Hypothesis-oriented algorithms.

Hypothesis-Oriented Algorithms The concept of clinical decision making by physical therapists in the management of patients remains so new that in the index of our literature there is no category for this important subject topic. As preparation for presenting this article, we examined the Sixty-Five-Year Index to Physical Therapy [1] and the yearly index of Physical Therapy to date. Clinical decision making is not a title that is used in any of these indexes. We became curious about how the original article on the hypothesis-oriented algorithm for clinicians (HOAC) by Rothstein and Echternach published in the September 1986 issue of the Journal was categorized in that year's index. (2) We would have to admit that we had a difficult time finding the category. It was indexed under the category Tests and Measurements--General. [2] This categorization compounded the difficulty in looking for other articles that have been published by physical therapists in the area of clinical decision making. In looking at the category of Tests and Measurements--General, we find an interesting collection of articles, but certainly no consistent theme on clinical decision making.

For the purpose of discussion of the HOAC, it should be understood that nearly everything we are going to present is based on the original concepts presented by Rothstein and Echternach. [3] Our original interest in this topic was stimulated by a mutual interest in the relationship between the examination of the patient and treatment planning. In our discussions with each other, we made the observation that many times the examination performed by the therapist seemed to have little relationship to the actual treatment plan that was then carried out. An example might be when the therapist would carefully examine a patient with low back pain, localizing the problem to specific structures and demonstrating specific loss of function but carrying out a treatment plant that is of a routine nature such as moist heat, massage, and Williams exercise. Other examples come to mind such as the use of a specific protocol for treatment of patients with total-joint replacements. Sometimes these protocols for treatment seem to replace the thought process associated with relating treatment to a patient's unique set of problems.

This observation led us to look at how evaluation or examination methods in the field of physical therapy have developed. Essentially, physical therapy has grown as a field in which the first physical therapy evaluations actually consisted of physician-ordered tests. A physical therapist would be requested by the physician to perform muscle testing or range-of-motion testing. The physician would then use the information gathered by the physical therapist in planning the patient's care. Very briefly, a summary of development of the evaluation or examination of the patients in the field then progressed to special examinations devised by individuals for specific purposes and then to more general evaluation schemes. An example of this progression is the development of evaluation or assessment methods that could guide treatment such as Brunnstrom's approach to the hemiplegic patient. [1] Following this development came examination methods that guided the process of describing the status of the patient but that were not tied to any one system of patient care. This development forces the therapist to integrate the information derived from the evaluation process with treatment planning.

The progression of this idea has led us to reexamine our definition of the word evaluation as "what a physical therapist does with a patient," and there has been discussion that the term examination better defines what a physical therapist is doing with a patient. More recently, there have been discussions of whether this process that a physical therapist goes through is one that leads to a diagnosis. [3] It seems obvious that this process of examination of a patient, treatment planning, implementation of treatment, and treatment reevaluation needs an overall clinical decision-making scheme to allow the physical therapist to function in a changing environment. The HOAC is an attempt to meet this goal.

A clinical decision-making model should meet the following objectives:

1. The model should serve the physical therapist in the setting of direct access and where increased autonomy is required.

2. The models that are developed need to be able to be applied in any setting where treatment is planned as a result of the physical therapy examination.

3. The clinical decision-making scheme should guide the process from patient examination to treatment planning and reevaluation and should appear as a logical sequence of activities.

4. The clinical decision-making model must be independent of treatment philosophies. This requirement becomes obvious when thinking about evaluating the ability of a particular treatment philosophy to change a patient's progress significantly. If the clinical decision-making model is independent of the treatment philosophy, it is more likely to be an objective method of assessing patient progress. The objectivity of the preexamination of the patient is crucial in assessing whether the patient is making the appropriate progress in meeting treatment goals. Because treatment goals should be stated in terms of function, it seems obvious that objectivity is greater if the system for examining progress is not tied to a specific treatment philosophy.

5. The clinical decision-making model should guide the physical therapist in knowing when to seek consultation or when to initiate referrals to other health care professionals, including other physical therapists.

Hoog, in defining the role of the physical therapist, stated that the therapist has five functions: 1) patient evaluation; 2) treatment planning; 3) treatment implementation; 4) communication with other health care professionals, primarily about the patient; and 5) communication and interaction with the public. [4]

The HOAC system serves four of these five functions. The algorithmic method used is also based on many of the same thought processes that go into the problem-oriented approach to patient care; however, it does differ from this system in certain important ways. [5] The HOAC places responsiblity with the therapist to define goals and determine whether they have been met. The physical therapist has the responsibility for patient management until the patient has reached the goals that have been set by the therapist. Another major difference is the HOAC's emphasis on the therapist generating a hypothesis about the cause of the patient's problem. The requirement for the therapist to generate hypotheses about the cause of the patient's problem is fundamental to the use of the HOAC system and is the step that most directly relates to effective treament planning.

The HOAC consists of two parts to guide the process. Part One is a guide to evaluation and treatment planning (Fig. 1), and Part Two is a guide to the anticipated clinical decisions that must be made (Fig. 2).

Part One--Guidelines for

Evaluation and

Treatment Planning

Initial Data Collection

The first step is to collect initial data (Fig. 1). This data collection is always done in the context of why physical therapy assistance is being sought. This initial data collection refers to information about the patient obtained from interview, history, chart review, and subjective information.

Problem Statement

The second step is to generate a problem statement. We think it is helpful if this generation of a problem statement is done as described in the problem-oriented format and is stated in terms of the patient's reasons for seeking help and not in terms of a diagnosis. [5] A functional loss can often be expressed in the same terms that the patient would use. Problems can be stated as anticipated problems (eg, those that may develop in the future as a result of the patient's current problems). An example that has been cited previously by Rothstein and Echternach [2] is the below-knee amputee who does not currently have any flexion contracture but might well develop the condition unless it is anticipated.

The generation of a problem statement and a problem list at this point in the algorithm is difficult for many therapist to understand on first examination of thsi method. The reasons for developing a problem statement prior to examination of the patient, however, are logical and consistent with actual clinical practice, particularly if thought processes of clinicians are examined. [6,7] A second important reason is that patients seek help because of their problems, which then become reasons for the physical therapist being sought out by the patient or by a referring health care practitioner. It is beneficial to both the physical therapist and the patient to identify the problem prior to any redefining of the problem by the physical therapist, which is more appropriate later in the sequence.

Goals

Following the generation of a problem statement, the therapist is then asked to establish goals that are measurable and functional with a temporal element. Goals should be stated in behavioral terms, indicating what the patient and the physical therapist hope to achieve. They should be stated in terms that are measurable, which in essence means terms that the physical therapist can define operationally. This process can be facilitated in the practice setting where those who practice together use lists of operationally defined terms. Goals are emphasized for the purposes of understanding why the patient is to be treated and defining when the patient should be discharged, that is, when the goals have been met. It should be understood that goals at this stage are modifiable and tentative. Setting goals prior to the evaluation improves the statement of goals in terms of patient outcome. A goal such as "The patient should be able to dress himself or herself or to achieve enough ROM of the shoulder to be able to put on a shirt" is more important in terms of the patient than the ability to achieve 100 degrees of shoulder flexion. This process tends to keep the focus on which goals are the most important to the patient. This process also serves to keep the physical therapist focused on understanding what the patient's expectations are. This understanding can guide the patient and the physical therapist to understand what can reasonably be expected from the treatment program.

Examination

The third step is examination and the collection of objective data about the patient. In this stage, the physical therapist physically examines the patient and collects data regarding the patient's health status. The physical therapist uses past experiences and methods to carry out the examination. This process may be guided by the practice group or departmental guidelines. In this stage of the process, the HOAC may incorporate other algorithms, such as those developed in institutional settings where physical therapists have been primary evaluators of patients' problems for some time. [8] Specific portions of the evaluation or examination might be related to treatment philosophies. For example, part of the examination might be guided by the neurodevelopmental treatment philosophy. The NDT philosophy may be related to treatment, but only a portion of this examination process should be related to a treatment philosophy.

Generation of a Hypothesis

The fourth step is generation of a working hypothesis about why the goals can or cannot be met at the present time. This generation of a hypothesis is based on the patient examination and integration of all information about the patient. At this point, the physical therapist develops a clinical impression from which he or she can generate a hypothesis about the causes of the patient's problem. During this process, more than one hypothesis may be generated. The treatment plan should be based on the hypothesis. An example might be that the patient has an ambulation problem. The patient reports knee pain after walking long distances. At this point, the physical therapist specifies a goal that the patient should be able to walk a specific distance without pain based on the patient's needs. During the examination, the therapist notes that the pain is caused by a lack of normal progression from heelstrike to mid-stance and that the patient does not flex the knee during this part of the gait cycle. When assessing the patient's ROM, the therapist discovers that the patient can flex the knee only to 15 degrees and experiences pain past 15 degrees of flexion. The therapist then hypothesizes that the functional loss is a result of a gait deviation and related to the ROM limitation and pain experienced during ROM efforts. If the treatment plan were designed to increase muscle power, it would not be appropriate because the hypothesis did not deal with the loss of muscle power but rather with the patient's loss of pain-free ROM. Theoretically, if no hypothesis can be generated, no treatment can be administered. The therapist would then be required to refer the patient to another practitioner or to seek consultation from another physical therapist who may be able to generate a hypothesis about the patient's problem. In this process, the consultation with another physical therapist might lead to additional examination procedures, which could then lead to a hypothesis. A hypothesis is a clinical impression based on assumption of causality. It is a testable idea. It is tentative, but at the moment that it is made, it is the best estimate of the problem. If the hypothesis is that the functional deficits of the patient are caused by a problem that the physical therapist is not prepared to treat, the physical therapist must then make a referral to the appropriate health care professional. For example, the patient's problem might be the loss of knee ROM caused by septic arthritis. [2] Referral, in this instance, might be the appropriate action.

At this stage, the therapist should ask whether the goals that have been set initially are viable. If the goals are not viable, they should be modified; if they are viable, the therapist should proceed to the next step in the algorithm.

Plan Reevaluation Methodology

This next step is to plan the reevaluation methodology and to schedule the dates for these reevaluations. These testing criteria are established before treatment begins and are independent of treatment goals. They serve to determine whether the hypothesis is correct. They also determine whether the treatment is having the desired effect and whether the treatment is effective in achieving the goals established by the therapist. This reevaluation methodology may resemble "short-term goals" of common usage, but this is a faulty way to think of the process. Increasing ROM is not a goal in itself, but is a goal only in terms of increased function.

In summary, the plan for reevaluation methodology should outline the procedures for reevaluation, including mechanisms for testing whether a patient has met the goals. A reevaluation schedule should be established. Obviously, a reevaluation should be performed earlier than scheduled if major changes occur in the patient's status.

Plan Strategy and Tactics

The next step is to plan a treatment strategy based on the hypothesis. This is the overall treatment approach. There are times when consultation will be needed by the physical therapist to be able to plan an overall treatment approach. This consultation may often be another physical therapist who has more experience or expertise in management of the kind of patient the treatment strategy is being planned for. The purpose of separating this step of planning an overall treatment approach from the next step, which is to plan the tactics of the treatment plan, is that there are times when the overall approach may be correct but the specific implementation may not be effective. It is useful to understand the differences between the two steps. The tactics are the specific treatments or means of implementing the treatment plan. They are closely linked to the strategy. The tactics of the treatment plan include not just those carried out by the physical therapist, but also what is done by the physical therapist assistant, the patients' family members, or the patients themselves.

Implementation of Tactics

The last step in the first part of this algorithm is the implementation of the tactics, or carrying out the treatment. The treatment is conducted for a specific period of time, and reevaluations should be conducted on schedule. If the patient's reevaluation must be scheduled earlier than planned, this fact may indicate that the goals and hypotheses may need to be reexamined. Such reexamination of goals and hypotheses entails starting over at the beginning of the HOAC. Obviously, some patient conditions may become exacerbated and thus present a need to reevaluate and to change the course of treatment.

Part Two--Branching Program

Part Two is the reassessment portion of the algorithm (Fig. 2). The first stage is reassessment and asks the question, Have goals been met? If goals have been met, then the patient can be discharged. If the goals have not been met, the physical therapist is then directed by the algorithm to a branching program to examine the previous steps in their reverse order. This sequence requires the physical therapist to ask questions regarding the most concrete issues first.

If the goals have not been met, the next stage of this part of the algorithm is to determine whether the tactics are being implemented correctly. In this stage, the therapist asks, Is treatment being implemented as planned? A variety of factors might explain why the goals have not been met. One reason might be that the patient simply failed to understand the treatment goals and therefore could not cooperate fully. Another possible explanation might be that the patient, or the patient's family, failed to comply with the treatment program. All aspects of the issue of compliance must be examined, including the possibility that treatment was carried out poorly by the physical therapist, by the physical therapist assistant, or by others. If implementation of the tactics was poor, then the problem should be corrected at this level and will involve establishing a new treatment and reevaluation schedule.

If the implementation was correct, however, then the therapist must consider whether the strategy was appropriate. At this point, the question should be asked whether one overall treatment approach is better than another. If the rate of progress has been slow, perhaps the problem has been unrealistic expectations. If the strategy and tactics have been correct, but the goals have not been met and the test criteria have been met, then a new hypothesis must be generated. This generation of a new hypothesis may require referral. An example might be that bracing is needed, and this treatment could be provided by an orthotist. Another example might be that surgery is needed, and a referral to an appropriately trained surgeon would be the next step. The goals may need to be revised based upon the generation of a new hypothesis and may indicate that lower functional levels for the patient have to be accepted.

Using the HOAC system, the patient can be discharged only under two clearly defined circumstances: 1) A referral is made to another health care practitioner, or 2) the goals, either modified or original, have been met. Documentation of the actions leading to either of these two outcomes is needed, especially when referral is made or when goals have been modified.

Summary and Conclusions

This has been a brief review of the HOAC system. Direct access and the use of other algorithms to guide this process fit well with the HOAC system. An example would be an algorithm designed to guide the physical therapist through the process of being the first evaluator of a patient with low back pain (Fig. 3) (Physical Therapy Department, US Public Health Service Hospital, San Francisco, Calif; unpublished data; 1976). The review of the system occurs at the first step (collect initial data). The examination of the patient occurs at step three. Step four would be to generate a working hypothesis that could lead either to referral to an appropriate practitioner or to continuation through the next step in this algorithm. The HOAC system requires the physical therapist to deal with defined problems and to document the actions that have been taken.

This system lends itself to peer review and quality assurance questions. It helps the physical therapist to review his or her own performance, and it can help to identify weaknesses in patient management and at which step in the process these weaknesses occurred (eg, at the goal-setting level, at the hypothesis-generation level). By creating a better understanding of the performance of the physical therapist, this model can help determine which continuing education needs must be met.

This method can also help us to understand the scientific basis for practice in that it requires testing of assumptions of treatment planning and goal achievement. It also serves the needs of the physical therapist in direct access by defining circumstances for both referral and consultation, and it requires a statement of the rationale for these decisions. In this sense, the HOAC system mandates a logical justification of any screening program and any subsequent interventions. The HOAC system can also help students in both the classroom and the clinical environment develop their framework for clinical decision making. Because it organizes patient management in a progressive and logical fashion, it can help students to integrate both theory and practice and it can help clinical educators use this system to identify student strengths and weaknesses in the management of patients.

It is the hope of the authors that the use of a system such as HOAC and the examination of other clinical decision-making models will create increased interest and publication in the area of clinical decision making. Increased interest and publication may lead to a separate categorization of future articles published in this area, making it easier for those of us concerned about this area to communicate with each other.

J. Echternach, EdD, PT, is Professor and Chairman, Department of Community Health Professions and Physical Therapy, and Director, Program in Physical Therapy, Old Dominion University, Norfolk, VA 23508 (USA).

J Rothstein, PhD, PT, is Associate Professor, Department of Physical Therapy, School of Allied Health Professions, Medical College of Virginia, Virginia Commonwealth University, PO Box 224, MCV Station, Richmond, VA 23298-0024.

References

[1] Sixty-Five Index of Physical Therapy. Alexandria, VA, American Physical Therapy Association, 1986

[2] Rothstein JM, Echternach JL: Hypothesis-oriented algorithm for clinicians: A method for evaluation and treatment planning. Phjys Ther 66: 1388-1394, 1986

[3] Rose SJ: Editorial: Musing on diagnosis. Phys Ther 68: 1665, 1988

[4] Hoog J: Professional standards for quality patient care. Phys Ther 49:1364-1368, 1969

[5] Echternach JL: Use of the problem-oriented clinical note in a physical therapy department. Phys Ther 54:19-22, 1974

[6] Elstein AS, Kagan N, Shulman LS, et al: Methods and theory in the study of medical inquiry. J Med Educ 47: 85-92, 1972

[7] Payton OD: Clinical reasoning process in physical therapy. Phys Ther 65: 924-928, 1985

[8] Overman SS, Larson JW, Dickstein DA, et al: Physical Therapy care for low back pain: Monitored program of first-contact non-physician care. Phys Ther 68: 199-207, 1988