Translating Evidence into Practice 1998 (continued, 4)

Session 4. Discipline-Specific Approaches to Evidence-based Practices

Moderator: Ernestine W. Murray, R.N., M.A.S., AHCPR

Resources to Facilitate Evidence-based Practice—Alba DiCenso, R.N., M. Sc., Ph.D., Ph.D., School of Nursing, McMaster University

Evidence-based Nursing (EBN), of which Dr. DiCenso is one of three co-editors, is a journal intended to look at strategies for overcoming the barriers and bridging the gaps for evidence-based practice. For each issue of EBN, articles relevant to nursing are summarized in the form of structured abstracts that describe the question, method, results, and evidence-based conclusions. Each abstract is accompanied by a brief commentary from a nurse clinician who is an expert on the subject matter and comments on the methods and clinical applications of the study findings. The author of the article is given the opportunity to review the abstract and the commentary before publication. Approximately two-thirds of the abstracts published in the first three issues of EBN have come from 30 journals, with the highest yield from the Journal of the American Medical Association, New England Journal of Medicine, Pediatrics, Archives of Internal Medicine, and the British Medical Journal.

Another strategy for bridging the gap is systematic reviews. The Cochrane Collaboration is an international initiative to undertake and maintain high quality systematic reviews of health care interventions. The evidence on a topic is extracted from studies, synthesized, and results are disseminated to consumers, health policymakers, and practitioners. Other strategies are clinical practice guidelines such as those produced by AHCPR and centers for evidence-based nursing that are emerging internationally such as the Canadian center at McMaster University.

Discipline Approaches to Evidence-based Practice—Norma M. Lang, Ph.D., R.N., F.A.A.N., F.R.C.N., Dean, School of Nursing, University of Pennsylvania

Dr. Lang said evidence-based practice has brought dramatic increases in clinical research focusing on nursing care, i.e., looking at patient problems and nursing interventions. Health services research focuses on the organizational context of nursing research. Classification systems and informatics are also related to nursing practice. The practice of nursing mandates looking at the patient populations with other descriptors such as diagnoses of pain, incontinence, immobility, confusion, skin breakdown, and inadequate sleep.

Research and clinical groups have developed classification systems for hundred of terms used by nurses to diagnose, intervene, and determine outcomes of care. The National Library of Medicine has included the recognized vocabularies in the meta-thesaurus of terms, an action that will help all the disciplines use each others literature. In Geneva, the International Council on Nursing is developing an international classification system for nursing practice.

The numbers of research and expert practice journals have increased radically and include: Applied Nursing Research, Journal of Nursing Care Quality, Clinical Nursing Research, Evidence-based Nursing, and Online Journal for Knowledge Synthesis for Nurses. Multiple general and special nursing organizations are developing best practices and guidelines. Twenty-five organizations have joined together to share best practices on the Internet.

Dilemmas and Realities of Evidence-based Dental Practice—Amid I. Ismail, B.D.S., M.P.H., Dr.P.H., Professor, School of Dentistry, University of Michigan

Nova Scotia is one of four provinces in Canada that has a universal publicly financed dental care program for children. Dr. Ismail, the epidemiologist on the committee to review changes in the plan resulting from cutbacks, said options were to cut services, reduce coverage, or find more efficient methods of care.

As a result of the study, a new program had guidelines that directed the budget allocation for covered services. All children (ages 0-10) are covered; all children are eligible for an annual dental exam; and based on the findings, the children may qualify for cleaning, fluoride, sealants, and treatment. The new program provided treatment to children with high risk of tooth decay; children with physical, medical, mental, and other challenges; and children whose parents could not afford the cost. The Nova Scotia Dental Insurance Program became effective in October 1997 and includes a Caries Prevention Service and a universal 15-minute appointment for plaque disclosure, flossing and brushing, and selective polishing.

Development of Practice Management Guidelines for Trauma—Michael D. Pasquale, M.D., F.A.C.S., Assistant Professor of Surgery, Pennsylvania

Dr. Pasquale said that when the physicians began to develop practice management guidelines, their first concern was to maintain quality. The trauma surgeons asked questions about patient problems, diseases, or processes and tried to critically appraise the findings.

The literature review included relevant literature with room for expert clinical opinion and was as comprehensive and specific as possible. A level 1 recommendation was given if the data were convincing and justifiable based on the science alone; level 2 if the data were reasonably justifiable and strongly supported by expert opinion; and level 3 where data showed science was lacking but widely supported by what was available. The group did not want to be held to a standard so they used level in their recommendations. In June 1998, the first four guidelines were published in the Journal of Trauma and are available on the EAST Web page (http://www.east.org/tpg.html).

Guidelines are put in a protocol that are then put at the bedside, and every bed in the trauma unit has a protocol book that gives nurses, residents, physicians access to about 30 protocols. Most of them are evidence-based. A modified consensus rule has been implemented, which states that persons do not have to entirely agree with the protocol or the guideline, but they must agree to try it.

Return to Contents

Plenary Session

President's Quality Commission Report and Recommendations on Evidence-based Medicine

Mary Wakefield, Ph.D., R.N., F.A.A.N., Professor, College of Nursing and Health Science, George Mason University

Dr. Wakefield discussed the Advisory Commission on Consumer Protection and Quality, which is set up to advise the President on changes occurring in health care delivery and to make recommendations about how to assure and how to promote health care quality. The Commission's first document is a consumer bill of rights and responsibilities. National aims for improvement need to be concise and selected based on certain criteria. The Commission offered aims for improvement, the first two of which are directly related to evidence-based medicine: expand research on new treatments and evidence on effectiveness and assure appropriate use of services.

The Commission's strategy of building capacity to improve quality focused on four areas: (1) fostering evidence-based practice and innovation, (2) encouraging organizations to adapt to changes, (3) engaging the health care workforce, and (4) investing in information systems. Recommendations from the Commission's chapter on evidence-based practice were summarized.

Return to Contents

International Perspectives

Translating Evidence to Practice: Models for Evidence-based Implementation—Richard Grol, Ph.D., Professor, Department of General Practice and Social Medicine, Centre for Quality of Care Research, University of Nijmegen, The Netherlands

Dr. Grol described the initiative undertaken by the Dutch College of Family Physicians to set national guidelines for family practice. About 70 guidelines have been set for common problems, using a systematic and rigorous development procedure. Acceptance and effectiveness of these guidelines indicates that about 80 percent of doctors know about them within 6-12 months, 60-70 percent discuss them in their practice teams, and more than 90 percent think they are useful for education and local arrangements. Although they are well accepted because they are developed by practitioners, an increasing number of doctors fear these guidelines will be abused by patients in malpractice suits, or insurers/health authorities will abuse them in the name of controlling costs.

The national strategy for guideline implementation is a comprehensive approach using the national level, regional/local level (education to groups of doctors/nurses, regional arrangements, regional coordinators), and practice level (outreach visits by trained facilitators). It has been very successful in the first 2 years, ascribed primarily to its comprehensiveness, addressing all potential obstacles and barriers, and using facilitators in the implementation process.

Toward Evidence-based Implementation: The Cochrane Effective Practice and Organisation of Care Review Group—Jeremy M. Grimshaw, M.B.Ch.B., Ph.D.,Programme Director, Health Services Research Unit, University of Aberdeen

Dr. Grimshaw said the Cochrane Collaboration (called EPOC) is "an international organization that aims to help people make well-informed decisions about health by preparing, maintaining, and ensuring the accessibility of systematic reviews of the benefits and risks of health care interventions." EPOC has reviewed the work of more than 100 researchers to review the evidence of effectiveness of different implementation strategies.

EPOC is conducting systematic reviews of studies of interventions designed to improve professional practice, including professional, financial, organizational, and regulatory interventions. Systematic reviews suggest the potential of a variety of strategies to be effective under certain conditions. The choice of strategy should be based on consideration of targeted activity, targeted health care professional groups, perceived barriers to change, available resources, and management of change processes. In most cases, multifaceted interventions that address several perceived barriers to change should be used.

The French Clinical Guidelines and Medical References Program—Hervé Maisonneuve, M.D., Director of Evaluation, Agence Nationale de l'Accreditation et de l'Evaluation en Sante, Paris, France

Dr. Maisonneuve said the objectives of the National Agency of Accreditation and Evaluation in Healthcare (ANAES) medical references program are to improve the quality of care and the cost effectiveness of health care in the ambulatory sector. Three evaluation tools are used at once—standardized computerized medical files in private practice, a national coding system, and the regulatory medical references (known as RMOs). The medical references program is composed of academic societies, experts, evidence-based medicine professionals, medical unions, insurance organizations, and ANAES staff.

Approximately 69 clinical practice guidelines have been implemented so far, about 15 per year. Methodology for the development includes support of expert working groups, review by peer reviewers, and literature review and gathering of expert opinions. Production time is 7-9 months, with two 1-day meetings of working groups and a third meeting after the peer review by mail.

Guidelines are disseminated using inhouse publications and medical journals, and sometimes executive summaries are mailed to 150,000 French doctors with high impact. Still to be decided is whether guideline compliance should be offered with positive incentives or negative controls, and what the impact of each approach is.

Return to Contents

Changing Behavior To Improve Quality of Care

Clinical Judgment and Physician Decisionmaking—Roy M. Poses, M.D., Associate Professor, Department of Medicine, Brown University

Dr. Poses said three modes of decisionmaking exist: cognitive, behavioral, and sociological. The key questions to ask about changing physician decisionmaking are: (1) Does the decisionmaking need to be changed? (2) What is the problem with the decisionmaking? and (3) How could the decisionmaking be changed?

The judgment and decision psychology literature, applied to medicine, indicates the following mental models of decision processes:

• Automatic/production rule/appeal to authority—no clear justification of the decision except by appealing to authority.
• Causal/pathophysiologic/theoretic—decisionmaking based on theories of causation; the reasoning makes theoretic sense, but it is not evidence-based.
• Ordinal/evidence-based/rational—decisionmaking based on balancing the likelihoods and values of outcomes in some logical, rational manner. (This is the model most often used by doctors.)
• Decision-analytic—decisionmaking based on quantitatively assessing the probabilities and utilities of outcomes, determining the expected value of decision options, and choosing the option with the best expected outcome.

What can go wrong with how decisions are made includes: failure to identify options; problems identifying outcomes of options; wrong outcome probabilities are postulated; wrong outcome values are assigned; and wrong method to combine probabilities and values is used. Causes of wrong outcome probabilities can be applied to how physicians think to make critical decisions, which include judgmental heuristics, judgmental biases, ego bias, value bias, and group polarization.

The Role of Information Systems in Improving Decisionmaking—Richard E. Ward, M.D., M.B.A., Director, Product Management, Oceania, Inc.

Dr. Ward stated that because behavior is the result of a process, the tools of process improvement can be used if the goal is to change behavior and, ultimately, outcomes. The Clinical Quality Improvement Cycle contains the core processes of decisionmaking and care delivery, with the effects of decision aids, quality indicators/feedback/other performance indicators, literature and experts, outcomes, guidelines/algorithms, and clinical policymaking. This cycle is a combination of quality and outcomes measurement.

Methods and technological tools support clinical process improvement. For simple processes, awareness of the problem should be increased, and clinical performance reports to provide feedback and offer reminders and alerts should be used. To improve complex processes, interventions should be set up using multidisciplinary clinics and developing and implementing a care map or protocol. It will be necessary to coordinate many tasks over time, often involving many different medical professionals practicing in different locations (home health care workers, pharmacists, ambulatory care clinicians, inpatient clinicians, etc.). Information tools should be developed that will assist in acquiring data directly from patients to evaluate overall outcomes of care, clinical messaging (to increase coordination within the care team), and workflow automation (to organize tasks among a group of people).

An example of a simple intervention is a study of an adult flu immunization intervention. First-year implementation reached 25,000 patients; by the third year, it reached 70,000 patients. Results showed a 5 percent improvement during the first year that translated to a savings of $118,000 in a nonepidemic year and $250,000+ savings in an epidemic year. An example of a complex intervention is a study of computer-assisted care management, a template charting application in which the clinician selects the starting template and structured text is inserted into the "note."

Changing Behavior to Improve Quality of Care—Stuart J. Cohen, Ed.D., Department of Public Health Sciences and Internal Medicine, Wake Forest University School of Medicine

Dr. Cohen noted that common reasons for discrepancies between guidelines and behavior include the belief that change is not needed (current care is appropriate or no gap exists), there is no problem, and change is not possible because of a perceived lack of skills or a perceived lack of support (patients do not want change, coworkers do not think it is essential, administrators will not support it, and/or the change is not affordable).

Producing change is sometimes more complex than originally envisioned. Theories of readiness for change state that change evolves through a series of stages that include: precontemplation, contemplation, preparation, action, and maintenance. This model is not unidirectional, and because these concepts pertain to patient behavior, they may also pertain to provider behavior.

Traditional strategies to motivate physicians include: continuing medical education, special workshops and training programs, national consensus guidelines, and pharmaceutical representatives. Alternative strategies to motivate physicians are the use of influential physicians, performance feedback, and financial incentives.

Especially with disease-preventive services, the problem may lie within the physicians's office system. Effective solutions must address the entire office system, including non-physician personnel, to create an ecology that helps physicians practice the way they want. Action cues to improve disease prevention services include chart flow sheets, reminder systems, use of nonphysician personnel, patient initiated cues, and academic detailing. Strategies for improving guideline adoption include: enlist influentials, use focus groups, use local guidelines, provide snapshots, provide gentle feedback, and use a systems approach.

Return to Contents
Proceed to Next Section