Section IV: Evidence Based Practice: Does It Work in Practice?

INTRODUCTION

Throughout the first three sections of this text, we have presented principles to guide us in each of the three elements of evidence-based practice. Section I focuses on our clinical judgment and decision making processes; Section II on the importance of knowing and respecting our patients' values and choices; Section III on identifying and appraising evidence from the literature. In all three sections, we have provided examples of clinical application of these principles. But a central question remains: Can and do clinicians actually use these principles in their own clinical practices?

In this concluding section, we explore the answers to this question. Research has shown us that human beings need time to adopt recommended changes and that there are some typical steps in that adoption process. In addition, we know that there are barriers to adoption that are also fairly consistent across all types of change in many different groups of people.¹ In Chapter 17, we present what is known about the diffusion of change in general, what we know about the adoption of evidence in health care, and the barriers to and opportunities for change in the future.

In health care we have come to rely on the patient record as the source of information about all aspects of patient care.² In Chapter 18, Kristin von Nieda discusses the importance of documentation as the basis for understanding our response to change in our practices, the importance of good outcome measures in this documentation, and the linkage of good documentation to quality measurement and improvement activities.

In Chapter 19, Christopher Bise and Anthony DeLitto delve more deeply into the connections of evidence based practice and quality improvement. They describe a successful program to assess the clinical performance of the therapists in their practice regarding a specific patient population. This example demonstrates that, indeed, we can put evidence based practice principles into practice!

REFERENCES

INTRODUCTION

The previous sections have presented a foundational introduction to the three main components of evidence-based practice (EBP). Clinicians will find that most decisions required of them can benefit from the integration of sound clinical reasoning, patient-focused preferences, and critically appraised evidence. Whether you are selecting a diagnostic test or an intervention, your odds of having the best patient outcomes are enhanced when you ground your decisions in EBP. We are finding increasing evidence to support this assumption.^1,2 But knowing and doing can be miles apart when it comes to human behavior, and clinicians are susceptible to barriers that make acceptance of EBP more difficult. Understanding the slow pace of change in the health care system makes it important to identify strategies to quicken our abilities to get the right care to the right patient at the right time. Increasing the rate of change to best physical therapy practice would result in significant improvements in our patient outcomes. Evensen et al³ refers to this as closing the evidence-practice gaps. Although most authors believe this is an appropriate goal, it will require the removal of barriers at several levels of health care, including the patient, the practitioner, and the health care organization.

Researchers and clinicians who identify problems with adherence to best practice, and who design and test interventions to decrease these barriers, are participating in what is called knowledge translation research or implementation science. The Canadian Institutes of Health Research define knowledge translation as:

Evensen et al³ have described a disappointingly low amount of high-quality evidence in the past decade for guiding our understanding of how to implement the changes we hope will close the evidence-practice gap. The literature in medicine reports a preponderance of observational studies or simply papers that restate the practice guidelines or treatment recommendations.

In this chapter we will discuss three elements that can affect, in either a positive or negative manner, the process of getting evidence into practice: dissemination, practitioner acceptance, and organizational support for EBP (Fig. 17-1). We will draw from the best available evidence in physical therapy to identify barriers and recommend interventions that have been effective. For some knowledge translation, the path to the desired change is rather straightforward, whereas for other innovations the path may be more complex, as illustrated in Figure 17-2. Whether you are attempting a simple or a complex change in practice, the information in this chapter will be useful for individual clinicians as well as communities of practitioners and managers who wish to facilitate best practice in their settings.

DISSEMINATION OF INNOVATION

Perceptions of Innovation

This term, innovation, may sound too grand for the steady changes we hope to see. We might think it is reserved for things like the creation of the Internet, a truly innovative development in society, so that recommended changes to our practice pale by comparison. However, any change that represents a new method or practice should be considered an innovation. Clinicians and scientists in physical therapy are innovators as they attempt to evaluate a new exercise technique or try a new position for applying joint mobilizations. Once the evidence supporting the novel approach to care of our patients has been tested and found effective, the task turns to disseminating this good news. Many say this is the point at which EBP meets a significant barrier. In a seminal paper on disseminating innovations in health, Berwick states, "In health care, invention is hard, but dissemination is even harder."^5,p1970 Jette,⁶ discussing these issues in physical therapy, comments that our profession must take studies of how best to diffuse innovations just as seriously as studies that produce the innovations.

Dissemination in Physical Therapy

The Institute of Medicine has drawn our attention to the problems of dissemination of evidence in health care in its classic report entitled Crossing the Quality Chasm.⁷ The report concludes, "Between the health care we have and the care we could have lies not just a gap, but a chasm." What do we know about the gap or chasm between the physical therapy care we have and the care we could have? Evidence supporting a diffusion of innovation problems in physical therapy is increasing. Several research studies have investigated the rate of physical therapists' adherence to professionally accepted practice guidelines.^{8,9,10,11,12,13,14} Mikhail et al¹¹ documented the typical practice decisions of 100 physical therapists for patients with acute low back pain and found 68% using interventions with strong or moderate evidence of effectiveness, while 90% also used interventions for which research evidence was limited or absent. Bekkering et al¹⁰ and Brennan et al⁹ studied patient outcomes for physical therapists provided with knowledge of current practice guidelines using one of several educational strategies and found mixed results both in practitioner adherence and patient outcomes, demonstrating how complex the problem of changing practice can be.

Classic methods of dissemination of research evidence in the past century required researchers to publish their work in peer-reviewed journals and clinicians to find that work and read it. Alternatively, they could attend professional conferences, at some expense, and attend research platform or poster sessions in search of practice innovations. This approach placed a huge demand on clinicians to access published research and to improve their abilities to critically appraise the work.

In Section III we demonstrated a variety of new strategies that are now available for accessing and understanding innovations. Clinicians are calling for a broader array of methods of dissemination to enhance the speed of change in clinical practice. A consensus panel of emergency medicine physicians, working to reduce barriers to dissemination, recommends that "researchers and advocates should disseminate findings through multiple forums beyond peer-reviewed publications when an ED-based (emergency department-based) public health intervention has enough evidence to support integration into the routine practice of emergency care."^15,p1133 Professional associations are playing a role in this aspect of dissemination by using e-mail to notify practitioners of current guidelines.

What types of dissemination strategies work? Spreading carefully collected evidence throughout a profession as diverse as physical therapy is a challenge. Practitioners in our field are diversified in setting and in specialization, such that targeting the information needed by each practitioner is a challenge. New technology can be quite helpful in pushing focused knowledge to practitioners in a prescriptive manner that affords good access of just the right amount and type of information. Barriers to dissemination have included limited access as well as limitations in comprehension.

Sufficient studies of methods of dissemination in physical therapy have been conducted for a systematic review of this work. Menon et al¹⁶ found moderate evidence from one high-quality randomized controlled trial (RCT) and two observational studies to suggest that the use of an active, multicomponent knowledge translation program to be successful in improving knowledge. They also found strong evidence to support the effectiveness of multicomponent interventions to change practice behavior, including increased self-perceived and actual adherence to practice guidelines. In this systematic review, the knowledge translation interventions used in the studies did not demonstrate significant changes in physical therapists' attitudes toward EBP.

Dissemination of Evidence to Consumers

What lies ahead in health care is an increasingly educated consumer who wishes to be an active participant in his or her health care decisions. In response, researchers have begun to study the process of creating, tailoring, and disseminating evidence to consumers. Tugwell et al state their aim this way: "to put the results of rigorous research into the hands of consumers so they can use it when making important decisions related to their health."^17,p1729 The Cochrane Consumers and Communication Review Group supports this dissemination to consumers and recommends several foci for intervention development, including interventions directed to the consumer and interventions for communication exchange between providers and consumers.¹⁸ Some of these interventions include the development of educational materials in various formats, termed decision aids, to prepare people to participate in "close call" decisions that involve weighing benefits, harms, and scientific uncertainty.^19,20 It is often said that to teach someone you must master the knowledge you wish to teach, and in the teaching you enhance your knowledge further. Thus, developing decision aids is perhaps another strategy for improving the dissemination of innovation to the clinician.

PRACTITIONER ACCEPTANCE

Seminal work on diffusion of innovation comes from a broad range of theoretical frameworks developed by researchers in fields as diverse as communications, public policy, health policy, agriculture, family planning, and nutrition. Everett Rogers, a well-known diffusion scholar, published his first text on the topic in 1962. Over the past five decades, knowledge about factors that influence the rate of dissemination for a beneficial behavior have been developed from studies as diverse as new hybrid seed corn in agriculture and public health studies of water boiling in a Peruvian village.²¹ Across these many studies Rogers has determined that the perception of the innovation as well as the characteristics of the individuals being encouraged to adopt an innovation can significantly affect the rate of adoption.

Practitioners' Perspectives

The perceptions of the innovations we desire in physical therapy are also likely to influence their rate of adoption. Clinicians should find the innovation to be beneficial to themselves or their patients, and the risks of adopting the innovation should not outweigh these perceived benefits. Related to the clinician's perception of benefit is the perception that the innovation is not in conflict with personal or professional values. Innovations that decrease the physical therapist's autonomy to make clinical decisions in conjunction with the patient will likely be rejected. A third perspective that can affect the rate of innovation adoption is its complexity. A complex innovation may not result in a positive cost/benefit analysis from the clinician's perspective. One solution in this situation is simplifying the innovation in some manner. Berwick states, "… innovations are more robust to modification than their inventors think, and local adaptation, which often involves simplification, is nearly a universal property of successful dissemination."^5,p1971

Rogers identified two additional significant perceptions that seem very applicable in a physical therapy clinical context: trialability and observability. Trialability is described as the ease with which an adopter might test the proposed innovation in a small context to determine the outcomes before a widespread implementation. One would imagine that this is a likely first step for trying out a new patient intervention. Observability refers to the ease with which a potential adopter might observe the innovation before testing it.²¹ This also may be an important and practical perception for physical therapists, who customarily discuss practice decisions with colleagues.

Practitioners' Innovation Adoption Patterns

Rogers' work identified a pattern of innovation adoption behaviors based on the proposed personal characteristics of individuals targeted for the innovation.²² He describes five categories, based on the mean adoption time. Figure 17-3 describes these categories in relation to a statistical normal distribution. The individuals who are much above average in time to adopt an innovation are called innovators. These are generally individuals who have a high tolerance for risk in decision making and are very interested in novel ideas. In health care, Berwick⁵ states, they might be considered as mavericks. The next group, approximately 13%, would also adopt an innovative idea more quickly than the average person and are termed early adopters. This group, considered to be composed of significant opinion leaders, is willing to test innovative ideas and has little personal difficulty with changing the way things are done. They are often respected leaders, and internal and external colleagues pay attention to their opinions. The next group, the early majority, includes those who watch the early adopters and take their lead from them. They generally adopt an innovation at or above the mean rate of adoption. They are more risk-averse than the early adopters and value an innovative idea with strong local importance rather than testing a novel idea that does not have a direct application to their work. The last two groups will adopt innovation at a rate below the average. The late majority tends to watch the early majority for signs of a "new status quo,"^5,p1972 choosing not to change their behaviors until it is quite certain that the innovation has merit. The last group to adopt are termed "laggards" by Rogers or "traditionalists" by Berwick. These individuals are strongly connected to the status quo and tend to be reticent to adopt innovation without significant proof of concept.

What influences a physical therapist to become an early adopter or innovator? Values and beliefs might be predictors of the inclination to help spread evidence based innovations. Our professional core values include the pursuit of excellence in practice, and these are reinforced by the core values of accountability and professional duty. It is true that clinicians with similarly strong core values could fall into any of the five rate of adoption categories described by Edwards, but this provides an excellent platform for debate of a particular innovation. Clinicians whose values foster an engaged attitude toward changing practice patterns can be identified and supported in their work by an organization committed to the pursuit of best practice models, the third element of the process of diffusion.

ORGANIZATIONAL CHARACTERISTICS

Berwick⁵ describes the third collection of factors that will influence the rate of dissemination as contextual or managerial influences. Health care organizations of all sizes can create nurturing environments for adoption and innovation of best practice policies. A practice or organization that wishes to allow some clinicians to pursue innovation would arrange time and resources for the creative work this entails, taking into account that productivity may be sacrificed as various new approaches are tested. Identifying a likely innovation in published and peer-reviewed literature is the first step in the process, and it is facilitated by providing resources to push these ideas out to practitioners for consideration, perhaps via automated literature searches or with the assistance of informatics professionals. Identifying and supporting a likely innovator is the second step, providing time to test the innovation on a small but particular scale in the practice.

Once the organization decides to pursue the dissemination of a particular innovation, one of several paths may be chosen: optional, recommended, or required behavior changes.⁵ If the innovation will be challenging to integrate into the organization, and also of as yet unquantifiable value, an optional approach might fit best. This optional path allows the collection of additional quality improvement data to further judge the value of the innovation in the setting. Choosing to label the change in practice as either recommended or required will communicate a very different message to organization members, one that they may have difficulty distinguishing. Matching the diffusion approach to the impact of the innovation is a crucially important management decision, as the use of one predominant strategy for all innovations is unlikely to be as successful. For example, Lankshear describes a strategy with multiple options to recommend a change to encourage physicians to document the stage of disease in cancer patients, including knowledge transfer strategies such as providing advice from opinion leaders in the setting, practitioner education sessions, or audit feedback on performance.²²

Successful research dissemination or knowledge transfer strategies are typically created to meet the needs of the clinicians who are to use the knowledge in making clinical decisions and to accommodate the environment in which they practice. MacDermit et al²² suggest it is important to consider organizational factors that may facilitate or impede knowledge transfer. To identify some of these factors in physical and occupational therapy practice, they proposed a study of knowledge transfer in the use of standardized outcome measures in adult and pediatric patients with chronic health conditions. The knowledge transfer outcome was described as strategies to "select health status measures for clinical practice, score/interpret results, incorporate measures into clinical reasoning, and recognize and address personal and organizational barriers and facilitators of change."^22,p6 Two types of organizational interventions were planned: a 30-hour session of faculty-led tutorials combined with facilitated independent group work in a problem-based learning format and a Web-based instructional presentation over a 6-week period using a similar problem-based learning format, with the learners not in the same location. Although the results of this work are not yet available, the researchers planned to interview subjects regarding their perceptions of facilitators and barriers to knowledge transfer in their practice setting.

A variety of management strategies to enhance knowledge transfer in practice are reported in the literature. Most traditional are continuing education courses in one format or another. In the most recent Cochrane review of the effectiveness of continuing education for health professionals, the evidence is very weak regarding the effectiveness of traditional continuing education as an intervention for changing practice. Forsetlund et al report a small effect for improving patient care following continuing education interventions, similar to those found for auditing of practitioner compliance, practitioner feedback, and educational outreach.²³

Practitioner audit and feedback are described as a process by which objective data on practitioner compliance are provided to the practitioner to enhance the quality of care given. Audits can take many forms and can be implemented electronically or in a more intense, personal counseling process. Jamveldt et al report in a Cochrane review of audit and feedback techniques for knowledge transfer that they can have a small to moderate impact on professional practice. The effects of audit and feedback were greater when the practitioner's baseline adherence to recommended practice is low and when feedback is delivered more intensively.²⁴ Foy et al²⁵ found that audit and feedback strategies are used extensively in the United Kingdom, but there is insufficient evidence regarding the best methods of applying this quality improvement technique in an organization.

Organizational strategies for quality improvement are also considered as strategies for knowledge transfer. Educational outreach visits can be targeted to a single clinical site or a community outreach to enhance the practitioner's awareness of an innovative practice. These visits often employ local opinion leaders or clinician colleagues who have been nominated or recognized as being influential in a community. The decision support systems discussed in Section III are also tools that may be structured to facilitate the adoption of a new practice guideline. For example, electronic reminders of a new initiative at a clinical site can help clinicians who wish to adhere to best practice models become compliant. Chapter 19 reports on a program instituted in one organization designed to improve the quality of care through use of clinical data and feedback to clinicians.

ADDITIONAL PERSPECTIVES ON KNOWLEDGE TRANSLATION

There may be many additional facilitators and barriers to changing practice in physical therapy than those covered in this chapter. We need additional evidence on this topic: success stories and descriptions of unique barriers in various settings or for the specific types of patients we treat. The research on knowledge translation itself must be scrutinized for criteria that indicate strength of evidence. Two journals created in the past decade are devoted to translational research: Implementation Science (www.implementationscience.com) and the Journal of Translational Medicine (www.translational-medicine.com). Majumdar²⁶ has reviewed three high-quality and successful knowledge-to-action studies in medicine and suggests some criteria for translational research. He recommends translational studies that provide sufficient detail for replication, indicate a clinically meaningful change score, and analyze both the process of practitioner change and the accompanying patient outcomes. He also suggests a research agenda for translational research that would investigate the persistence of practitioner change behaviors as well as the occurrence of relapse to old habits.

The barriers to implementation of physical therapy EBP are similar in many respects to those for other areas in health care. Salbach et al in a study of Canadian physiotherapists identified two organizational barriers to EBP for patients who have had a stroke: a lack of training in EBP and time in the work day to search and interpret the literature.²⁷ In a later study, Salbach et al reported similar organizational barriers of time and training and recommended the incorporation of technology in the form of personal digital assistants for ease of accessing evidence, as well as educational sessions led by experts with opportunities for case-based learning and skill training.²⁸ This combination of technology aids and face-to-face educational sessions was considered the strongest for actually changing practice.

However, in many countries physical therapists may experience a unique barrier that relates to autonomy in clinical decision making. This was a finding of a qualitative study of 43 physiotherapists in Belgium.²⁹ In addition to identifying obstacles to dissemination of knowledge, the physiotherapists were concerned with their limited autonomy in clinical care as compared to that of the referring physicians. This perspective is a primary barrier to knowledge translation, for a practitioner who lacks self-efficacy in clinical decision making is unlikely to risk adopting an innovation in practice. We consider autonomy in clinical decisions to be a necessary condition for knowledge translation.

In terms of access to knowledge, the clinician's dilemma is not only about having too little evidence to guide practice; it is also about the translation of the right evidence into practice in a timely manner. The practice community must depend on the research community and those who work to prepare valid and unbiased summaries of evidence for practice. Tricco et al³⁰ suggest that knowledge translation that depends only on individual studies runs the risk of a new practice being adopted without sufficient support and that knowledge syntheses should be "the basic unit of knowledge translation."^30,p.11

Straus suggests that we are wise to avoid the knowledge translation imperative that all knowledge must be implemented.³¹ Our profession will benefit from a mature and selective approach to knowledge translation so as to avoid the whiplash phenomenon associated with repeatedly contradictory health advice, such as we have experienced in human nutrition. In our profession, the now rejected "no pain, no gain" advice provides an example of the difficulty of presenting a trustworthy message to the public about exercise.

CONCLUSION

Shea,³² reporting on a decade of knowledge translation research, points to several major success stories, including a significant reduction in heart disease attributed to a multifaceted knowledge translation project on smoking cessation. She also reminds us that in the 21st century we have many more tools available to engage a community in knowledge translation, including media and Internet social networking sites. These tools can facilitate the goal of knowledge translation to enhance the health-related decision making of all—patients, practitioners, managers, and policy makers. If we keep patients at the center of this process, as the targeted benefactors of the best evidence and clinical decision making we can provide, we can enhance physical therapy care for all.

SELF-ASSESSMENT

1. There are several studies (listed below) that support the value of exercise in people like Mr. Ketterman with or at risk of Alzheimer's disease. What barriers would you expect to find in convincing your colleagues to implement exercise programs for this patient group?

2. What strategies would you use to overcome these barriers?

References for Self-Assessment Activity

• Abbott RD, White LR, Ross GW, Masali KH, Curb JD, Petrovitch H. Walking and dementia in physically capable elderly men. JAMA. 2004;292(12):1447–1453.

• Larson EB, Wang I, Bowen JD, McCormick WC, Teri L, Crane P, Kukull WA. Exercise is associated with reduced risk for incident dementia among persons 65 years of age and older. Ann Intern Med. 2006;144:73–81.

• Scarmeas N, Luchsinger JA, Schupf N, Brickman AM, Cosentino S, Tang MX, Stern Y. Physical activity, diet, and risk of Alzheimer disease. JAMA. 2009;302(6):627–637. doi: 10.1001/jama.2009.1144.

• Teri L, Gibbons LE, McCurry SM, Logsdon LG, Buchner DM, Barlow WE, Kukull WA, LaCroix AZ, McCormick W, Larson EB. Exercise plus behavioral management in patients with Alzheimer disease. JAMA. 2003;290(15):2015–2022. doi: 10.1001/jama.290.15.2015.

• Weuve J, Kang JH, Manson JE, Breteler MMB, Ware JH, Grodstein F. Physical activity, including walking, and cognitive function in older women. JAMA. 2004;292:1454–1461.

• van Gelder BM, Tijhuis MAR, Kalmijn S, Giampaoli S, Nissinen A, Kromhout D. Physical activity in relation to cognitive decline in elderly men: The FINE Study. Neurology. 2004;63:2316–2321.

CONTINUED LEARNING

• The Agency for Healthcare Research and Quality of the U.S. Department of Health and Human Services has supported research on the concept of pay for performance (http://www.ahrq.gov/qual/pay4per.htm), which is designed to align payment incentives with appropriate care based on sound evidence. Review one of these programs and reflect on:

1. Is the evidence sound?

2. Are the incentives based on sound principles?

3. What might you do differently?

REFERENCES

INTRODUCTION

In Section III, we discussed identifying, assessing, and applying, as appropriate, evidence from the research literature to decisions about patient care. Yet, it is apparent that the research literature does not hold all the answers to all the questions we have about patient care. In Chapter 19, Drs. Bise and Delitto discuss an approach that uses the data generated in everyday care as a source of evidence. This approach, as with many evidence based activities, relies on documentation to be successful.

The goals of this chapter are to examine the multiple purposes of documentation and how these purposes may be met in producing accurate and useful documentation about a patient's condition. Thus, improved documentation can become a source of evidence that aids in decision making.

THE HEALTH CARE RECORD

The patient/client record serves multiple purposes. According to the American Health Information Management Association (AHIMA), documentation has many purposes. For most clinicians, most of the time, the primary purpose of documentation is to provide a basis for clinical reasoning in planning care and assessing its success for the individual patient. Ideally, the patient/client record also serves as a communication tool for providers involved in the care and supports patient/client care delivery. Appropriate and meaningful documentation is important to support optimal decision making, to minimize risk, to promote continuity of care, and to determine the best plan of care for the patient/client.

Documentation is also a mandatory practice, subject to applicable jurisdictional and regulatory requirements. There are specific legal requirements about when and how to document patient care and charting errors. There are additional requirements for signature and other means to demonstrate the authenticity of the record. Payment for services is justified through documentation in the patient/client record. Different payment sources have specific mandates for documentation to justify both approval for services and to determine which, if any, services are eligible for payment.

All of these purposes matter in patient care, but the last one, a means for data collection and retrieval, may be the most important in terms of evidence based practice. When the clinical record contains standardized information, then these data can be retrieved and analyzed, both within and across practices. Secondary purposes are education, regulation (compliance and accreditation), research, policy making (allocating resources and business planning), and industry needs (including research and development).¹

Standards for Quality in Documentation

AHIMA² has defined comprehensive guidelines the patient record should meet if it is to serve as a basis for data collection through the collection of high-quality data. The characteristics are:

• Accessibility

• Accuracy/Validity

• Comprehensiveness

• Consistency

• Currency

• Clear definition

• Granularity (the correct level of detail)

  • Relevancy

  • Precision

• Timeliness

The purposes for which documentation is currently used extend beyond the reasons cited above and include risk assessment and management, quality assessment and improvement, peer review, and productivity measurement. When taking the manifold purposes of documentation into consideration, the task of documenting clearly, correctly, comprehensively, and meaningfully becomes even more important.

Documentation is defined in the Guide to Physical Therapist Practice³ as "any entry into the client record, such as a consultation report, initial examination report, progress note, flow sheet/checklist that identifies the care/service provided, re-examination, or summation of care." The American Physical Therapy Association gives general guidelines in its Guidelines for Physical Therapy Documentation that expand on specific elements that may be included. (The Guidelines may be found at www.APTA.org and at the F.A. Davis Web site for this book.) Documentation is a complex process that has many discrete parts. The Guidelines provide both lists of the various types of documentation that are seen in physical therapy and guidance for the standards that should be used in creating them.³ Standards include the need for clear authentication as to the physical therapist of record for the patient. Table 18.1 shows some of the many different kinds of documents created in the physical therapy record.

BARRIERS TO QUALITY DOCUMENTATION

If all documentation possessed the characteristics discussed above, we would be able to readily use it for developing evidence about the success of care. But there are disparities in much of the physical therapy documentation that preclude it from serving as a meaningful database. Practitioners, including physical therapists, often complain about the time needed to complete documentation as a primary reason for not producing documentation of the best quality.^4,5,6,7,8,9 Several other reasons are also apparent, as are discussed later.

Focus on Intervention

Current documentation commonly does not address all the appropriate elements listed in the APTA Guidelines. The examination may be limited to the identification of impairments and functional loss and may lack a meaningful, complete evaluation, diagnosis, and prognosis. Throughout the continuum of care, documentation commonly focuses on identifying impairments and listing interventions. Flow sheets and checklists are frequently used to record interventions, but there may be no additional documentation to substantiate the patient's/client's responses to the interventions and the subsequent need to revise the plan of care. Too often, the documentation lacks apparent connections between interventions, functional problems, and outcomes. Physical therapists commonly document a discharge summary but may fail to include pertinent discharge criteria about anticipated goals and expected outcomes.

Payment Incentives

Often, practitioners focus on payment for services as the primary reason for documentation of patient care. Economic factors can influence symptom reporting, as can pending litigation with potentials for monetary rewards. Payment for services may promote unfavorable behavior changes in practitioners, such as misrepresentation of information in the patient/client record to obtain reimbursement from insurance companies. Wynia and colleagues¹⁰ surveyed 720 practicing physicians and reported that 39% used falsified patient information or exaggerated symptom severity to obtain insurance authorization. As an explanation for this deceptive documentation, the authors offered the term covert advocacy to describe the behavior that results from the tension between patient advocacy and cost containment,.¹⁰

Nadler et al (2004) reviewed documentation of a straight leg raise (SLR) test in 200 people with a diagnosis of lumbar radiculopathy or disc herniation. Documentation of a positive SLR was more likely if patients were covered under personal injury insurance programs than if they were covered under a managed care workers' compensation program. The likelihood of a positive bilateral SLR diagnosis was even more strongly associated with a personal insurance program. The authors concluded that there might be numerous explanations for these differences, including exaggeration of symptoms by participants, covert advocacy, differences in education levels of providers performing the test, and poor technique or improper interpretation.¹¹

Although these studies involved physicians, not physical therapists, it is not too far a leap to imagine that deceptive documentation could occur in physical therapy practices as well. For example, many physical therapists believe that Medicare patients in the acute setting would not be admitted into a rehabilitation setting if the documentation said that they could walk more than 150 feet, so they record that the patient walks 50 feet times 3, an example of covert advocacy. Or they believe that patients would not be considered homebound if they walk more than 100 feet; therefore they may well feel the need to practice covert advocacy and record that a patient can ambulate fewer than 100 feet when this is not true.¹² Interestingly, Medicare guidelines focus on the need for skilled care but do not specify number of feet as a decision determinant.

Lack of Standardized Terminology

The lack of standardized terminology across health professions has affected the quality of documentation. Poor documentation leads to uncertainty in interpreting patient information, impairs communication among providers, and decreases the continuity of care. Müller-Staub et al (2007) evaluated the impact of an educational program to improve implementation of standardized nursing diagnoses (NANDA), interventions (NIC), and outcomes (NOC) based on the premise that standardized language would lead to improved data management and enhanced quality assessment of nursing practices in the areas of diagnosis, intervention, and outcomes. Results indicated that standardized language led to an improvement in the formulation of nursing diagnoses, identification of signs/symptoms and correct etiologies, specific nursing interventions, and documentation of nursing-sensitive patient outcomes.¹³

Oostendorp et al investigated the intrarater and interrater reliability and internal consistency of physical therapy documentation based on a set of quality indicators developed for the study. The authors also looked at the clinical utility of the set of quality indicators and evaluated the quality of physiotherapy patient documentation. They developed a set of 44 indicators, organized into six clusters, representing phases of the physical therapy examination/evaluation process. They are:

1. General (general patient information)

2. Intake

3. History taking

4. Examination

5. Treatment

6. Discharge

Three individuals, a physiotherapist/movement specialist, a physiotherapist, and a health researcher, conducted 147 chart reviews from 160 practices in the Netherlands using the set of quality indicators from the six clusters and a scoring system from 0 to 3 (0 = absent; 1 = somewhat present; 2 = mostly present; 3 = always present). After determining internal consistency, intrarater reliability, and interrater reliability, they concluded that the evaluation instrument has apparent clinical utility and that the quality of the documentation was what they termed moderate (mean score 1.6).¹⁴

IMPROVING THE UTILITY OF DOCUMENTATION AS A SOURCE OF EVIDENCE

If documentation is to become a reliable source of evidence, these barriers must be overcome. The recommendations to improve documentation, and thereby improve patient care, have two major themes in common: increasing use of electronic record keeping and improving the links between the documentation and the actual outcomes of care.

Use of Electronic Record-Keeping Systems

Electronic record keeping has been proposed as one solution for improving the quality and consistency of documentation. For example, Knaup et al developed a software system, eardap, that was based on an extensible architecture for using routine data for additional purposes. Just as with clinical research based on a standard terminology, the impetus for the development of this software was the concept that standardized documentation would lead to effective communication through sharing of data based on standardizing terminology.¹⁵ Medicare, along with numerous policy organizations and many insurance companies, has supported the move toward submission of electronic records for payment of services on the premise that electronic record keeping could:

• Improve the quality of documentation

• Increase the efficiency of communication, including that for payment

• Improve patient safety and therefore improve patient outcomes¹⁶

Despite the promise of electronic record keeping, evidence shows that computer-based systems are not yet the complete answer. A systematic review by Delpierre et al showed that while computer-based systems seem to increase practitioners and patient satisfaction, they were as likely as not to produce positive changes in practitioner behavior related to improved patient care.¹⁷ Because of the anticipation that properly implemented computer-based documentation will have great benefits, policy groups continue to try to identify the barriers to full implementation and the means for overcoming those barriers (Box 18.1).^18,19

A review of the available literature in physical therapy ²⁰ summarized the benefits (improved reporting, operational efficiency, interdepartmental communication, data accuracy, and access to data for future research) and barriers (need for workflow or behavior modification, software or hardware inadequacy, inadequate staff training) found in previous research about the use of computerized record systems in physical therapy. Based on analysis of the factors that seemed to lead to better success in implementation, the authors of the review recommended that the following principles be used in implementing computer-based systems for physical therapy documentation:

• Incorporate workflow analysis into the system design and implementation

• Include end users, especially clinicians, in the system development activities

• Devote significant resources to training

• Plan and test carefully to ensure adequate software and hardware system performance

• Pursue the efficient capture of coded data

• Commit to data standards

Support for more widespread adoption of computer-based systems has also been expressed in various APTA activities, including adoption of the policy Support of Electronic Health Records in Physical Therapy in 2008²¹ and by support of the Physical Therapy and Society Summit (PASS) in 2009. The Summit focused on a futuristic perspective of health care, including the many opportunities for the use of computer-based systems. One of the major themes of the PASS was also the concept of "disruptive change." Similar to the recommendations from the National Research Council (see Box 18.1), this concept emphasizes the need to be able to recognize and embrace change that makes fundamental differences in current systems.²²

Description of Actual Patient Outcomes

If documentation is to provide us with better information about patient care, one of the ways we need to improve is by using more standardized outcome measurements, when such measurement is available. By using standardized tools for measuring outcomes, the data in our documentation can be used in observational studies to describe the results of various diagnostic labels and interventions when used in actual practice, in other words, to help measure the effectiveness of our care.

There are many different kinds of outcomes in which we are interested as physical therapists. We need to measure outcomes across the disability model, starting with any changes to pathology that we might achieve, and then measure changes in function, including body structure, activity level, and participation in behaviors related to role. There are also many other areas where we would be interested in measuring outcome, including patient satisfaction, costs of care, and general, health-related quality of life. A good standardized outcome measure is based on many of the same principles discussed in Chapter 12. Standardized measures must be reliable, valid, and generalizable to the patient population for which we are documenting care. We need to know the ability of the test to measure both clinically meaningful and statistically significant change. All outcome measures should be as objective (free from measurement bias) as possible. Subjective measures are those that are much more susceptible to the bias of the observer.²³

In choosing an outcome measure, we can ask several questions:

1. How much system specificity do we need?

   A tool containing items that are clearly not relevant to the patient wastes time, but the more specific the measure, the more limited our ability to compare outcomes across large groups.

2. Does the tool measure the range of outcomes in which we are interested?

   Outcome tools have what is termed ceiling and floor effects. This means that some tools are very good at measuring outcomes in people with severe impairments and decreased function and other tools are much better at measuring outcomes in patients who started care with fewer impairments and a high level of function. Very few tools can do both.

3. Over what time frame are we concerned about changes in performance?

   Some tools are best at measuring change across the course of a typical episode of care, such as days or weeks. Others are best at measuring change over long periods, such as months and years, as would be appropriate for patients with long-term or chronic health problems.

4. Are we interested in measuring changes in one person over time or in comparing the results for groups of patients?

   This question is certainly also related to time frame but deals with another fundamental difference. Sometimes we need a tool to help us measure the individual patient's progress toward specific goals, and sometimes we are interested in comparing our overall programs, as in benchmark activities, or in creating databases for evidence based practice or clinical research.

5. Is the tool completed by observation of patient behavior or by patient report?

   It is not possible for a therapist to observe all behaviors, nor is it desirable. For example, we often see a patient after trauma and/or surgery and we need to rely on reports from the patient or family about prior level of function as the baseline outcome measure. Some tools, but not all, have been shown to reliably measure behavior through patient report.

6. Was the tool really created to measure outcomes across time or did it have another purpose?

   Many of the tools used in clinical practice have been developed for a particular purpose. For example, the Berg Balance Score was developed as a screening tool.²⁶ The Functional Impairment Measure (FIM) was developed as a way to compare group data across multiple rehabilitation facilities.²⁷ Neither of these was developed as a means to track outcomes for individual patients across time, although subsequent work has been done to assess their appropriateness as tools for this purpose.^28,29 Until the tool has been validated for our intended purpose, it is best to use it for its original, validated purpose.

7. Can I design a tool that is specific to my practice; can I adapt an existing tool to make it more specific to my practice?

   It is tempting, and all too common, to develop "home-grown," idiosyncratic tools for documenting outcomes in patient care. But such tools have not been studied for what is called their psychometric properties. In other words, we don't know their reliability, validity, or generalizability. Therefore, these tools may not be telling us very much at all about real changes in impairment or function based outcomes. And they certainly can't help us in developing databases for evidence based practice.

Just as with other aspects of evidence based practice, the adoption of appropriate standardized measures has been slower than would be desirable. Researchers examining the reasons for this have identified that in addition to the usual barriers, such as perceived lack of time and lack of knowledge, physical therapists also believe the standardized tools do not always provide sufficient description of the quality of movement being observed.^30,31,32 Stevens et al have described in a case study the adoption of more standardized outcome measures in one practice, a process very similar to that discussed in Chapter 17. They emphasized that implementation of outcome measures has to be inclusive of documentation strategies but also emphasized that the most important aspect of their plan was incorporating the use into therapists' clinical reasoning.³³ Deutcher et al described the benefits of integrating an electronic system with implementation of standardized measures. They concluded that such a system leads to the types of analysis described by Delitto and Bise in Chapter 19.³⁴

New documentation systems for physical therapy have been devised that build on these principles. For example, the American Physical Therapy Association (APTA) has created, in partnership with an outcomes measurement firm, a product known as APTA CONNECT. APTA CONNECT is designed to be a point-of-care, computerized patient record system. In addition to the ability to schedule and register patients, the system contains many standardized outcome tools. The data entered into APTA CONNECT are then loaded into a national outcomes database, which is used to better understand the effect of physical therapy care. In addition, APTA is expanding this national database to allow submission of data from all physical therapists, whether or not they are using APTA CONNECT. (www.apta.org/CONNECT/)

QUALITY IMPROVEMENT SYSTEMS

A full discussion of quality measurement and improvement in health care is beyond the scope of this book. Documentation has been linked to the concept of quality for as long as health-services researchers and clinicians have been discussing quality. Avedis Donabedian is considered one of the founding theorists in this area. He contended that there are three approaches to the evaluation of quality:

• Appraisal of structure, including physical structures, qualifications of heath-care personnel, and administrative organization. These structural elements are seen as necessary, but not sufficient to assure quality care.

• Appraisal of process, focused primarily on what the health care practitioner does. This approach relies on being able to identify a standard for acceptable, or even best, process.

• Appraisal of outcomes, evaluating the end results of care in terms of health and satisfaction. This approach is successful only if all parties can agree on what successful outcomes actually are and is based on a belief, preferably evidence, that the outcomes are actually a result of the care provided.³⁵

This model has been the basis of much of the work done about quality over the past 50 years. Evidence based practice clearly has roots in the model. Figure 18-1 shows the model as well as specific physical therapy examples in each category.

Donabedian pointed out that patient care documentation is an essential source of information for all three aspects of the model. He documented the prevailing issues of incompleteness, inaccuracy, and potential bias in the patient record, issues with which we continue to grapple today. Considerable efforts have been made at all levels of the health care enterprise (see side bar) to put in place quality assessment techniques designed for all three types of appraisal. Almost all of these techniques rely on good documentation.

Studies of implementing various means of improving documentation generally show some benefit. For example, Gumery et al (2000) reported the effect of chart audits done pre- and post-training on improving physical therapy documentation for patients in an adult cystic fibrosis unit in Birmingham, England. The purpose of the study was to assess adherence to professional standards of the Chartered Society of Physiotherapy (CSP) and the Association of Chartered Physiotherapists in Respiratory Care (ACPRC) by all physiotherapists working in the adult cystic fibrosis unit. The authors concluded that chart audits and training were useful tools in improving documentation and can lead to five key outcomes:

• increased knowledge and understanding of effective practice,

• increased effectiveness of service provision,

• improved practice,

• improved quality of patient care, and

• demonstration of commitment to evidence-based practice.³⁷

CONCLUSION

Efforts to use documentation as a means of improving the quality of health care have been under way for at least a century, yet documentation still does not always achieve the quality required for it to be readily used as an accurate record of care, as a basis for developing evidence from practice, and as a mechanism for measuring and improving quality. What can we do to change this situation? We appear to be at the point where electronic systems can provide all of the elements needed to make documentation a successful tool for quality assessment and improvement, including ease of use, access at the point of care, and ability to document the complexity of patient care. Zhou et al³⁸ bring us full circle with their work showing that simply implementing electronic systems is not sufficient. Success is linked to the system's ability to help the practitioner make better decisions. Chapter 19 describes just such a program in actual physical therapy practice. Other examples will be available at the F.A. Davis Web site for this text. We can also recognize the issues discussed in Chapter 17 to help reduce barriers to successful implementation. Perhaps the growth of point-of-care decision-support systems, as discussed in Chapter 16, offer the best possibility for the future.

SELF-ASSESSMENT

1. Review the documentation provided about Mr. Ketterman in the Appendix. Although it contains more than would typically be found in a clinical chart, review those elements that would be contained in a patient chart. What suggestions can you make to improve this documentation? What can you learn from the example to improve your own documentation?

2. Review three or four clinical charts. Do they meet the standards identified in this chapter? What suggestions would you make to the clinicians who documented in the charts to improve the record?

3. Identify the documentation system used in your clinical practice or in one with which you are familiar. Is there a system that you can identify? Does the system have the right elements to support good physical therapy documentation?

CONTINUED LEARNING

• The APTA has developed Defensible Documentation as a tool to help clinicians improve their documentation skills. Downloadable documents are available to members only at www.apta.org/Documentation/DefensibleDocumentation/, and an Internet-based course is available to all at learningcenter.apta.org/. Either of these tools has a great deal of information that can be helpful in improving documentation in a wide variety of settings.

• Identify the outcome measures used in your practice. Identify what is known about the reliability, validity, and generalizability of these measures for your given patient population. The book, Is Change Real? by Riddle and Stratford (Philadelphia, FA Davis, 2013) can be useful in determining whether you should adopt or change outcome measures in practice.

• Review the quality-assurance activities in your practice. Determine whether they measure the quality of the structure, process, or outcome of care. Do you have ideas on how to improve these activities?

REFERENCES

INTRODUCTION

In previous chapters of this book, you have had the opportunity to gain competence in the nuts and bolts of evidence-based practice (EBP). That is, you have used published research and other forms of evidence to increase your knowledge and skills in determining the diagnostic and prognostic value of tests and measures, evaluating clinical diagnostic strategies, making accurate prognoses, and deciding on the most efficacious treatment interventions for your patient. You are now ready for the next step in the EBP process: implementation of this newfound competence into your everyday practice. By implementing EBP, your increased competence and knowledge will potentially change your attitude about your approach to patients by not only considering what the latest evidence has to offer but actually realigning your behavior when appropriate, as measured by your own clinical performance. This EBP paradigm is illustrated in Figure 19-1.

Achieving Change in Performance

It becomes very important to distinguish between clinical competence and clinical performance in the evidence based process. We cannot assume that competence (e.g., knowledge of best practice principles and skill of administering) immediately translates to evidence based clinical performance, where the latter depends on aligning clinical behavior to be consistent with best practice principles.

The evidence is very clear that the translation of evidence into clinical performance is lagging in our health care system, a fact best illustrated by the Institute of Medicine's (IOM) characterization of the "Quality Chasm."¹ The extent of disparity between best practice standards and clinical performance is an unfortunate fact that transcends all health professions, including physical therapy. For example, in 1994 the Agency for Health Care Policy and Research (now known as the Agency for Health Care Research and Quality) published evidence based practice guidelines for managing acute low back pain, some of which were highly relevant to physical therapy.² Among the myriad of interventions for acute low back pain, two interventional procedures were assessed that fell well within the scope of practice for physical therapy: use of physical agents (e.g., heat, cold) and manual therapy. The guideline recommended against the use of physical agents and in favor of using manipulation for acute low back pain. Several years after this guideline was published, it is clear from the literature that physical therapists' attitudes toward these interventions and subsequent clinical behaviors (e.g., choices of treatment interventions) were clearly not aligned with the recommendations. In fact, evidence related to clinical performance demonstrates that overutilization of physical agents and underutilization of manual therapy is the rule in physical therapists' management of low back pain.²

Although many potential barriers to translation have been proposed, some researchers point to competency shortfalls as a primary reason for clinical care not aligning optimally with best practice standards.³ As the evidence based literature grows, clinicians are increasingly pressured to keep up-to-date and must negotiate and digest an ever increasing body of literature related to their patients and practice environments. We hope that the strategies and tools described in the preceding chapters will be helpful. In addition, we suggest strategies of lifelong learning, such as the judicious use of continuing education, can be used to stay current with contemporary knowledge and skills necessary to effectively treat patients.

An integral part of any lifelong learning plan is a candid assessment of your own performance. We are seeing competency assessment become increasingly common in graduate curriculums and ongoing assessment of advanced practice knowledge and skills in the clinic. Unfortunately, competency assessment alone does not ensure that clinical practice will be consistent with evidence-based, best practice standards.

In addition to staying up-to-date with competencies relevant to your patients and practice environments, perhaps the most important component of EBP is the implementation of newfound knowledge and skill in daily practice and review of your performance. Only by reviewing performance will you convince yourself that your patients are better off as a result of your efforts to implement the latest in EBP principles. The EBP paradigm (see Fig. 19-1) involves gaining new knowledge and skills from published literature through searching and evaluation that will lead to changing attitudes toward tests, measures, and interventions, which will eventually lead to changing clinical behavior (e.g., implementing tests, measures, and interventions based on evidence in lieu of those you may now be using). In order to assess performance with regard to adherence to EBP principles, you must measure it comprehensively.

The remaining portion of this chapter will focus on strategies for routinely evaluating your clinical performance with an emphasis on implementing best practice standards that are evidence based. In addition, we emphasize the importance of tracking patient outcomes.

Reviewing Performance: A Continuous Quality Improvement Approach

To gain competence in evaluating and managing patients using the best available evidence requires a framework by which you can critically appraise evidence and tools and that allows you to gather, collate, organize, and digest the ever increasing body of new evidence as it is published and disseminated. Previous chapters in this book have given you the tools with which to search, appraise, and summarize literature relevant to your practice environment. Now this information must be transformed into a measurable process of care for the patients that you see in everyday practice. The remaining portion of this chapter will attempt to show how quality improvement (QI) principles apply in the measurement of clinical performance.

A very simple old adage in QI literature is: you cannot manage what you do not measure. Performance assessment in any domain requires a comprehensive measurement approach. In clinical care, we attempt to measure clinical performance by routine use of chart audits, direct observation, and the use of specific clinical performance instruments. These approaches are largely unsuccessful because of unwarranted variability in the process of care and documentation of the same, both of which inhibit a comprehensive assessment of performance.

We have provided an overall framework for clinical performance appraisal and how it fits in the EBP process (see Fig. 19-1). More specific steps will allow you to collect and collate clinical data that is critical to performance appraisal. Ultimately, these steps lead to an accessible clinical database that becomes the tool used for performance appraisal.

Framework for QI-Based Performance Assessment

Figure 19-2 outlines four steps necessary for accurately evaluating clinical performance as it relates to implementing an evidence based approach to patient care in a clinical setting. The key component includes measuring clinical performance, with the goal of evaluating how well you are adhering to evidence based practice standards that you have set for yourself.

Step 1: Formulate the data fields for the clinical database that will be used for clinician performance assessment. This database will include patient information from the following categories:

1. Patient history (including pertinent medical history)

2. Tests and measures selected

3. Clinical decision rules used

4. Interventions selected

5. Patient outcomes

In planning and formulating the database, it is helpful to keep in mind the end goal, which is to be able to evaluate adherence to best practice standards and clinical performance. Several factors are worth discussion:

• Comprehensiveness versus exhaustiveness: Whenever a clinical database is formulated, the temptation arises to collect too much information. The more data to be collected, the greater the likelihood you will not have a complete data set on all patients because of the workload involved. The collection must be precise but extensive enough to be representative of the variables being assessed. Keep in mind that in a quality improvement initiative, the tolerance for missing data must be kept at a minimum.

• Environmental considerations: As you prepare to collect the data, you need to consider how environmental factors can affect both the data and the collection process. Data will need to be collected and entered into a database and then monitored. Poor adherence to any one of these processes can lead to incomplete data sets, which will compromise the QI effort. Barriers to complete data sets are numerous in a fast-paced clinical environment with limited resources. To ensure completed data sets, some system of surveillance must be incorporated. The methodology can include clerical staff or an automated information technology (IT) approach.

• Cultural considerations: Finally, you need to consider the culture in which you are collecting the data. As a profession, physical therapists have long relied on intuition and remain somewhat skeptical when presented with change. As you begin to collect and analyze the data, the culture around you may not embrace the process because it suggests the possibility of change.

Step 2: Collect data over time. Again, with the goal in mind, pick a defined period over which to collect the data. The duration should be of sufficient length to allow for an accurate snapshot of the behavior or practice patterns being analyzed. Keep in mind that the data collection must be comprehensive—it must include all data—and can sometimes be a tedious process. Careful attention must be given to the type of collection system used. A system that is labor-intensive might produce the most data but may begin to compete inappropriately with patients for the attention of the clinical staff. Collection that is too lean may not give the full picture of clinical performance. The optimal system will be quick and easy for clinicians to use and easy to read or input for those responsible for data input.

Once in the database, the data need to be accessible to you and your research (or QI) team and presented in a way that is easy to understand. Avoid complicated statistical computations. Try to present the facts in a format that will allow those involved with the process to evaluate the metrics being measured and then draw conclusions about performance. Businesses outside of health care have long used metrics to evaluate their performance. With the increased use of technology and the availability of real-time sales data, a "dashboard system" has become the norm in large companies. The "dashboard" presentation of the database makes the data easy to obtain, read, and analyze. You can then begin to make appropriate changes in clinical behavior based on this easily available record of performance.

Step 3: Evaluate clinical performance. After the data have been collected, collated, and presented in a useable form, you can begin to draw conclusions about clinical performance. The first component for analysis is compliance. How compliant was the individual practitioner with the data collection? Remember, data collection must be comprehensive or it doesn't reflect the true nature of performance. Second is adherence to the protocol. Our research has investigated the results of compliance with clinical decision making protocols. The data collected evaluated whether individual therapists correctly classified patients and then initiated the proper interventions regarding low back care. The final step was evaluating the outcomes achieved and their correlation with a set of predetermined benchmarks. Benchmarking is the process whereby individuals can evaluate their own personal performance against that of a group or a predetermined set of metrics. Our research specifically evaluated the downstream costs for low back care. Therapists who adhered to the protocol were compared with those who deviated from the established low back care guidelines. The established care guidelines were the benchmark. (See below for a fuller description of this research.)

Step 4: Structure an improvement plan. The final step in performance assessment is to create a plan for improvement. In the last step, compliance and adherence were evaluated. Problems in these areas could be addressed through the human resource department. Improvements in compliance allow an analysis of a comprehensive data set that more accurately reflects clinical practice.

After the first data set has been analyzed, adherence to specific treatment protocols can be ascertained. Changes in adherence can be measured over time and then compared to the baseline. The only remaining question is why clinicians do not adhere to the protocol. In our experience, most of the answers to this question can be broken down into two categories: nonadherence based on lack of knowledge and nonadherence based on lack of interest in or agreement with the QI process. The first is easily remedied: Educate the clinicians. The second, and more critical, will require an educational component but may also require the intervention of a third party. With national health care reform looming on the horizon, the third party may be the insurer or in some cases the consumer. Current research has focused on answering one question: What will it take to get clinicians to "do the right thing"?⁴ (Chapter 17 discussed this issue in more detail.) In the remainder of this chapter, we will give an example of QI research and how it makes the transition from instrument of measure to instrument of change.

Continuous Quality Improvement in Action

At the University of Pittsburgh, students are involved in a clinical performance final project. The project is intended to give students a tool for evaluating their clinical performance as they become licensed therapists. Early in the evidence based practice portion of the curriculum, students begin to select interventions and outcome measures for specific pathologies based on the strength of current evidence. The intervention and its outcome are then part of the data the student collects and then uses to populate the QI database fields. Data collection occurs during the first 6 months of the final year of clinical education. After 6 months, students analyze the data for adherence to best practice standards, compliance with data collection, and patient outcomes. Each student then identifies deficiencies, makes a plan for change, and begins the data collection process again. The final step of the project is a presentation of the initial and final data sets with a critical evaluation of clinical performance and ability to evaluate and change deficiencies over the clinical year. The result is an invaluable resource for evaluating personal change. You may say at this point that this is an academic assignment, required of students, and could never be executed in a clinical situation. McGee et al showed that not only can this system of performance be implemented in clinical practice but it can actually be used to decrease the total overall cost of delivery of physical therapy care.⁵

In 2007, a research group at the University of Pittsburgh identified difficulties with therapists' adherence to a treatment protocol for low back pain (LBP). The advantages of a treatment-based classification (TBC) approach were identified as far back as 1995 and culminated in a series of articles.⁶ Finally, a randomized controlled trial (RCT) published in 2006 offered undeniable evidence that a treatment-based classification system offered superior outcomes over other interventions in the treatment of LBP.⁷ Despite the evidence, care for LBP by physical therapists went relatively unchanged. The 2007 University of Pittsburgh research group speculated that adherence to TBC (Fig. 19-3) would provide better outcomes while reducing total overall cost of care.

The Pittsburgh study was an excellent example of a quality improvement initiative put into practice. Let's look back at the four steps in the quality improvement process and compare it to the study. The first step is formulation of the clinical database. The data used in the study was gleaned from a database collected by a large health system as part of its low back initiative. This initiative began as an attempt to identify the costs associated with low back care. The data collected provided all the markers needed to identify those physical therapists who were adherent to the TBC and those that were not, as well as the patient outcomes. Step 2 involves choosing a period of time for data collection. The researchers chose to collect the initial data over 12 months. This allowed for an appropriate number of patients to be admitted to the data set and provided an appropriate length of time for analyzing trends that might confound the data set. Failure of the therapist to collect or complete the minimum data set would confound further analysis.

Step 3 consists of data analysis. We have previously noted the importance of having complete minimum data sets (Fig. 19-4) for each patient. Early in the collection process the study coordinator noted an increase in the number of patients admitted to the study but no subsequent increase in the number of complete data sets. After the first 2 months, only 18% of the minimum data sets (MDS) had been completed. Lack of data severely compromises the ability to draw conclusions about performance. With the help of the HR department, a surveillance program was put in place to guide data collection.

This program began as a biweekly data report that identified each therapist and the number of complete vs. incomplete data sets. The facility directors were then e-mailed by the company QI officer, and the therapists who had missing data sets were asked to complete the MDS. If completion was not possible, the therapist was asked to write a short note explaining why. Through a series of accountability measures and in-service activities for clinicians, data collection slowly started to improve. Most clinicians responded to the biweekly verbal follow-up by the facility managers. As data collection improved, the frequency of verbal reminders decreased. Those clinicians who still had difficulty with collection were contacted by the directors and presented with a nonpunitive internal incentive. After 4 months of surveillance, the percentage of completed data sets rose to greater than 95%. With the increase in the number of data sets and the availability of data the study was ready to continue with the third stage of the QI process.

Analysis of the data was then completed. As Figure 19-2 shows, the three steps of evaluation of clinical performance are:

• Compliance: Is there a data set for every eligible patient?

• Adherence: Do interventions match prospectively determined clinical decision rules?

• Outcomes: Are outcomes achieving prospectively determined benchmarks?

Compliance was addressed in the previous section. Adherence and outcomes are the final two elements that need to be addressed. The purpose of the study was to analyze the adherence to a protocol. Figure 19-5 breaks down the totals by the treatment-based classification. Of the 750 patients enrolled, just over 50% were treated with a program adhering to the guidelines outlined by the TBC, a number that certainly leaves room for improvement. However, when the numbers were broken down into their respective categories, there were opportunities for improvement in adherence to the manual therapy and the stabilization protocols: 82% of the manual therapy and just over 50% of the stabilization patients were treated in a manner that was judged nonadherent to the TBC.

The final element is outcomes. During analysis, we need to consider patient outcomes, which will then give insight into our performance. During typical analysis, the first element to be considered is, "Are our patients achieving their predetermined benchmarks?" This question can be answered using the collected data, which will allow analysis of both patient outcomes and therapist performance. To continue following the example from previous sections, we note that even though patient outcomes are evaluated, therapist compliance with the treatment-based classification (TBC) system was the variable subjected to analysis in this particular activity. Since the researchers were observing adherence to the protocol, no specific benchmarks were set with regard to patient outcomes. Rather, the benchmark being evaluated was compliance with the TBC. Again referring to Figure 19-5, areas for improvement in compliance can be easily identified. Once identified, the areas for improvement would be evaluated based on set benchmarks by comparing data previously obtained with data obtained as Steps 2 and 3 are repeated.

Step 4, the final step of the performance improvement process, is the development of an improvement plan. Based on the data in the example, we can speculate on a possible plan. Earlier, low compliance with data collection was addressed by implementing a system-wide surveillance program supervised by the company QI officer. The response to the program resulted in 97% compliance in completion of the MDS. The more difficult problem may be therapist adherence to a specific treatment or protocol. Why is adherence low? Data analysis reveals poor adherence when thrust procedures are indicated. This may be explained by therapists' lack of familiarity with thrust procedures; some may not be comfortable executing these treatments based on education or knowledge. Other issues identified by therapists include the opinion that benchmarks or standards infringe upon clinical autonomy.

Though these responses need to be considered, the ethical implications of nonadherence must also be considered. The TBC has been shown to provide superior outcomes and better quality of care for those with LBP.⁸ If therapists are nonadherent with the classification system, are they providing the best care for patients? As the QI benchmarks are set, they will clearly address adherence to TBC. An increase in adherence is imperative for change to take place. If knowledge is the issue, further education of clinicians would augment the performance improvement process. If indifference is the issue, then a plan needs to be set in place to guide change over time. From here, we return to the clinic and start the process again.

CONCLUSION

QI may seem to be an academic debate right now, but health care is changing rapidly. The final analysis of the study described above evaluated the cost of delivery and additional downstream health care costs over the next year. During this period the authors showed a total direct health care savings in excess of $283,000 for those patients treated on protocol.⁵ Remember, over 50% of the patients were treated off protocol. Imagine if adherence increased a modest 10% to 15%, increasing total overall adherence to 65%. Based on this savings we could conclude that the cost savings alone would be attractive to payers while dramatically improving quality of care for the patients involved.

The future of health care will not be built on art; rather, it will be built on evidence and performance. We can either begin to evaluate our own performance and make changes or we can let others do it for us. The QI initiative we have described continues and we are at a steady state of greater than 95% compliance with completed data sets for eligible patients. We are assessing the effectiveness of an educational program on improving adherence to TBC guidelines as well as other educational initiatives, many of which are generated by feedback to the physical therapists on their performance related to the QI initiative. Either way, quality improvement is not the wave of the future; it's the reality of now.

SELF-ASSESSMENT

1. reflect on either a group of patients like Mr. Ketterman, a group of patients that make up a large portion of your practice, or a group you would like to treat regularly. Develop a database that contains the essential elements that would help you determine that the right care is being provided to this group of patients. As is shown in this actual case, these elements may include both the processes and the outcomes of care.

2. Using the same database, develop a minimal data set that could be used by you and your colleagues in your patient documentation to build an actual database for this group of patients.

CONTINUED LEARNING

• Examine the literature for one or two patient populations of your choice to identify therapist behaviors that you believe would result in improved care, both in process and in outcome. What benefit would you predict from implementing those behaviors?

REFERENCES

INTRODUCTION

We began this book with these words from Hippocrates, and they seem fitting for the book's conclusion as well. Over 2000 years ago, he identified that health care practitioners reach excellence by understanding the art of care, by recognizing the patient as a partner in care, by making accurate diagnoses and prognoses, and by knowing what helps (and harms) our patients—all the elements of evidence-based practice (EBP).

If these principles have been with us for 2000 years, one might ask why they have not been adopted fully into health care. We've discussed the barriers to individuals adopting change in Section IV, but it is also worthwhile to identify the reasons that reflective practitioners have opposed the adoption of EBP because of serious concerns about its appropriateness and the barriers that arise from the very structure of the health care system.

We will explore here the ways that EBP has been implemented by physical therapists across the country and will conclude with some thought about the possibilities for the future.

CONCERNS ABOUT ADOPTING EBP

Reasonable and reflective people have expressed some reservations about the very premises of EBP. Because these are such important issues, we will explore them here.

EBP Interferes With the Clinician's Decision Making

This concern about EBP is generally offered in one of two ways. Some complain that EBP is simply a tool of insurance companies to be used in preventing practitioners from providing the right care in order to decrease the amount spent on care.¹ While it is possible that an insurance company can misuse a particular piece of evidence to make incorrect decisions, these actions can be challenged and corrected. This possibility is far outweighed by the improvement in both quality and cost of care that can be achieved when practitioners use those tests and interventions that are shown to be more effective than others. This potential has led policy groups to recommend that the use of evidence be incorporated into health care practice. One such group is the Institute of Medicine (IOM), an independent group composed of scientists and practitioners and part of the National Academy of Sciences. They have examined the issues around EBP and have recommended that it is imperative that resources are made available to conduct systematic reviews and to develop trustworthy clinical guidelines. They state that such efforts will succeed in making health care decisions that:

• Constrain health care costs

• Reduce geographic variation in the use of health care services

• Improve quality

• Enhance consumer-directed health care²

Academy Health, a group designed to help move forward the principles and findings of health services research, has developed a briefing paper on how to use evidence to design health insurance benefits. They recommend developing better research to understand the effects of various coverage decisions. They also state that use of EBP to establish benefits should:

• Establish consistent expectations or standards for evaluating new treatments or technologies

• Move beyond the narrow scope of current efforts in evidence based insurance design

• Improve understanding of how different forms of incentives work

• Recognize implications of insurance coverage for health and economic equity³

Think tanks such as Academy Health propose a deliberative process for incorporating evidence into health insurance coverage policies, which will serve practitioners and consumers more successfully than the hit-and-miss technique followed by some individual insurers. Rapid fluctuations in covered services can have a negative effect on both quality and consumer confidence.

A second complaint about decision making is that EBP means "cookbook" care. This complaint claims that EBP forces clinicians to use certain tests and measures, regardless of the individual patient needs, thereby depriving both the clinician and the patient of the benefits of the clinician's clinical wisdom and of personalized care.⁵ We hope that this text helps to dispel at least a portion of this complaint. We have demonstrated that EBP actually means a reliance on good decision making by the practitioner. Only an astute practitioner using good decision making processes can make the decision that a particular piece of evidence actually applies to a particular patient. Another way to look at this complaint is from the perspective of unwarranted variation.⁶ Variation in care can occur because of patient preference or circumstances but should not occur because practitioners have not chosen the most effective care. When the variations occur for no discernible reasons or for reasons related to poor practitioner knowledge or decision making or to financial gain, then those variations are unwarranted and the variance should be reduced. Adopting well-designed clinical guidelines is one way to reduce unwarranted variation.

EBP Is Unethical

This concern is a very important one to consider and address, as we owe our patients the highest possible ethical behavior. Often these concerns revolve around the injustice of applying a particular intervention to a group of patients without regard to the clinician's own judgment or the patient's own preferences. There are also concerns about adopting too rigid a definition of what acceptable evidence is, relying too heavily on the need for randomized controlled trials. There are additional concerns about the ability of even the best research to fully identify some of the more complex outcomes for care, such as reduction of pain or quality of life, and therefore lead clinicians to make decisions based on more easily measured but less relevant outcomes.^7,8

These concerns represent a view of EBP as a simple application of research results with no filter for their applicability to a particular case. A clear understanding of the need to integrate evidence with a clinician's expertise and a patient's preferences should relieve much of these concerns. They are also balanced by the perspective that ignoring what the evidence tells us may also be unethical. For example the Code of Ethics for physical therapists, promulgated by the American Physical Therapy Association, states, "Physical therapists shall demonstrate professional judgment informed by professional standards, evidence (including current literature and established best practice), practitioner experience, and patient/client values."¹⁰

One might argue that any practice that does not serve the patient optimally could be considered unethical practice, but such a definition excludes the motivation of the practitioner from the judgment. Clearly, a practitioner who searches diligently for the intervention that facilitates the best outcomes is practicing ethically, even if the outcomes fall short of the patient's goals. It is the practitioner's value for openness to innovations in health care that ensures ethical practice.

Evidence Is Not Always Based on Theory

Theory is focused on the reasons why a phenomenon occurs. In health care, theory attempts to explain the causes of illness and disability and the reasons why particular care meets with success. Theory is often the basis for postulating the potential of new approaches to patient care management. It is rooted in the sciences that support health care. For physical therapy, those sciences include anatomy, physiology, neuroscience, psychology, sociology, ethics, and teaching and learning.¹¹ In an early editorial on the issue of theory in physical therapy, the authors state, "Theory is a general, abstract body of interrelated principles, concepts, and constructs that presents a systematic, scientifically acceptable view of phenomena. Theory is conjecture that is inferred from a set of logical propositions that, in turn, are based on empirically derived evidence. Acceptable scientific theory is internally consistent, empirically testable, parsimonious, and congruous with existing knowledge. Theory also should be important scientifically."¹²

It is important to note that the word theory implies that there is an element of doubt about the accuracy of the theory. In fact, in the physical sciences, when sufficient doubt has been removed the word theory is often replaced with the word law. Evidence, on the other hand, is actually defined as that which tends to prove something. We have two terms that may be at odds with each other: theory—a not yet proved description of a phenomenon, and evidence—proof.

This tension between theory and evidence is well documented. There are numerous examples of interventions designed and used because they "make sense" from a theoretical perspective but are later shown to be ineffective or even harmful when tested in clinical trials.^14,15 An opposing view states,

One of the central concerns of these critics is the reliance of much of EBP on the randomized clinical trial (RCT). Critics believe RCTs are both too limiting for studying all the many phenomena that make up health care (see Chapters 8 and 12) and can overinflate the importance of results, particularly in studies with small numbers of subjects and small effect sizes.^16,18

Scholars have pointed out that some of the evidence provided about interventions in physical therapy has been done in an atheoretical context, which might account for the rather small, albeit positive, effects seen in this research. For example, recent work on classification of patients with low back pain to match them to selected interventions has explicitly stated that this work is based on empirical observation, without regard for theoretical explanations of why these approaches might be successful. Researchers contend that years of seeking the mechanism of injury as a means of guiding care have not resulted in success and that this empirical approach helps patients improve. Critics of this work contend that we must do a better job of identifying the mechanisms of injury that cause the back pain if we are want to achieve the necessary level of success in identifying appropriate care.^19,20,21 Others have taken a more philosophical approach to their critique, noting that wholesale adoption of any single way of looking at the world leads to a loss of possibility, growth, and creativity.²²

But rather than seeing only the tension between the two terms, they should also be seen as enlightening each other. In the best of all worlds, wouldn't a theory about how and why something works help drive the empirical research to show if it does, and wouldn't identifying that something does work help develop the theory as to why? Development of theory and production of evidence about our care should be twin goals, as together they can lead to a sound body of knowledge that matters to our patients. This really represents the inevitable link between research and practice.

Eugene Michels, one of the leaders in supporting and developing research in physical therapy, wrote about this topic often. In one letter to the editor of Physical Therapy he noted the deep need to develop theory in physical therapy, saying,

In a letter to the editor a year previously, he commented, "If we are to develop our own science for understanding and doing socially useful things about human movement dysfunction, that science must first be well grounded in the definition, identification, and study of relevant, observable phenomena and variables and later explained and elaborated by appeal to more abstract and organic levels of reduction. The robust sciences work forth and back between the nonabstract and the abstract levels of reduction."^{24,p1138,1139} This is all advice we would do well to heed today.

EBP, QUALITY, AND COSTS

The adoption of some specific recommendations from the evidence may actually increase the costs of care. But the hope is that widespread adoption of changes identified by trustworthy evidence will result in reduced costs of care. As we discussed in Chapter 9, the growing costs of health care in this country arouse concern about our ability to maintain those costs, about inequities in access to equivalent care across the U.S. population, and a recognition that those costs do not seem to be buying a commensurate improvement in overall quality.

Gawende has identified the wide disparity in costs in specific small regions, without any positive relationship between costs and quality. He has advocated for the need to compare the effectiveness of both the specific care provided and the systems designed for the care delivery, supporting the concept that more systematic and centralized responsibility for care can help reduce costs and at least maintain, if not improve quality.²⁶

The IOM has recommended that emphasis be placed on comparative effectiveness research, which can result in trustworthy clinical practice guidelines, selecting from among the array of diagnostic and prognostic tests and measures and interventions to demonstrate the best options for the entire plan of care for a patient (see Chapter 16).^2,27,28 This would meet at least part of the recommendations needed to ensure that a cost-quality balance is reached. A class of studies we have not explored, but which is important, are cost-effectiveness analyses. This line of research is designed to identify which aspects of care are able to produce the same outcomes with lower costs. Of course, the goal would be to only do cost-effectiveness studies once the actual effectiveness of various options has been identified.² Work has begun to more clearly identify the cost-effectiveness of physical therapy.²⁹ These types of studies will also contribute to achieving the cost-quality balance.

An interesting example of research designed to identify the impact of practice focused on controlling costs is a study by Milstein and Gilbertson. They identified several practices that were able, through a variety of tactics, to provide quality care at 20% or less cost than other practices, with no reduction in quality. If the 20% reduction in costs identified in this study can be extrapolated to the entire country, it would represent a saving of $640 billion, as of 2011.³⁰

In Chapter 17 we discussed many of the barriers for individual clinicians and individual facilities to making changes in their practices based on the evidence. But there are also barriers that are intrinsic to the health care system in the United States that interfere with adoption of the best practices identified through the evidence. (See Chapter 9 for a discussion of the complexity of the health care system in the United States.) Why doesn't the health care system adopt these changes and realize these savings? Fuchs and Milstein³¹ state that the prime reason is that the key players in the system each have their own incentives to not change. Here are some examples of the stakeholders they identify and reasons that inhibit change:

• Large insurance companies do not want to adopt standardized national benefits because that could decrease their competitive advantage in local settings, thereby decreasing profits for their shareholders.

• Large employers do not want to encourage competition among health plans so as to not alienate employees; small employers can't afford more than one plan; employers in general mistrust government solutions and recommendations that seem to restrict care. All of this means that the competition among plans based on cost reduction cannot occur.

• The general public resists change, not recognizing that health insurance benefits are actually a form of income and that carefully reducing costs could actually increase their discretionary income. In addition, the general public often has little ability to really understand the issue of risk analysis and the benefits that could arise from certain changes.

• Legislators at both the state and federal levels are influenced by the health care system lobbying effort, which can be more focused and, therefore, more successful than a diffuse effort by individuals.

• Hospital managers fear that reducing specific procedures could reduce their ability to cover the large fixed costs they have for buildings and equipment.

• Drug companies and medical equipment manufacturers believe that their best success is in developing unique products whose prices they can control and, therefore, resist attempts to standardize care.

• Health care practitioners resist change for all the reasons that we discussed in Chapter 17 and for the concerns we have discussed above.

These perspectives are primarily economic in nature and may not be the perspectives that individual health practitioners typically have, but they do describe well the difficulty in achieving system-wide changes. As health care practitioners, we are obligated to place the needs of patients first. Shouldn't we be the leaders in helping the system come together to bring about necessary and appropriate changes?^2,27

EBP IN USE IN PHYSICAL THERAPY

While preparing this text, we asked our colleagues, through multiple listservs, for examples of the use of EBP, particularly the integration of the three elements of clinical judgment, respecting patient values, and using evidence from the literature. We received dozens of examples in a few short days.

The examples ranged from relatively small efforts, such as setting up a regular journal club using a Critically Appraised Topics (CAT) model, to developing comprehensive evidence based guidelines for use in a large facility. We will highlight just a few of these ideas here:

• Several facilities used data from their own patients to determine which particular populations were at risk for falls. They then used the literature to build better screening for fall risk, assessment of the causes of falling, and better fall prevention programs for these populations.

• Several therapists reported using specific information from the literature to explain to patients or physician colleagues the rationale for specific interventions, with increased acceptance of these changes in practice.

• The inpatient physical therapy team at one large facility holds case conferences twice a month. The first meeting is a discussion of a patient case. One week prior to that meeting, a team member provides the group with the history and tests and measures from the evaluation of one of his or her patients. The team members read this information, write their own clinical impression and plan of care, and come prepared to discuss their thoughts. In that session the group discusses differences taken by each team member. They compare evaluation statements and plans and talk about differences. At the end of the meeting (about 1 hour), they identify a clinical question that could be answered through a search of the literature. Each member of the group searches the literature, and the team leader then chooses one article in particular to use as a point of discussion at the second session. After the discussion of the article, the group asks, "Would we do something different if we saw him/her again or another, similar patient in the future?" After the second meeting, the team leader writes up a summary of the month's work and shares it electronically so all staff can access it. The team leader says, "This process has been successful and very well received by our staff. They frequently read past write-ups and those from other teams when they have a question about a patient."

• One therapist described the need to develop a specific program for care of patients with osteoporotic fractures in advance of the research evidence and base it on the known information from the study of anatomy and human movement, modifying the program as evidence became available.

• Another therapist was the first to introduce evidence based assessment tools to a group of therapists in Kenya.

• A facility developed a best practice protocol for the treatment of idiopathic toe walking. It is a treatment algorithm based on current research and clinical practice. The protocol is presented to the family with best treatment options explained based on the child's presentation. The family is then given the opportunity for input, with the goal being to educate families on current research and improved clinical outcomes as they make their decisions.

• One large inpatient facility has a practice committee. The committee has developed protocols, chronological care patterns for major patient populations seen, and standards of care documents, which include the natural history of the problem, details for the physical therapist's examination and intervention options, and copies of recommended tools and references that support the choices. What is most remarkable about this is that the committee has developed over 40 protocols and 70 standards of care. And it has made all of them publicly available at the facility's Web site, to be used by not only the therapists but patients and families and health care colleagues.

• A facility developed the position of clinical research educator (CRE), who spends 50% of his or her time on evidence-related activities. The CRE functions to identify and assess published evidence, to teach colleagues to search and appraise the literature themselves, and to develop practice-Specific evidence by collecting and analyzing data derived from their own practice.

• One facility asked the therapists to develop PICO questions based on their clinical experience and then had students participating in clinical education conduct the searches and analysis of evidence, thereby building on the strengths of each group.

• In addition to that example, several academic faculty offered examples of collaboration between the educational program and related clinical programs, where faculty offered their expertise on search and assessment and clinicians identified questions of clinical importance and their judgment in the application of the evidence to particular populations and patients.

As can be seen from these examples, representing all aspects of practice and all parts of the country, clinicians can find the resources to develop and utilize EBP skills.

WHAT DOES THE FUTURE HOLD?

We have emphasized the thinking processes used by clinicians in Section I. Understanding decision making is a complex science. Bell and Raiffa have identified a taxonomy used with great frequency by decision analysts, those scientists who study decision making, to clarify some of the approaches used in studying this topic. They have identified three different processes used in thinking about decision making:

• Descriptive

  1. Decisions people make

  2. How people decide

• Normative

  1. Logically consistent decision procedures

  2. How people should decide

• Prescriptive

  1. How to help people make good decisions

  2. How to train people to make better decisions.^32,p1,2

We have discussed two sources of normative information, patient values and circumstances, in Section II, and the evidence from the literature about our tests, measures, and interventions in Section III. We have discussed descriptive information, what we learn about how clinicians actually think, in Section I. We have offered some ideas on how to improve our decision making in Section IV. But we will need to continue to test these methods, and others, as they are developed to see if they produce the changes that are desired. We hope that we will see continued growth in the study of decision making in physical therapy.

As we have discussed throughout this text, evidence based practice holds much promise to improve the care that we provide our patients. The fulfillment of this promise rests in the full integration of all three elements of EBP. Information from research is useless if thinking clinicians do not apply this information in collaboration with their patients. There continues to be growth in advocacy for the patient to be recognized as the center of our care. The federal government, as well as multiple foundations, has supported this perspective.^33,34,35,36

At an IOM conference, placing the patient at the center was described as a Copernican shift.³⁷ Copernicus forced us to completely rethink our understanding of the universe. Will this focus on patients really help us change our care as completely as placing the sun in the center of the universe has done? In physical therapy, we have prided ourselves on engaging patients, because we know that our care is not passive. If the patient does not participate, then we cannot achieve reduction of impairments or improvement in function. But, have we always placed the patient really in the center? Do we collaborate in goal setting? Do we respect our patients even when they reject our advice? Have we found ways to help patients understand the evidence and judgment that (should) drive our recommendations? We hope to see continued growth in these behaviors by physical therapists and the research that demonstrates the benefits of patient-centered care.

CONCLUSION

The single best practice physical therapists can follow to improve their decision making is to reflect on decisions, individually or with colleagues or learners. This commitment to reflection in practice will never fail to provide the practitioner with more questions to pursue, strengthened values for excellence, and a stronger relationship with the patient.

Epstein, in The Mindful Practitioner, tells us, "Exemplary physicians seem to have a capacity for critical self-reflection that pervades all aspects of practice, including being present with the patient, solving problems, eliciting and transmitting information, making evidence based decisions, performing technical skills, and defining our own values."^38,p833 What a wonderful description of the behavior we all aspire to as clinicians who have fully embraced all aspects of evidence based practice!

REFERENCES