Attribute dimensions that distinguish master and novice therapy clinicians in orthopedic settings.

There is a long tradition of study of the phenomenon of expertise within professional groups such as teachers, nurses, and physicians. (1-3) More recently, expertise has been studied in physical therapy and occupational therapy. (4-6) Much of this research focuses on what the expert practitioner knows and how this knowledge is related to the efficacy of action or clinical decisions. These studies focusing on knowledge of the expert practitioner are frequently grounded in theories from cognitive psychology and include a broad range of investigations from the composition of expert knowledge (1,7) to clinical reasoning and the decision-making process. (2-5,8)

An essential consideration in investigating the work of professionals is that skillful action is adapted to its context; that is, through repeated practice and reflection on practice, the professional develops relevant specialized skills. (9,10) Schon (10) uses the term "knowledge-in-action" to describe the knowledge that is embedded in the skilled action of the professional. The challenge for researchers is to "get inside the heads" of practitioners in order to see the world as they see it and understand the manner in which professionals think about, construct, and solve clinical problems. (11,12) We know that experience is a necessary, but certainly not sufficient, condition for expertise. In addition, we still know little about the process of how expertise is acquired, even though much research has been done on expert-novice differences, particularly differences in problem solving or reasoning. (12)

Clinical Problem Solving

Clinical problem solving has been the focus of several investigations concerning physicians' performance. (3,7,8) In this research, expert and novice physicians' differences were attributed to the early generation of hypotheses and a testing of the hypotheses until a fit was found with cues from the clinical data and disease presentation. The use of this hypothetico-deductive strategy, however, did not distinguish successful from unsuccessful clinical problem solving. (3) The differences were found primarily in experts' recall of meaningful relationships and patterns, that is, in the structure of the knowledge rather than in a problem-solving strategy applied to the problem. In a descriptive study analyzing the clinical reasoning process of 10 expert physical therapy clinicians, Payton (13) found that therapists use a process similar to the hypothetico-deductive reasoning process of physicians.

Researchers also claim that problem-solving expertise is case specific and highly dependent on the clinician's mastery of a particular content domain. (3,7) May and Dennis (5) used a questionnaire to gather data on cognitive styles from 400 American and 384 Australian physical therapists who were considered by their peers to be expert clinicians. From their findings, they suggest that clinicians may use different cognitive processing styles for different clinical problems. For example, therapists working with patients with orthopedic disorders reported more frequent use of a receptive (suspending judgment until all possible data have been collected) data-gathering style and a systematic (performing an ordered search for information) information-processing style. Alternatively, therapists working with patients with neurological disorders reported more frequent use of a perceptive data-gathering style (seeking and responding to cues and patterns as a guide to data gathering) and an intuitive information-processing style (keeping the total problem in mind and considering alternatives simultaneously).

Research on expertise in the field of education, like practice within health care professions, has become a major focus of inquiry in recent years. Current investigations on the master, or expert, teacher have broadened the problem-solving, decision-making analysis to include a fuller understanding of the knowledge practitioners require in the context of their performance. This focus on context has resulted in a shift from more experimental, process-product research in classrooms to more qualitative investigations that describe not only what teachers do or decide but how they understand and perceive what they do. (11,12,14)

Work of the Physical Therapist

The work of the physical therapist, like the work of other professionals, is a complex process that can be conceptualized and investigated in many ways using different conceptual and theoretical frameworks. Our initial research was a qualitative investigation of the work of the physical therapist in the natural practice environment. (15) Eight physical therapists, representing three levels of experience, were observed in four different adult outpatient orthopedic settings. A conceptual framework was generated that provided a preliminary structure (16) for studying the physical therapy practice environment (Fig. 1). The primary focus of this research was investigation of the therapeutic intervention. We argue that the therapeutic intervention is the filter through which all other factors related to patient outcome in physical therapy practice are processed. These factors include characteristics of the therapist, the client, and the organizational setting. For the purposes of our research, we defined therapeutic intervention as the strong, one-to-one relationship between client and therapist, centered around the identification and rehabilitation of the client's movement dysfunction problems. Five themes, which represented differences in the behaviors of experienced and less experienced clinicians, emerged from the analysis of these data (Tab. 1).

[FIGURE 1 OMITTED]

The purpose of this qualitative case study was to further investigate the work of the physical therapists with a small number of expert, or master, clinicians (ie, experienced and proficient therapists) and novice clinicians using the previously identified conceptual framework and initial five themes as a basis for elaboration and revision.

Method/Research Design

Conceptual Framework

A qualitative case-study approach (17) was used to collect the data because our intentions were (1) to investigate clinicians at work, in context, in the real-world setting--the clinic--and (2) to elaborate and revise the conceptual (theoretical) framework about the work of physical therapists (Fig. 1). (15) The focus of our observations, both in the prior study as well as in this study, was on the complex and unknown processes that occurred during the therapeutic intervention. In this study, we were particularly interested in determining whether the five themes from the original conceptual framework, which distinguished between the therapeutic intervention of less experienced and more experienced therapists and their patients, were the same for novice and identified master clinicians (Tab. 1). We prefer to use the term "master clinician" in lieu of "expert" to represent a therapist who has expertise as well as proficiency of skill and artistry.

Throughout the course of this research, we reflected on and discussed all the components and relationships among those components that were part of the initial conceptual framework, particularly the therapeutic intervention component. As a result of these data, a refined and reorganized conceptual framework will be presented in the "Discussion" section.

Sample

The sample consisted of three master clinicians and three novice clinicians, all of whom practiced in orthopedic outpatient physical therapy settings. We chose to limit our sample to therapists engaged in orthopedic practice to control, in part, for client characteristics and the type of physical therapy techniques used (Fig. 1). The selection process for master clinicians was panel nomination on the basis of reputation. (18) Clinicians were identified by nominations obtained from Academic Coordinators of Clinical Education (ACCEs) (N=12) in three different regions of the United States. Eight ACCEs were from the Northeast, 2 from the Mid-Atlantic, and 2 from the Southeast. The ACCEs were asked to identify up to five physical therapists who, on the basis of experience and expertise, they considered to be master clinicians in the orthopedic area. Therapists nominated by more than 1 ACCE were then contacted to determine whether they would agree to participate in this study. The ACCEs were also asked to identify two novices, that is, recent graduates who had worked for less than 1 year in orthopedic outpatient settings. Novice clinicians were contacted according to geographic convenience to the researchers.

We used a nonprobability sampling strategy called "purposive sampling" or "criterion-based sampling." (17) Purposive sampling is based on the assumption that certain characteristics of standards of the sample are central units in the investigation. One selects a sample that fits these identified standards. In our sample, the characteristics of experiences and expertise were identified as criteria for the selection of novice and master clinicians. We identified these master clinicians because of their experience and identified competence; therefore, the clinical status of the subjects (ie, whether they were expert or novice clinicians) was known to the researchers.

Our master clinicians were three therapists, two with 13 years' experience and a third with 23 years' experience. One therapist held an advanced master's degree in physical therapy, and the other two therapists held baccalaureate degrees. All three therapists were involved in clinical teaching as well as teaching clinical content in one or more academic settings. Two of the therapists conducted frequent local and state continuing education sessions. These master clinicians all practiced in outpatient orthopedic clinics in institutional settings, two at university medical centers and the third in a rural community hospital.

The novice clinicians were recent graduates of physical therapy programs, two at the baccalaureate level and one at the master's degree level. The orthopedic clinical settings for these therapists consisted of a private practice, an outpatient clinic in a community hospital, and an outpatient clinic in a university medical center. All clinicians and all patients observed gave informed consent prior to their participation in the study.

Data Collection

Data were collected in three different regions of the country--the Northeast, the Mid-Atlantic, and the Southeast. One of three researchers (GMJ, KFS, JG) collected data on a master clinician and a novice clinician at each of the three sites. Data collection consisted of on-site observation of each clinician with at least three patients, audiotaped treatment sessions, patient interviews, interviews with clinicians regarding perceptions of their own decision-making and clinical skills, and reviews of patient records.

Following observation of treatment sessions, patients were briefly interviewed to determine whether their perceptions of their clinical problems (type and course of treatment and expected outcomes) (Appendix 1) were similar to the perceptions held by their therapists. We then interviewed the therapists to find out more about their decision-making process and to gain further insight into their own self-assessment (Appendix 2).

Data Analysis

The first step in the analysis process was transcription of field data gathered through observations and audio-taping of the clinical interaction between therapist and patient. These transcriptions were then coded using our previously developed coding scheme (Appendix 3). (15) The researcher responsible for the data collection then wrote individual case descriptions for each master and novice clinician she observed. In our previous work on novices and experienced clinicians, we generated five themes that defined the major characteristics on which the therapeutic intervention of experienced and novice clinicians differed (Tab. 1).

Analysis of the data from this study (observational and audiotape) in the context of these five themes was a central part of each case report. The case report also included the following descriptions: (1) the clinical setting (observational data), (2) the demographic background of the therapist (therapist interview data), (3) the patient's perceptions about his or her physical therapy (patient interview data), and (4) the therapist's approach to clinical decision making (therapist interview, observational and patient-record data).

The initial analysis of the data included a review of the original data by all four researchers (ie, review of the transcripts of the three therapist/ patient interactions, the therapist interviews, and the patient interviews). This review allowed each researcher to analyze the original data and identify the confirming and disconfirming evidence related to the original five themes. The six case reports for the three master and the three novice clinicians were then examined by all four researchers and contrasted with their individual analyses of the original data. The analyses of these case reports were first performed individually, then refined in a data-analysis meeting of the four researchers. Analysis at this level demonstrated that revision of our initial five themes was needed. Erickson (19) calls this process of data analysis "the generating and testing of assertions." We found that our initial themes (assertions) did not fit conceptually with our present data. We reanalyzed the data by additional categorization, comparison, and interpretation of the case data. Conflicting interpretations of the case data were discussed and supported with verbatim transcript data until consensus was reached. This process resulted in the development of five revised themes, which we called "attribute dimensions." These attribute dimensions appeared to distinguish between the therapeutic interventions of the master and novice clinicians (Tab. 2). These attributes were the new assertions that were used to revise our individual master and novice case reports into case studies. The case studies were constructed using a consistent descriptive framework. (20) This framework included the descriptive material from the case reports as well as discussion of each of the five attributes supported by evidence from the field data (observations, interviews, and artifacts).

A cross-case comparison of the six case studies was then performed, first across all novice clinician cases (n=3), then across all master clinician cases (n=3), and finally across all six clinician cases. The analysis strategy we used was described by Yin (20) as a combination of pattern matching and explanation building. One researcher (GMJ), experienced in cross--case study analysis, (21) then performed an initial analysis of the cases (ie, the three novice cases followed by the three master cases). This analysis consisted of first looking at the patterns, that is, the supporting data for each of the five attributes, across the cases. This examination revealed consistency within the novice and master cases. Next, a cross-case comparison was done using an explanation-building process that examined the contrast between novice and master cases. This process involved comparing the data in the cases with critical insights from research and related theory. (1-3,11,14) This explanation-building analysis resulted in our division of the attributes into one knowledge attribute and four attributes related to improvisational performance. The process of pattern matching and explanation building was then repeated using the same sequence (novice cases, master cases, all cases) when the researchers met as a group. Conflicting interpretations of the case data were again discussed until consensus was reached. Changes made were supported with evidence from the field data.

Internal reliability in the study was addressed by having the researchers use the same analysis and review process for identification and interpretation of general constructs (ie, our attribute dimensions). (22) In qualitative research, agreement is sought not on the frequency of events, but on the description and composition of the events. (16,17,22) We used several strategies designed to address internal validity from a qualitative perspective. Nonparticipant observation was used to collect data in the natural setting (the clinic) to represent the reality of the experiences of the participants (therapists and patients). Informant interviews were used to provide therapists and patients with the opportunity to talk about and respond to what we had observed. Verbatim data were collected through audiotaping and subsequent transcriptions, which allowed us to use a primary, low-inference source of data in support of our assertions. We used multiple sources for data collection, including observation, interviews, and review of physical evidence. This use of multiple data sources, or methodological triangulation, provided support for our attribute dimensions. The use of multiple researchers also helped ensure internal validity by providing a process of corroboration and confirmation through pooled judgment versus the judgment of one. (17,22)

Our "Results" and "Discussion" sections will focus on documenting and discussing the differences we found through our analysis process between the novice and master clinicians.

Results--Patterns in Master and Novice Cases

Ability to Control the Environment

We found that the master clinicians were able to fully control the treatment session and make efficient use of time, yet maintain an intense focus on the patient, regardless of a busy clinical setting. Two of the master clinicians practiced in busy clinics, yet observation of their treatment sessions revealed a calm atmosphere in which interruptions were integrated without problems. For example, in one case the therapist was evaluating a patient in the midst of a busy gym, yet was able to keep the patient's attention focused on her evaluation. In another case, the master clinician smoothly integrated responses to interruptions, as demonstrated here.

   Patient 1: You want it out (of the
   splint), right? Master Clinician 3 (MC-3):
   Yep. (Turns to aide, who has
   brought the mail in, riffles through it,
   and asks,) Any money in there? (Patients
   all laugh; MC-3 turns immediately
   back to patient and says,) Okay,
   let's see how much you can straighten.
   If I hurt you, holler. [MC-3:3]

In the third case, the master clinician worked in a hospital-based outpatient clinic but had control over the logistical operation of the clinic. He controlled scheduling and spent an average of 48 minutes of uninterrupted time with each of the patients observed.

Two of the novice clinicians, one employed in a private practice and the other in a hospital outpatient setting, were in situations, similar to MC-3, in which they were simultaneously treating two or three patients. Unlike the master clinicians, the novices did not make efficient use of their time or take steps to control their environments. The novices appeared to be juggling time and personnel, making some treatment decisions based on "buying more time" for themselves as they tried to keep up with their schedule. One novice clinician (NC-1) spent an average of 20 minutes with each patient and had to cope with an average of four interruptions per patient. Having patients independently review their exercise programs was her method of moving on to the next patient. Another novice therapist reported her growth over the past year in terms of increasing her efficiency in a hectic clinical environment.

   NC-3: I have learned to be a little more
   efficient. My time-management skills
   are sharper. I would get caught in a rut
   where I would let people talk and talk
   and talk and talk, and I would be afraid
   to cut them off. Now I know how to do
   that a little bit better. I know how to
   direct them out of the conversation so
   I can leave the booth and go do
   something else. [NC-3:2]

Evaluation and Use of Patient Illness and Disease Data

The master clinicians consistently used an evaluation framework that included the patient's history and physical examination. They spent their initial time with patients filling in that framework and were able to deviate from the framework when it was necessary to probe deeper to gather more detailed, patient-focused information. The master clinicians not only collected disease data (ie, findings that help validate or invalidate a diagnosis), but they also gathered illness data (ie, data that reflected the patients' perception of how the disease affected their lives). (23)

The master clinicians placed emphasis on documentation of items such as range of motion, muscle testing, and reflex status, and they attempted to tie the data directly to the patient's needs. As one clinician stated when asked how he knows whether he has been effective,

   MC-3: The numbers ... The numbers
   tell me they're better. I mean their grip
   strength is better, their job performance
   is better. It's got to be objective.
   [MC-3:2]

This commitment to documentation is coupled with a strong tie between the physical examination data and information gathered about the patients' perception of their condition. For example, these clinicians described "using their (the patients') history a lot. Listening to what the patient tells you." [MC-2:5] and "if the patients do not perceive they have been helped, they have not been helped." [MC-1:3]

Master clinicians also reported designing their exercise interventions not only to fit with their evaluation data, but to fit the patient's environment as well.

   MC-1: I try to think of the patients in
   the environment, about their work and
   recreational activities. Then, I go on
   and think about structural and
   functional considerations. [MC-1:5]

Like the master clinicians, two of the novice clinicians also gathered subjective and objective data. These novices placed heavy emphasis on attempting to get information by implementing standard evaluation routines. As one novice therapist described,

   NC-1: I have a standard routine. Observation,
   palpation, posture, gait, manual
   muscle testing, range of motion, and
   any special test for each joint. Then, of
   course, some functional activities ... I'm
   constantly taking objective data.
   Maybe too much of it. I don't know.
   Right now, I have to do it, and I'm not
   comfortable knowing what will
   happen. [NC-1:4]

The third novice clinician reported gleaning information wherever she could find it: "So I would say my subjective and my objective [evaluations], what I can get from the doctors--and I use all my resources from my colleagues here." [NC-3:3] For the novice clinicians, filling in the evaluation framework was a primary goal. They did not take opportunities to deviate from their evaluation frameworks and pursue patient perspectives.

Focus of Verbal and Nonverbal Communication with the Patient

All three master clinicians demonstrated an intense, focused connection with patients. Two of these therapists were observed working in one-to-one settings, and the third therapist was observed working with a group of patients. The researchers' descriptions of these master clinicians are as follows:

   MC-2's attention to the patient during
   these evaluation sessions is intense.
   She directs her questions in a pattern
   to track down the source of the patient's
   problems. Her body language
   shows complete attention to the patient,
   from sitting on a stool directly at
   the feet of the patient to take the history
   to keeping contact with the patient
   physically throughout the examination.
   [MC-2:4]

   Patients appear to be at ease (social
   interchange, laughter, exchange of
   information about their physical status)
   working at their tasks close together
   under the watchful eye of MC-3. His
   eyes rove over the patient group constantly,
   assessing/monitoring their
   progress. In an individual interview,
   when patient 2 was asked why MC-3
   was an effective therapist, she replied,
   "He is very dedicated to his work. He
   goes from one patient to another, and
   that is not a problem.... He listens
   well. Everything is important." [MC-3:1]

We observed the novice clinicians using a variety of verbal communication patterns with patients. One therapist's focus on trying to fill out an evaluation dominated her interactions with patients. At times, her questions were long and involved, but she paid little attention to the answers. Another novice clinician communicated a definite sense of warmth, caring, and enthusiasm, as verified by patient comments such as these:

   Patient 1: She is such a lovely child.
   She has such a sweet smile and positive
   atmosphere. She thinks everything
   is going to be better. [NC-1:4]

   Patient 2: She is friendly and positive.
   [NC-1:4]

For the third novice therapist, 35% of her responses during treatment of four patients were coded as social interchange. From the researcher's notes:

   When she elicited information from
   her patients, it was primarily related to
   the mechanics of treatment (eg, "Do
   you want your head up at this end?")
   or mechanics related to prescriptions
   or appointments (eg, "Did you call
   [your doctor] yesterday?").

Patients reported, "What a great gal she is!" and "She knows her job and is good at it." [NC-1:4] These descriptions of therapist behaviors indicate differences between master and novice clinicians. The patient interviews, however, provide evidence that patients held positive perceptions of their therapists, regardless of the therapists' level of expertise.

Importance of Teaching to Hands-on Care

The master clinicians described teaching as one of their most important clinical skills, and our observations support their self-assessments. For example,

   MC-2 comments that one of her most
   important clinical skills is her instruction
   in a home exercise program. This
   is reinforced by the amount of emphasis
   she gives this part of the treatment
   time. Over the years, she has learned
   to give more responsibility to her patients
   and to make them less dependent
   on her as their physical therapist.
   She believes this is directly related to
   better outcomes for her patients.
   [MC-2:6]

   MC-3: I think I'm good at teaching
   them what to do and impressing upon
   them that they have to do it. I never,
   never let them think that I'm going to
   make them better. Okay? They have to
   make themselves better. All I can do is
   be their coach and their cheerleader.
   But they have to play the game. They
   don't think that the things that happen
   to them in here are making them better.
   They don't come to therapy to get
   better. They come to learn how to get
   better. [MC-3:3]

Again, the novice clinicians differed from the master clinicians. They were quick to identify patient rapport and hands-on skills as important treatment skills, but they did not identify teaching as a similarly important skill. For example,

   NC-3 (In response to the question,
   What does communication mean?):
   That means I explain to them all the
   machines I am using. Try to develop a
   good rapport with them to develop
   their trust. I try to act interested in
   something they did outside of therapy
   so that I sort of have a balance. You
   know, you have a problem, but you
   also have other things in your life.... [NC-3:3]

   NC-1: I like touching, and I want to
   continue to use that, and I know that
   will take a lot of time. I know it is real
   powerful with lots of patients.... Just
   the human touch, and so many people
   need that humanistic approach. You
   need to approach from all angles, not
   just walking in and slapping on a
   treatment. [NC-1:5]

Confidence in Predicting Patient Outcomes

The master clinicians had a much more elaborate framework for prediction of patient outcomes, which appeared to be grounded in their confidence in gathering and interpreting clinical data. This confidence appears to be the result of the large reservoir of knowledge they have gathered about clinical problems over many years of experience. The master clinicians described more than the mere data-gathering stage of the patient evaluation. They linked the systematic collection of data to ongoing interpretation. For example, one master clinician described her evaluation process as working toward establishing a working diagnosis by performing a selective tissue examination to identify the structure at fault. She admitted that you cannot always be certain of your diagnosis and that it is a process of ruling out one thing and another until you have a collection of signs that point to one particular problem. She stated,

   And I think there's a lot of intuition
   and a lot of going with what your gut
   feeling tells you. You are going to get
   multiple signs that all indicate that, but
   can you ever prove it? The answer is
   no. New graduates sometimes can't
   make that distinction, or they jump to
   conclusions a little quicker ... you
   have to watch that. [MC-2:6]

The other two master clinicians expressed the same integrative approach, correlating the data collected as well as considerations of individual patients and the need to consider other factors.

   MC-1: I continually try to correlate
   subjective and objective data. I also try
   to think about the patients in their
   environment, about their work and
   recreational activities.... [MC-1:1]

   MC-1: I make dosage considerations
   for exercise and treatment, and I feel
   they are environmentally dependent. If
   patients have postural problems, I will
   consider that, even if they have strong
   signs. Once the signs are more stable,
   then I can look for other things. I try
   to withhold my judgment when I look
   at preexisting history. [MC-1:7]

   MC-3: So you establish that the diagnosis
   is correct by history, examination,
   and so forth. Then you should immediately
   be able to construct what the
   potential problems are with that diagnosis ... then
   you identify which are
   problems ... then you just have to
   judge how aggressive you are going to
   have to be with this, the dosage of that,
   how often do you have to see the
   patient, are there any particulars that
   are different about this patient than the
   ordinary, am I going to have to spend
   more time emphasizing this part of the
   education or convincing the patient to
   do this or whatever.... Then you just
   sit down and think.... [MC-3:5]

The novice clinicians held firmly to their evaluation framework as a basis for decision making, but freely admitted their frustrations with not knowing whether they could predict patient outcomes. As one novice describes,

   NC-1: I'm continuously taking objective
   data. Maybe too much of it. I don't
   know, but I feel like right now I have
   to do it, and I'm not comfortable
   enough with knowing what will happen ... I
   just note changes and try to
   deduct what is getting back to normal
   and what is not and getting worse. You
   know, why do we learn all this stuff if
   we don't use it everyday? What's supposedly
   normal for that person at that
   time. Until you know what "normal" is,
   you can't really know what "abnormal"
   is. [NC-1:1]

Another novice clinician related a similar account. She felt very dependent on her history-taking skills to lead her into her examination with a probable diagnosis in mind, but related that if the diagnosis did not emerge, things were harder: "But if I'm totally lost with the history, a lot of times I'm lost the rest of the way through." [NC-2:5] The researcher described the novice's process of examining patients as an attempt to get all the right information, hoping that a diagnosis would soon become clear. All novice therapists acknowledged other, more experienced therapists as a source of consultation.

Discussion--Explanations for Master-Novice Differences

The five attribute dimensions identified in the "Results" section represent differences we observed in the therapeutic interventions of our master and novice clinicians. Figure 2 represents our present understanding of the therapeutic intervention. We have followed Livingston and Borko's model of analyzing how expert and novice teachers think and act in a classroom setting. (14) Similar to the act of classroom teaching, which mandates that teachers cope and make sense of a vast amount of stimuli, the therapeutic intervention of the physical therapist can be characterized both as a complex cognitive skill (knowledge) and as improvisational performance.

Figure 2. Knowledge and performance attribute characteristics
that distinguish between the therapeutic intervention of novice
clinicians (n=3) and master clinicians (n=3).

THERAPEUTIC INTERVENTION

   KNOWLEDGE

ATTRIBUTE         MASTER         NOVICE

* Confidence      Elaborate      Collecting
   in Predicting   comfortable    data and
   Outcomes        schema for     hoping to
   interpreting   find a clear
   data and       direction;
   predicting     seeking
   outcomes       help from
   colleagues

   IMPROVISATIONAL PERFORMANCE

ATTRIBUTE         MASTER         NOVICE

* Ability to      Interruption   Interruption
   Control         controlled     not
   Environment                    controlled

* Focuses         Intense        Medley of
   Verbal and      patient-       approaches
   Nonverbal       centered       to gather
   Communication   encounters     data and
   maintain
   rapport

* Evaluation      Dynamite       Focus on
   and Use of      elicitation    finding the
   Patient         and use of     right data
   Illness and     data speci-
   Disease Data    fic to the
   patient

* Importance      Reliance on    Focus on
   of Teaching     teaching as    patient
   to Hands-on     an essential   rapport and
   Care            clinical       hands-on
   skill          skills

Knowledge Different Frameworks for Masters and Novices

One attribute dimension indicative of possible differences in master and novice knowledge was how our master clinicians demonstrated confidence in predicting patient outcomes. We did not discover this attribute in our previous work, (15) possibly because we did not interview clinicians about their treatment approaches. The master clinicians in this study presented a much more elaborate cognitive framework than did the novice clinicians. This cognitive framework was based, in part, on their knowledge of pathology and their clinical experience with patients, including an understanding of how patients may respond and adapt to their disease or disability. Alternatively, the novice clinicians were much more rule-governed, intent on collecting data to fill in their evaluation forms in the hope of finding an understandable diagnosis. Like experts in other fields, (1,2,24) the master clinicians appeared to have well-developed and easily accessible schemata that allowed them to evaluate and treat patients efficiently and with confidence.

The concept of schema represents how acquired knowledge is organized and how the structure of the knowledge can facilitate its use in certain situations. (9,24,25) A fundamental distinction made among types of schema is the difference between declarative and procedural knowledge. Declarative knowledge is static and deals with recognizing a specific case or providing the data to perform the skill, whereas procedural knowledge is dynamic and involves not only recall but transformation of information. Both forms of knowledge are necessary for skilled performance. One distinction drawn between experts and novices is that experts have more procedural knowledge (ie, they know how to perform their craft). (25) In our cases, master clinicians knew when they needed to gather more data and how to selectively gather important data versus gathering all the data possible.

These master clinicians also were comfortable with the uncertainty of not knowing the immediate diagnosis and were able to plan accordingly. They implemented a process of ruling items in and out until they had a collection of clinical symptoms and signs. Schon (10) describes the actual practice environment as the swampy lowland of professional practice. He argues that the professional practice situation defies the strict application of technical skills and that the professional must recognize the situation as unique and identify the problems at hand. The master clinicians in our study all spoke of the necessity of correlating subjective and objective data, weighing the evidence, and withholding judgment until they had corroborated their data. These three master clinicians practicing in orthopedic settings may have been using a receptive data-gathering style and a systematic information-processing style, similar to the orthopedic clinicians studied by May and Dennis. (5) The novice clinicians, however, were less selective of the data they gathered and more intent on gathering as much data as possible until the diagnosis emerged.

Improvisational Performance Different Acts by Master Clinicians and Novices

Improvisational performance refers to what actually happens when the therapist is treating patients. Though therapists draw from their limited or vast knowledge and experience, their actual work demands a great deal of improvisation to respond to individual patient disease or disability states and patient-specific needs. Analysis of our other four attribute dimensions demonstrate differences in improvisational performance between master and novice clinicians.

Even though the novice and master clinicians worked in similar clinical settings, the master clinicians, like the experienced clinicians in our previous work, as were notably different from the novice clinicians in that they set the stage for an intense therapeutic intervention through their control of the clinical environment. The master clinicians took charge of how their time was allocated and of when and how they would respond to the numerous interruptions that were an inherent part of a busy orthopedic clinical setting. Conversely, the novice clinicians were much more reactive to external stimuli and struggled to manage individual patient care amidst the confusion of competing demands from multiply scheduled patients and constant interruptions from intercom and staff requests. This attribute--ability to control the environment--represents a combination of components from two themes identified in our prior work: allocation of treatment time and impact of the therapeutic environment (Tab. 1).

Another attribute dimension that differentiated master from novice clinician performance was the ability to maintain focused verbal and nonverbal communication with the patient. This attribute dimension is a combination of two themes outlined in our previous study--degree of responsive therapeutic interaction and therapist integration of therapeutic and social interaction (Tab. 1). (15) Our analysis suggests that the master clinicians were able to elicit and integrate data related to the patient's clinical problem, provide patient education, engage in social interchange, and render hands-on evaluation and treatment in a smooth, integrated fashion. They were "with" each patient in an episode akin to an encounter. In contrast, our novice clinicians were much more intent on filling in evaluation forms, attending to the mechanics of treatment, or engaging in social exchange. Our observations of novices demonstrated that at this point in their careers, they were not able to integrate all the multiple aspects of the therapeutic process.

Research on experts in teaching demonstrates that experts show strong self-regulatory, or meta-cognitive, capabilities. (26) These capabilities enable experts to skillfully plan and implement activities and to use their time sensibly. The novice clinicians, unlike the master clinicians, may be limited in their ability to attend to multiple tasks during the therapeutic process, because they must concentrate on attending to filling in their framework of content knowledge.

A third attribute dimension indicative of differences in master and novice clinician performance was evaluation and use of patient illness and disease data. This attribute includes components described in three previously identified themes: types and use of information gathered from patients, degree of responsible therapeutic interaction, and therapist integration of therapeutic and social interactions (Tab. 1). (15) Master clinicians collected necessary objective data, and listened intently to how their patients interpreted their clinical problems and recounted the social and emotional sequelae. These clinicians used both types of data as they tailored treatment programs specifically to their patients' needs. Patients' perceptions of their conditions was a central aspect of master clinician evaluation and treatment. The master clinicians, just as researchers have found in experts in other fields, were more sensitive to task demands and the social structure of the job situation. (1,3,11) The novice clinicians, however, could be described in a remarkably similar fashion to the novices (physical therapy students) studied by Thomas-Edding:

   It was also clear that the experts constantly
   used cues to lead them in their
   data collection and interpretation. Students,
   on the other hand, spent less
   time amassing information and conducted
   routine interviews, asking standard
   questions. For example, all students
   asked about the patient's
   occupation but did not respond to
   several cues that were given. Experts
   asked the same question but continued
   the questioning to learn exactly what
   hand movements were needed for the
   patient's job. (4(p103))

A fourth attribute dimension of differences between master and novice clinicians was recognition that teaching the patient was as important as hands-on care. This attribute dimension was modified from the same three initial study themes noted earlier for evaluation and use of patient illness and disease data (Tab. 1). (15) Master clinicians explicitly valued teaching patients to assume responsibility for their own health care as an essential dimension of their performance in the therapeutic intervention. These clinicians reported they have learned from their experience that when patients are empowered to take charge of their own health, the result is better health care outcomes. These master clinicians may also have realized that a patient's presence in a physical therapy clinic is only a momentary phenomenon in the span of that person's life and that teaching patients to care for themselves is a way of increasing the value or usefulness of that short event. Conversely, the novice clinicians, at this stage of their professional development, were more intent on mastering their hands-on skills and making the patients feel comfortable in therapy and receptive to their ministrations.

Although our primary focus was investigating the work of novice and master clinicians and identifying differences, one similarity that arose was patient perceptions regarding their therapists. All of the patients interviewed held positive perceptions of their therapists. This preliminary finding of patients' positive perceptions, regardless of level of therapist expertise, raises an interesting question for future investigation.

Implications

Data from the master and novice clinicians revealed important differences in knowledge and improvisational performance that are supported by other studies of expertise. (1,2,4,14,24) These differences, based on a very small sample of master and novice clinicians working in orthopedic settings, are part of an emerging theory of expert practice in physical therapy. As more data are gathered on larger and more diverse subjects, these attributes and other aspects of the conceptual framework will be revised and expanded. The methods used in this study did not permit us to standardize variables across subjects (eg, age is a factor that might influence improvisational performance and knowledge of the therapist). Neither did our design permit us to be blind to the master or novice status of the therapist prior to the analysis of the data. Our particular strategy of sample selection is consistent with other research on the nature of expertise. For example, studies of expertise frequently use samples for which the researcher has prior knowledge of the subjects' level of expertise. These investigations focus primarily on understanding the "wisdom of the practitioner" and are often done by contrasting expert performance with novice performance. (2,3,8,11,24) Our primary goals were continued exploration of professional practice and elaboration and refinement of a conceptual framework. We were interested in illuminating the thoughts and actions of clinicians in a particular context. We acknowledge, however, that there are other variables that could well account for some of the differences we observed.

Our work represents only a beginning point for further exploration and investigation of the multiple dimensions that are part of clinical practice. Additional research, using similar methods for data gathering and analysis, is necessary for further expansion and revision of our conceptual framework. In qualitative research, the issue of replicability is important and addressed through the confidence one has in the investigator's interpretation of reality. Does the investigator present a coherent, illuminating description of reality that is consistent with a detailed study of the situation? (27) This concept is described well by Strauss and Corbin:

   [P]robably no theory that deals with a
   social/psychological phenomenon is
   actually reproducible, insofar as finding
   new situations or other situations
   whose conditions exactly match those
   of the original situations ... unlike the
   study of a physical phenomenon, it is
   very difficult to set up experimental or
   other designs in which one can recreate
   all of the original conditions.... Another
   way of denoting reproducibility
   is ... given the same theoretical
   perspective of the original researcher
   and following the same general rules
   for data gathering and analysis, plus a
   similar set of conditions, another investigator
   should be able to come up with
   the same theoretical explanation about
   the given phenomenon. Whatever
   discrepancies that arise can be worked
   out through reexamination of the data
   and identification of the different conditions
   that may be operating in each
   case. (28(pp250-251))

Our use of qualitative research methods lets us investigate the phenomenon of expert practice in context and from the eyes of the actors. This line of research is in concert with investigations in other disciplines in which activities are focused on analyzing expertise and identifying aspects of practical knowledge, that is, knowledge that emerges from observing and investigating practice. (3,14,29) Our suggestions for future research are as follows:

1. A much larger study of master and novice clinicians engaged in therapeutic interventions is needed to affirm or revise our five attribute dimensions and add more information to other components of our conceptual framework. Subsequently, similar studies will be needed of master clinicians working in different settings such as pediatric and adult neurological, cardiopulmonary, and arthritis settings to add to our understanding of the role of disease and disability in altering what happens in the therapeutic intervention.

2. Longitudinal studies are needed to enhance our understanding of how one gets from the chaotic world of the novice clinician to the organized world of the adept master clinician. For example, how and when does one learn the importance of teaching as a clinical skill, and when does one learn how to elicit and use rich patient-specific illness data?

3. Research is needed to determine whether master clinicians produce different and more effective patient outcomes than do novice clinicians. In this study, patients' perceptions of care appeared to be unaffected by the improvisational performance slips and knowledge misses of the novices. The patients interviewed pronounced the novices to be effective practitioners.

Continued research in these three areas holds important implications for the admission of students to physical therapy education programs and for both entry-level and continuing education of physical therapists, specifically related to the efficacy and quality of patient care.

Conclusions

The work of the physical therapist was further investigated with a small number of master and novice clinicians. Five attribute dimensions from an elaborated conceptual framework--one on knowledge and four on improvisational performance--appear to distinguish between our sample of master and novice clinicians. Follow-up investigations will be needed to further develop and expand our understanding of an emerging theory of expert practice in physical therapy. Such investigations are crucial to the improvement of physical therapy practice and professional education.

Table 1. Original Themes Distinguishing Between Novice and
Experienced Physical Therapy Clinicians (15)

Theme

1. Allocation of treatment time (actual time with patient and
what is done during that time)

2. Impact of the therapeutic environment on the treatment
session (management of interruptions and tasks outside of direct
patient treatment)

3. Types and uses of information gathered from patients
(building upon patient's responses; giving patient information;
social interchange)

4. Degree of responsive therapeutic interaction (integration of
verbal and tactile cues)

5. Therapist integration of therapeutic and social interactions
("entering the patient's life" to elicit and give relevant
information about the disability)
Table 2. Attribute Dimensions (Revised Themes) that Distinguish
Between Master and Novice Clinicians

Attribute Dimension

1. Ability to control the environment

2. Evaluation and use of patient illness and disease data in a
context-rich evaluation

3. Focused verbal and nonverbal connection with the patient

4. Equal importance of teaching to hands-on care

5. Confidence in predicting effective patient outcomes based on
knowledge of pathology and experience with the course of healing

Appendix 1. Interview Guide Questions for Patient Interviews

1. What do you think caused your illness/disability? What happened?

2. How severe is your illness/disability? Will it have a short or long course?

3. What kind of treatment do you think you should receive?

4. What are the most important results you hope to receive from this treatment?

5. What are the chief problems your illness/disability has caused you?

6. What do you fear most about your illness/disability?

Appendix 2. Interview Guide Questions for Physical Therapist Interviews

1. What kind of data did you use to arrive at your assessment?

2. How (by what process) did you arrive at your decision?

3. What kind of treatment approach did you use and why?

4. What are your most important skills in patient care?

5. How do you know when you have been effective?

6. How have you grown most as a clinician over the years?

Appendix 3. Coding Categories and Examples for Observations
of the Physical Therapist and the Patient (a)

Coding Category                 Example

Physical therapist

   Seeks Information (SI)        "Have you been doing the
                                 exercises?"
                                 "Is the pain the same as before?"

   Rewards/Encourages (R)        "That is excellent! Much better!"
   (praising and reassuring,     "You are moving better than last
   not educating, patients)      week!"

   Simple Commands or Comments   "Stand up and let me check your
   Related to Mechanics of       back."
   Treatment Procedures (Mp)     "Let me know if that gets too hot."
                                 "I think five repetitions is
                                 enough."
                                 "Relax and I'll be right back to
                                 increase the weights."

   Instruction to Patient on     "After you get into position,
   How to Do an Activity (I)     relax and lift your heel and leg
                                 so your knee is straight."
                                 "See if you can hold it longer and
                                 slowly stretch rather than
                                 bounce."

   Explanation (E)               "You have a minor muscle pull that
   (statement that informs       is causing the pain."
   or explains why to a          "You are more flexible in that
   patient)                      direction because your joint is
                                 looser."
                                 "You eventually want to get your
   heel all the way down."

   Social Interchange (SO)       "Did you have a good time on
   (exchange unrelated or,       Saturday?"
   peripherally related to       "You came in feeling great and
   treatment (humor)             leave feeling lousy. How's that
                                 for therapy!" (laughter)

   Miscellaneous (M)             "Tell the next patient to wait
   (information not directed     "I can't find the extension cord."
   to the patient or sched-      "When can you come in next week?'
   uling appointments)

Patient

   Seeks Information (SI)        "Why does that feel so tight?"
                                 "Is it all right to swim?"

   Rewards Physical Therapist    "Thank you."
   (R)                           "You have very gentle hands."

   Reports Status or Responds    "That exercise is very hard to do."
   to Physical Therapist         "I'm feeling the heat right now."
   (RS)                          "I can go pretty much the whole day
   without feeling any pain."

   Social Interchange (SO)       "I have a friend help massage my
                                 legs."
                                 "Dr Johnson is a good family
                                 friend."
                                 "Do you like working here?"

   Miscellaneous (M)             "I'm not sure where I put my car
                                 keys."
                                 Thursday afternoon is a good time
                                 for me to come in again."

(a) Adapted and reprinted with permission. (15)

Acknowledgments

We are truly indebted to the physical therapists and patients who graciously allowed us to observe them during treatment sessions and thoughtfully responded to our questions regarding their reflections, opinions, and perceptions of the work of the physical therapist.

References

(1) Berliner D. In pursuit of the expert pedagogue. Educational Researcher 1987;16(6):5-13.

(2) Benner P. From Novice to Expert: Excellence and Power in Clinical Nursing Practice. Menlo Park, Calif: Addison-Wesley Publishing Co Inc; 1984.

(3) Elstein AS, Shulman L, Sprafka S. Medical problem solving: a ten-year retrospective. Evaluation and the Health Professions. 1990;13(1): 5-36.

(4) Thomas-Edding D. Clinical problem solving in physical therapy and its implications for curriculum development. In: Proceedings of the 10th International Congress of the World Confederation for Physical Therapy; May 17-22, 1987; Sydney, Australia. 1987:100-104.

(5) May BJ, Dennis JK. Expert decision making in physical therapy: a survey of practitioners. Phys Ther. 1991;71:190-202.

(6) DePoy E. Mastery in clinical occupational therapy. Am J Occup Ther. 1990;44:415-422.

(7) Patel V, Groen G. Knowledge-based solutions strategies in medical reasoning. Cognitive Science. 1986;10:91-116.

(8) Barrows H, Feltovich P. The clinical reasoning process. Med Educ. 1987;21:86-91.

(9) Calderhead J. Exploring Teacher Thinking. London, England: Cassell Educational Ltd; 1978.

(10) Schon D. The Reflective Practitioner New York, NY: Basic Books; 1987.

(11) Shulman L. The wisdom of practice in teaching: managing complexity in medicine and teaching. In: Berliner D, Rosenshine B, eds. Talks to Teachers. New York, NY: Random House Inc; 1987:369-386.

(12) Kagan D. Teaching as clinical problem solving: a critical examination of the analogy and its implications. Review of Educational Research. 1988;58:482-505.

(13) Payton OD. Clinical reasoning process in physical therapy. Phys Ther 1985;65:924-928.

(14) Livingston C, Borko H. Expert-novice differences in teaching: a cognitive analysis and implications for teacher education. Journal of Teacher Education. 1989;40(4):36-42.

(15) Jensen GM, Shepard KF, Hack LM. The novice versus the experienced clinician: insights into the work of the physical therapist. Phys Ther. 1990;70:314-323.

(16) Miles M, Hubermann A. Qualitative Data Analysis: A Sourcebook of New Methods. Newbury Park, Calif: Sage Publications Inc; 1984; 27-48, 79-203.

(17) Merriam S. Case Study Research in Education: A Qualitative Approach San Francisco, Calif: Jossey-Bass Inc Publishers; 1988:1-21, 48-49, 163-184.

(18) Helmer O. Looking Forward: A Guide to Futures Research. Newbury Park, Calif: Sage Publications Inc; 1983:58-60.

(19) Erickson F. Qualitative methods in research on teaching. In: Wittrock M, ed. Handbook of Research on Teaching. 3rd ed. New York, NY: Macmillan Publishing; 1986:119-161.

(20) Yin R. Case Study Research: Design and Methods. Newbury Park, Calif: Sage Publications Inc; 1984.

(21) Jensen GM. The work of accreditation onsite evaluators: enhancing the development of a profession. Phys Ther 1988;68:1517-1525.

(22) LeCompte M, Goetz J. Problems of reliability and validity in ethnographic research. Review of Educational Research. 1982;52:31-60.

(23) Kleinman A, Eisenberg L, Good B. Clinical lessons from anthropologic and cross-cultural research. Ann Intern Med 1978;88:251-258.

(24) Chi M, Feltovich P, Glaser R. Representation of physics knowledge by experts and novices. Cognitive Science. 1981;5:121-152.

(25) Cervero R. Effective Continuing Education for Professionals. San Francisco, Calif: Jossey-Bass Inc Publishers; 1988:38-56.

(26) Leinhardt G, Greeno J. The cognitive skill of teaching. Journal of Educational Psychology. 1986;78:75-95.

(27) Schofield J. Increasing the generalizability of qualitative research. In: Eisner E, Peshkin A, eds. Qualitative Inquiry in Education. New York, NY: Teachers College Press; 1990:201-232.

(28) Strauss A, Corbin J. Basics of Qualitative Research: Grounded Theory Procedures and Techniques Newbury Park, Calif: Sage Publications Inc; 1990:250-251.

(29) Mattingly C, Gillette N. Anthropology, occupational therapy, and action research. Am J Occup Ther 1991;45:972-978.

GM Jensen, PhD, PT, is Associate Professor and Research Coordinator, Department of Physical Therapy, Samuel Merritt College, 370 Hawthorne St, Oakland, CA 94609 (USA). Address correspondence to Dr Jensen.

KF Shepard, PhD, PT, FAPTA, is Professor and Assistant Dean for Graduate Studies, College of Allied Health Professions, Temple University, Philadelphia, PA 19140.

J Gwyer, PhD, PT, is Assistant Professor, Graduate Program in Physical Therapy, Duke University, Durham, NC 27710.

LM Hack, PhD, MBA, PT, is Assistant Professor, Department of Physical Therapy, Temple University.

This study was approved by the institutional review boards of The University of Alabama at Birmingham, Duke University, and Temple University.

This article was submitted May 7, 1991, and was accepted June 3, 1992.