intervention. (As one frequently notes in these papers, one of the two successful sites had a full-time nurse champion, and guideline adherence was greatest at the three sites that reported nurse-to-physician contact as the most common method of notification for patients who qualified for the guideline.) The authors' said:

Despite a standardized protocol, in-person training sessions, periodic audits, careful monitoring by the authors of the guideline, and best efforts by the local champions, the implementation process and subsequent change in patterns of care differed across sites, indicating that the effective translation of clinical guidelines outside their original setting is a complex process. (Lichtman et al., 2001, p. 47).

Health care experts from the USA and Europe held a roundtable in October, 1999, to discuss international experiences with implementing evidence-based guidelines (Gross, 2000). In reviewing the literature on guideline implementation, they came to the following conclusions:

  • The use of multiple implementation methods is more likely to succeed than the application of a single method.
  • The effectiveness of specific strategies, such as academic detailing and use of opinion leaders, varies.
  • Didactic educational efforts usually fail unless they are interactive.
  • Computer-assisted management programs are effective.
  • Non-physician providers (I'll call them nurses!) can facilitate implementation.
  • Assessment and feedback are important.
  • Local selection and adaptation of guidelines are critical.

Why are the challenges so great? If we are preparing the best individual clinicians in the world, why are there so many barriers to use of evidence? In a study by Cabana and colleagues (1999), they reviewed 5658 articles about clinical practice guidelines and selected 76 published studies that described at least one barrier to adherence to guidelines, practice parameters, or national consensus statements. The barriers identified could be organized into a theoretical framework of factors related to knowledge, attitudes, and behavior.


  • Lack of familiarity - volume of information, time needed to stay informed, guideline accessibility
  • Lack of awareness - volume of information, time needed to stay informed, guideline accessibility


  • Lack of agreement - variations in interpretation of evidence, applicability to patient, not cost-beneficial, lack of confidence in guideline developer.
  • Lack of agreement with guidelines in general - opposition to cookbook medicine, guideline too rigid to apply, challenge to professional autonomy, not practical.
  • Lack of outcome expectancy - clinician believes that performance of recommendation will not lead to desired outcome.
  • Lack of self-efficacy - clinician believes that he/she cannot perform guideline recommendation.
  • Lack of motivation - inertia of previous practice, habit, routine.


  • Patient factors - inability to reconcile patient preferences with guidelines.
  • Guideline factors - presence of contradictory guidelines.
  • Environmental factors - lack of time, lack of resources, organizational constraints, lack of reimbursement, perceived increase in malpractice liability.

The authors concluded that multiple strategies to induce change will be necessary, but, depending on the nature of the variables in any clinician group or organization, different strategies may apply.

These barriers are highly similar to the barriers to research use by nurses identified by Sandy Funk and colleagues (1995). The barriers are not unique to physicians - or nurses, but to all health professionals.

In another approach to identifying what works with guideline implementation, investigators used a modified nominal group/Delphi process with twelve people who had extensive experience in leading clinical guideline implementation and were thought to have particular insight into the process (Solberg et al., 2000). They generated 87 variables they thought affected the success of guideline implementation and 25 strategies that were considered to be essential or key elements of successful implementation. These 87 variables clustered in categories that were then ranked in order of importance by the panel.

In order of importance, the variables believed to affect implementation were (Solberg et al, 2000):

  • Organizational capacity for change - strong support and interest at all levels of leadership, along with a well-developed infrastructure, capability, and culture for CQI and change management.
  • Implementation infrastructure - standardized organized process for making change, access to adequate resources, retraining capability.
  • Medical group characteristics - presence of organized systems, priorities for quality vs. finance, financial resources, clinician cohesiveness to shared mission/policies.
  • Guideline characteristics - relative advantage of the new care process, importance of the guideline topic to the clinicians, potential for change that would close the gap, guideline developer credibility.

None of these key elements have anything to do with individual clinician-level issues. The elements considered important by persons who are experts in the implementation of evidence-based change are, by and large, organizational.

At the Institute for Healthcare Improvement meeting in December, 2000, Steve Shortell, another expert in this field, reported the results of a consensus conference. At this conference it was concluded that achieving sustainable quality only occurred under conditions of both a high organizational commitment to quality improvement and change AND conditions where the specific change was of strategic importance to the organization. Clearly, from multiple sources, we must conclude that huge organizational commitments are required if we are to overcome the prerogatives of professional autonomy.

Educating Health Professionals for Evidence-Based Care

Once again, I ask, are we preparing students for these realities? Do we expose students to interdisciplinary teams that are making attempts to change practice based on evidence? Are nurses who have been change champions for successful improvements in care invited to lecture? Do we value nurses who are performing these roles? Are they our heroes'¦or do we dismiss them as handmaidens?

Moving back into academe from a practice setting where I had been working with improvement science methods and many nurse change champions, I was surprised that few faculty members recognized the tools that were available to evaluate whether changes were improvements. Statistical process control charts were a mystery. The underlying relationships of control charts to the statistics we teach in graduate school were unknown. Yet these are the tools nurses need when they enter today's health care environments if they are to be leaders or even good members of improvement project teams.

I've had the pleasure of working as part of a national interdisciplinary faculty group led by Dr. Paul Batalden at Dartmouth. Two nurse, three physician, and two health care administration faculty members work together as "Batalden's Kitchen Cabinet". Each summer, we plan a symposium for 40-50 people who are members of our three professions (national and international participants). We gather for a week and push ourselves to continuously learn and innovate about teaching for the improvement of care. In the summer of 1997, we spent the week trying to perfect a statement of core content that we believed all health professions students should learn. The eight domains of content we believed were important were (Batalden et al, 1998):

  • Health care as process, system. Experiences that help students understand the interdependent people (patients, families, caregivers), procedures, activities, and technologies of health care-giving that come together to meet needs of individuals and communities.
  • Variation and measurement. The use of measurement to understand variation across and within systems so that clinicians have a basis for evaluating changes and using data to make changes that decrease unwanted variation or improve an otherwise standardized process.
  • Customer/beneficiary knowledge. Identification of the person, persons, or groups of persons for whom health care is provided or may be provided in the future, an understanding of their needs and preferences and the relationship of health care to those needs and preferences.
  • Leading, following and making changes in health care. The methods and skills for designing and testing change in complex organizational caregiving arrangements, including the general and strategic management of people and the health care work they do in organizations ... and including the ability to contribute effectively to the development of a supportive internal organizational climate for working, learning, and caring.
  • Collaboration. The knowledge, methods and skills needed to work effectively in groups, to understand and value the perspectives and responsibilities of others, and the capacity to foster the same in others. Experiences that help students develop an appreciation for the importance of interdisciplinary participation in the establishment of organization-wide goals.
  • Social context and accountability. An understanding of the social contexts (local, regional, national, global) of health care and the way that expectations arising from them are made explicit ... an understanding of the financial impact and costs of care.
  • Developing new locally useful knowledge. The recognition of the need for continuous learning, and the need for new knowledge in daily health professional practice ... and the skills to develop that new knowledge through empiric testing.
  • Professional subject matter. And yes --- it's also essential to learn the knowledge of our disciplines, along with the ability to apply and connect it to all of the above.

Some of this content is content that nursing can take pride in doing well. Some we hardly address at all. Some additional thoughts about which symposium participants reached consensus about teaching/learning about the improvement of health care were:

  • Learning about improvement is stimulated when it is linked to real clinical/organizational work.
  • Faculty development is probably necessary and should take place within the context of practice.
  • Some learning should occur in the context of inter-professional teams.
  • Early content with core ideas is important, especially with regards to inter-professional collaboration.
  • Improvement learning should be a thread throughout health professions education.

The irony was that three years later, in the 2000 Dartmouth Summer Symposium, we approached the topic in a totally different way. Faculty from each discipline shared the stories of heroes in their fields from past and present, and the participants analyzed those stories for what could be learned about what made these people heroes. We found that most characteristics were common across the disciplines, though not all. The following ideas emerged as priorities for what would be needed in health professionals of the future:

  1. Able to identify the human elements of need and put the beneficiary at the center of the enterprise. Balances individual and population needs with a new synthesis of the two.
  2. Able to function as a member of an effective integrated team, including being part of collaborative support for patient self-management.
  3. Able to create and be part of a learning community. Demonstrates continuous learning inside and outside of health care.
  4. Values a commitment to service.
  5. Able to listen well and use those skills in interpersonal relationships.
  6. Able to use improvement methods and principles. Understands systems thinking.
  7. Manages value conflicts.
  8. Gives evidence of a moral, ethical compass.

Each discipline then went to work on the question: What would it take in your profession to educate this health professional of the future? Here are the thoughts that emerged from the group of nursing faculty participants (Cronenwett, 2001):

  1. Increase the availability of role models who exemplify the values and behaviors desired in the health professional of the future.
  2. Increase opportunities for students to see and learn about inter-professional collaboration.
  3. Embed commitment to improvement (personal, team, system) in academic and experiential learning. It can't be taught if it's not practiced.
  4. Assure that people and human needs are the focus of our teaching. Although technologies will continue to emerge, we need professionals who will stay focused on human beings.
  5. Focus on the big issues - the truly important human needs of our time.
  6. Professional silos cannot support real learning, for faculty or for students.
  7. Retain strategies that have nurtured nursing heroes of the past:
    • A commitment to service.
    • A focus on human beings and their responses to birth, death, illness, and disability.
    • Patient and community advocacy.
    • Professional values that support human dignity.

So here we are, having begun with a focus on evidence-based practice, now ending with a discussion about what health professionals really need to know if we're ever to achieve the goal of improvement in health care. Admittedly, we ended up pretty far afield from the content that normally comprises our research and even our research utilization courses. Somehow we have to widen our lens so that we don't focus solely on professional knowledge. Likewise, we can't focus solely on the learning of individual clinicians if it's at the expense of their learning about the work of improving health care, or we will continue to stumble in our path towards evidence-based care.


We've touched on the issues of policy throughout my remarks today. As far as I'm concerned, the issues regarding use of evidence in practice apply equally to the issues surrounding use of evidence in policy. We have made some steps forward in the policy arena, primarily through the establishment of AHCPR, now AHRQ, because we finally achieved funding for projects that were about the application of knowledge for the good of patients and society. We had nursing pioneers like Ada Jacox and Nancy Bergstrom who chaired early guideline development and thus affected millions of patients and families through the policies for practice that resulted from their work. The policy agenda for AHRQ is critically important to health care, and we're lucky to have leaders like Marita Titler and Mary Wakefield on AHRQ's Advisory Council, affecting that agency's policy agendas today (Maddox, Wakefield, & Bull, 2001).

But we face one major challenge and that is the scarcity of nursing leaders in policy-making roles. During a time when people are looking for nurses - begging us to name someone qualified to participate at the policy table - we have too few. The situation is not unlike the scenario of "too few scientists."

I think the major issue, if we were to name it, is the nature of interdisciplinary leadership, that is, nursing's position with respect to the other dominant players in health care--medicine and health care administration. Neither profession embraces nursing as a full partner. After several experiences where a team's accomplishments are attributed to physician or administrative leaders, nurses sometimes retreat to work on problems of lesser scope that do not require interdisciplinary work.

If nurses retreat, we give away our place at tables where nurses' voices in generating solutions to the big problems in health care are required. We cannot allow members of our profession to retreat. Given the interdependence of nurses, physicians, and health care administrators when the patient is the focus of concern, nursing leaders need to continue to expose students to functioning, effective interdisciplinary teams who are working on the big problems of human need. In addition, the professional lives of nursing faculty members need to demonstrate the priority placed on this work.

The Call to Be Heroes

I think the times call for everyone in this audience to reflect on what he or she can do to nurture the larger competencies required for rethinking and revamping our health care system. The time calls for health care professionals who are heroes. Nurses too often say, "What can I do? They are to blame. They have the power, not me."

In this day and age, no health professional is feeling all that powerful, believe me.

Or nurses might say, "But I don't know where to go or what to do to improve the health care work I do (or my organization does)." You know what? No one else knows what to do either. We're all learning and making it up as we go along.

Karl Weick (1995), author of Sensemaking in Organizations, tells a story to illustrate a point.

An incident related by a Hungarian Nobel Laureate happened during military maneuvers in Switzerland. The young lieutenant of a small Hungarian detachment in the Alps sent a reconnaissance unit into the icy wilderness. It began to snow immediately, snowed for 2 days, and the unit did not return. The lieutenant suffered, fearing that he had dispatched his own people to death. But on the third day the unit came back. Where had they been? How had they made their way? Yes, they said, we considered ourselves lost and waited for the end. And then one of us found a map in his pocket. That calmed us down. We pitched camp, lasted out the snowstorm, and then with the map we discovered our bearings. And here we are. The lieutenant borrowed this remarkable map and had a good look at it. He discovered to his astonishment that it was not a map of the Alps, but a map of the Pyrenees. (p.54)

Weick says, "This incident raises the intriguing possibility that when you are lost, any old map will do."

Just think about the energy we derive from people who have outlined a new idea ... a new path ... and enlist our help to make things better. We, too, can find or create evidence that can guide us'¦can give us our bearings. And if, by chance, we make it - back to camp - or having made care better for the patients and communities we serve - then we will be on the consensus panels, up on the stage speaking to inspire others. If everyone commits to some aspect of heroism, there will be enough nurses in policy roles throughout the country.

What else do we need besides '˜any old map' --- and just think how much better an evidence report might be! Weick (1995) says that resilient organizations have people who maintain an "attitude of wisdom." Ignorance and knowledge grow together. To be wise is not to know particular facts but to know "without excessive confidence or excessive caution" - to maintain curiosity and openness and a willingness to engage in complex sensemaking.

Don Berwick, the CEO of the Institute for Health Care Improvement (IHI), in his "Escape Fire" keynote address at the 1999 National Forum (December 9, 1999, New Orleans, LA), talked about preconditions to being able to make new sense for our organizations and patients during rapidly changing health care conditions. He says "the first is the toughest - to face reality. This is very, very hard. Often the threat is not too small to see but too big to see. Some problems are too big to name." He says, "I now think that this is where we have come in health care. I have been radicalized. Our challenge is not to develop more sensitive ways to detect our risks, our errors, our flaws, our variation, our indignities, our fragmentation, our delays, our waste, our insults to the people we say we exist to serve. Our challenge is to have the courage to name clearly and boldly the problems we have - at the size they occupy - immense. We must find ways to do that without either marginalizing the truth-teller or demoralizing the good people working in these bad systems."

In his speech, Berwick drew on work by University of Michigan education Professor David Cohen who says that no education occurs until what we call the "inert" assets (books, teachers, rooms, curricula, rules, budgets, and so on) come into interaction with each other and with students. People in educational organizations, Cohen says, often act as if the inert assets were essential and the interactions not. They fight political wars over the inert assets and spend little time defending and perfecting the interactions among them through cooperation, communication, teamwork, and building knowledge about students.

Berwick says, it's the same in health care. Care is not nurses, doctors, hospitals, computers, books, rules, or medicines. These are inert. Care is interaction among our assets and between these and patients. To perfect care, we must perfect interactions.

What is my commitment? I'm trying to nurture cooperation, communication and teamwork within a system (that is, a university) that rewards professional autonomy above all else. I worked to secure a seat on the strategic planning board for UNC Health Care System. I work collaboratively and visibly with physician colleague Deans in the School of Public Health and the School of Medicine, benefiting from their wisdom and power and grateful that they share power with me.

I am working with all of our health affairs deans to create an interdisciplinary certificate program on health outcomes with a lecture series in which members of all disciplines participate. I have worked with deans from the other disciplines on grant proposals related to patient safety. I stay involved with the Institute for Healthcare Improvement and Paul Batalden's health professions educator group as my source for stimulation and continued interdisciplinary learning.

None of my choices are unique. We have incredible heroes in this room. I ask you to make your personal commitment to taking some next step on the path to evidence-based care. It can start right here in this meeting. As you listen to research presentations, comment on what you have gained from interdisciplinary work. Let students see those role models. If you are studying a topic because the users of the evidence begged you to find answers for them, say so. It's as important as describing the theory that guides your work. If you have a clinician partner who helped design an intervention that is now proven to work, let the audience know. If you have led an evidence-based change in practice, tell the story. Let us applaud you, regardless of the outcome.

When you are listening to a presentation and you think someone is using a map that is going to improve the care of patients in a major way, give them encouragement. If you see a link between someone's work and the goals of a professional organization, give them a tip.

This path is worthy. In fact, it's essential if we're ever going to get out of the mess we're in. We're going to need new tools and we're going to have to create them as we go along. The world needs all that nursing can deliver. Don't hesitate. As Arthur Ashe said: start where you are, use what you have, and do what you can.

Thank you.


Linda Cronenwett, Ph.D., R.N., F.A.A.N
Email Address:

Dr. Linda Cronenwett is Dean and Professor of the School of Nursing, University of North Carolina at Chapel Hill. Her professional experience includes fourteen years as a nurse researcher and administrator at Dartmouth-Hitchcock Medical Center, Lebanon, NH. Prior to her appointment as Dean, she was the Sarah Frances Russell Distinguished Professor of Nursing Systems at UNC-Chapel Hill.

Dr. Cronenwett is a member of the Editorial Advisory Board of the Joint Commission Journal on Quality Improvement. For the past three years, she has been a faculty member of the Dartmouth Summer Symposia on Building Knowledge for the Leadership of Improvement of Health Care.

Disclaimer: Mention of a Web site does not imply endorsement by the author, OJIN, or NursingWorld.


Batalden, P.B., other faculty and participants of the 1997 Summer Symposium. Eight domains for the improvement of health care. Presentation for the Institute for Healthcare Improvement, Boston, MA, September, 1998.

Berwick, D.M. (1999). Escape fire.  Plenary address. Eleventh Annual National Forum on Quality Improvement in Health Care.  New Orleans, LA.

Cabana, M.D., Rand, C.S., Powe, N.R., Wu, A.W., Wilson, M.H., Abboud, P.C et al. (1999). Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA, 282, 1458-1465.

Clancy, C. (2000). Health outcomes and evidence-based medicine: Translating research into practice. Keynote speech for the 2000 Program on Health Outcomes Lecture Series: University of North Carolina at Chapel Hill, September 8, 2000.

Cronenwett, L. (2001). Educating health professional heroes of the future: The challenge of nursing. Frontiers in Health Services Management, 18 (2), 15-21.

Eisenberg, J.M. (1999). Outcomes research: The road to better outcomes. Keynote speech for the 1999 Program on Health Outcomes Lecture Series: University of North Carolina at Chapel Hill, September 10, 1999.

Funk, S. G., Tornquist, E.M., & Champagne, M.T. (1995). Barriers and facilitators of research utilization: An integrative review. Nursing Clinics of North America, 30(3), 395-407.

Greer, N., Mosser, G., Logan, G., & Halaas, G.W. (2000). A practical approach to evidence grading. The Joint Commission Journal on Quality Improvement, 26, 700-712.

Gross, P.A. (2000). Implementing evidence-based recommendations for health care: A roundtable comparing European and American experiences. Joint Commission Journal on Quality Improvement, 26, 547-553.

Institute of Medicine. Guidelines for Clinical Practice: From Development to Use. In M.J.Field & K.N. Lohr (Eds). Washington, DC: National Academy Press.

Lichtman, J.H., Roumanis, S.A., Radford, M.J., Riedinger, M.S., Weingarten, S., & Krumholz, H.M. (2001) Can practice guidelines be transported effectively to different settings? Joint Commission Journal on Quality Improvement, 27, 42-51.

Lohr, K.N., & Carey, T.S. (1999). Assessing "best evidence": Issues in grading the quality of studies for systematic reviews. Joint Commission Journal on Quality Improvement, 25, 470-479.

Lohr, K.N., Eleazer, K., & Mauskopf, J. (1998). Health policy issues and applications for evidence-based medicine and clinical practice guidelines. Health Policy, 46, 1-19.

Maddox, P.J., Wakefield, M., & Bull, J. (2001). Patient safety and the need for professional and educational change. Nursing Outlook, 49(1), 8-13.

Shortell, S. (2000). Evidence-based improvement: How do we know it works? Panel presentation with R. Baker and J. Ovretveit. Sixth International Scientific Symposium on Improving Quality and Value in Health Care, San Francisco, CA, December 6, 2000.

Solberg, L.I, Brekke, M.L., Fazio, C.J., Fowles, J., Jacobsen, D.N., Kottke, T.E., Mosser, G. et al. (2000). Lessons from experienced guideline implementers: Attend to many factors and use multiple strategies. Joint Commission Journal on Quality Improvement, 26, 171-186.

Titler, M.G., Mentes, J.C., Rakel, B.A., Abbott, L., & Baumler, S. (1999). From book to bedside: Putting evidence to use in the care of the elderly. Joint Commission Journal on Quality Improvement, 25, 545-556.

Weick, K.E. (1995). Sensemaking in organizations. Thousand Oaks, CA: Sage Publications.

Citation: Cronenwett, L. (February 19, 2002). "Research, Practice and Policy: Issues in Evidence Based Care." Online Journal of Issues in Nursing. Vol. 7 No. 2.