Objective measures.

Objective measures. Phys Ther 69:590-593, 1989 To address all the questions relevant to objective measures is not possible in a brief presentation. I, therefore, will limit the following to three questions: 1) What are the objective measures? 2) Why do we need objective measures? and 3) How are we doing with objective measures, and where might we go from here?

What Are Objective Measures?

Regarding the substance of objective measures, I would like to deal with two concepts. One is the relative independence between objective measures and the examiner. The second is the persistent dependence of our objective measures on the patient.

Rothstein has distinguished objective measures as those that are not dependent on the examiner. [1] All of the measures that therapists use, however, are probably dependent in part on the examiner. In elaborating on Rothstein's statement, I would define objective measures as those that are not dependent primarily on the judgment of the examiner. Such measures typically involve the use of instrumentation and are expressed in real numbers. Thus, manual muscle testing, which involves the judgment of the examiner, may be considered subjective rather than objective. Hand-held dynamometry may be dependent, like manual muscle testing, on the strength of the examiner. [2] Nevertheless, it involves the use of instrumentation and results in quantitative and nonjudgmental values of muscle force production. Fixed force gauges provide the same quantitative objective values but are dependent on the examiner only in regard to how he or she sets up the test and reads the gauge. Shoulder subluxation is another example. The shoulder might be judged subjectively as "clearly subluxed" or "not clearly subluxed." [3] For greater precision, albeit still subjective, the distance between the acromial process and humeral head could be judged by the clinician "in terms of the possible number of fingers that could be inserted." n4 Truly objective measurements of shoulder subluxation can be obtained using linear measurements from radiographs taken by various techniques. [4] Of course, measures of muscle strength and should subluxation are only two examples that involve a spectrum of subjective and objective measures.

In physical theraphy, many, if not most, of the measures we use are dependent on the patient. Thus, the patient's cooperation can have a lot to do with the results obtained. No matter how objective the measure itself, variability inherent in the patient is bound to be manifest. Active joint range of notion, voluntary muscle performance (particularly endurance), gait speed, and maximal work capacity, although all measurable using objective means, can be influenced strongly by patient effort. The aforementioned tests of our domain can be contrasted with some of those practiced by physicians. For example, blood values (eg, hematocrit, blood urea nitrogen, and creatine kinase) are influenced by the patient only in that the patient must allow the blood for the tests to be drawn. Such medical tests, although further removed from the function in which we have a specific interest, do have distinctive advantages.

Why Do We Need

Objective Measures?

The rationales for objective measures can be categorized as those common to all measures and those that distinguish objective measures from subjective measures. Each category will be addressed in the following paragraphs.

Whether objective or subjective, measures can be used for at least four basic purposes: 1) to describe status, 2) to document change, 3) to explain performance, and 4) to predict outcome. I will use the example of muscle strength to illustrate. Therapists can measure muscle force or torque production to describe a patient's status, say in relation to normative values or the contralateral side. The therapist can remeasure muscle force or torque production to determine whether the patient's strength has increased. Knowing the patient's present muscle force or torque production, the therapist can explain, to a limited extent, the functional performance that a patient will demonstrate. For example, gait, [5,6] rolling, [7] and transfer [8] performance can be explained in part by muscle strength measures in stroke patients. Knowing the strength of some muscles of a stroke patient, his or her walking performance and transfer capacity can also be predicted across time, albeit incompletely. Beyond the basic purposes outlined heretofore, measures are needed for other reasons. Weiss and Bucuvalas summarize numerous such reasons including to clarify the relative advantages of alternative choices, to understand the background and context of program operation, to stimulate review of policy, to focus attention on neglected issues, to provide new understanding, to clarify thinking, to reorder priorities, to make sense of what is being done, to offer ideas for future directions, to reduce uncertainties, to create new uncertainties and provoke rethinking of taken-for-granted assumptions, to provide a sense of how the world works, to justify actions, and to support positions and persuade others. [9] Bohannon has published the results of a study in which productivity measures were used to fulfill the last three purposes. [10]

A need exists for objective measures above and beyond the need for measures in general. This need is tied to the advantages inherent in some objective measures. Before discussing their advantages, a sense of balance requires that I admit that the therapist could conceivably go overboard with objective measures, that objective measures may require more time and effort than subjective measures, and that objective measures require equipment that may elevate their cost and limit the settings in which they can be applied.

On the positive side, objective measures have in some cases three potential advantages over subjective measures: 1) sensitivity, 2) validity, and 3) reliability. Objective measures are more sensitive or precise than subjective measures. This fact has been documented repeatedly in regard to measures of muscle strength. Beasley, using objective strength measures, demonstrated over 30 years ago that subjective manual muscle testing in children produced "Normal" test results when strength was only 50% of Normal. [11] He also demonstrated that differences in strength of 20% to 25% were not detected by manual muscle testing. [11] Griffin et al, [12] Koller and Kase, [13] and Miller et al [14] have reported similar findings. The ramifications of such sensitivity are many. Among the questions addressed by such sensitivity are 1) Is there a deviation from normal and thus should therapy be initiated, continued, or discontinued? and 2) Has a change occurred and thus is there reason to believe that an intervention has been appropriated and is likely to be reimbursed? Improvements in status documented by objective measures can be provided earlier than those provided by subjective measures. Such improvements can be encouraging to the patient and therapist alike. They can also serve to heighten the image of physical therapy in the eyes of those receiving services, reimbursing services, and referring patients for services.

Because they are more sensitive than subjective measures, objective measures can sometimes provide a more valid indication of patient status than measures that are subjective. Three examples will follow. Bobath has claimed that the listing phenomenon in stroke patients is the result of spastic muscles pulling patients toward their paretic side. [15] Using electromyography to objectively document activity of the lateral trunk muscles, Bohannon and colleagues found that the activity of the muscles on the paretic side was always less than on the nonparetic side in listing patients (unreported research, 1987). Thus the objective evidence does not support Bobath's subjective judgment. Similarly, observation of movement in stroke patients has led people to believe that the elbow extensor muscles are more paretic than the elbow flexor muscles after stroke. Quantitative measurements of elbow flexor and extensor muscle force contradict this impression. [16] Yet another example relates to pelvifemoral rhythm. Although some authors recommend measuring hip flexion ROM at the point where palpation reveals that "motion of the femur causes posterior tilting of the pelvis," [17] objective measurements obtained by cinematography demonstrate that posterior tilting of the pelvis begins within 10 degrees of the start of the hip flexion movement. [18]

Perhaps because much of the error contributed by the examiner is eliminated in objective measures, the measures can be more reliable than subjective measures. Observational gait assessment, for example, is known to be associated with limited reliability. [19] Gait speed, on the other hand, has been shown to be a measurement with very high reliability. [20,21]

How Are We Doing with

Objective Measures, and

Where Might We go

from Here?

I would provide personally a mixed review of how we are doing as a profession. On the one hand, an increased interest in objective measures seems to exist among researchers and those concerned with reimbursement. On the other hand, we still lag behind medicine in much of our clinical practice. In that practice, clinicians appear to be quite limited in their use of objective measures. They also seem to lack a knowledge of objective measures and where to find out about them.

An interest in objective measures does appear to exist among some therapists. Physical Therapy and other journals frequently publish articles describing, comparing, and assessing the reliability of objective measures. A recent article examining the realiability of measurements of lumbar lordosis with a flexible ruler is a case in point. [22] I am often confronted by physical therapy department administrators and private practitioners (particularly owners) who are concerned that reimbursement is, or certainly will soon be, dependent on our ability to document change objectively. Additionally, an increasing number of continuing education courses are being offered on the topic of documentation.

Despite the aforementioned positive trends, physical therapy has a limited number of commonly accepted and understood objective measures for which validity and reliability are established. This situation is particularly apparent when our profession is viewed relative to some aspects of medicine. We may even be retreating from the medical model in some areas. There seems to be a growing interest among some in the less objective aspects of our practice. Much of the debate over myofascial techniques has focused on this issue. A casual observation of some clinical settings will reveal a limited use of commonly accepted tools of objective measurement (eg, goniometer). When I interviewed clinicians about how they knew that their patients were improving (unpublished research, 1988), they often indicated that they knew intuitively. Although their intuition may be correct, it is not objective and is not likely to satisfy the demands of those reimbursing services. Many clinicians seem to be unaware of the multitude of methods that can be used to objectively document patient status and change. I have often heard statements such as, "No one has done anything regarding so and so." Yet dozens of articles are often available and provide evidence to the contrary. A longtime physical therapist assistant, who is also a student of mine, commented that he had never seen clinicians use many of the simple clinical measures presented in a recent article. [23] He questioned whether they are being used in any clinics. The results of my doctoral disseration research suggest that clinicians do vigorously seek information, but not in journals, where much of the information about objective measures resides. [24] The clinicians interviewed for the dissertation were able to identify, on the average, less than three journals of use in their practice. This number stands in sharp contrast to the vast number of journals relevant to physical therapy. [25] My doctoral dissertation research showed also that clinicians have a highly limited knowledge of where to look, other than to colleagues, for information. [24] This situation is inseparable from how we are doing with objective measures.

My recommendations for where we might go from here are organized around entry-level education, practice, and research. Although each target is important, improvements in the use of objective measures will probably depend on progress on all fronts.

In entry-level education, students must be exposed to the purposes, value, and variety of objective measures. Educators have an opportunity not only to teach their students objective measurement techniques that are being used clinically but also to encourage the use of new or underutilized objective measures that may be of possible value in the clinic. The curriculum should be updated constantly from the literature to include appropriate information about valid and reliable objective measures. Students can be prepared to be change agents at their future affiliations and sites of employment. One possible way of facilitating this preparation is for professors to promote the value of progress and the literature, not only in a course on research but in all other courses and by other means as well. [26] Clinical educators have a similar opportunity.

In practice environments, much can be done to promote the use of objective measures. A good first step is for clinicians, even at the risk of being socially deviant, to personally begin using or increasingly use such measures. Among the measures that may hold particular promise, but are not routinely used, are quantitative measures of muscle performance and electromyography, [27] quantitative measures of balance and gait speed, [28] and measures of pulse rate. [29] Hamzeh et al have proposed an interesting derivation using the last two measures in combination. [30] Other processes that may assist the adoption of objective measures are the purchase of appropriate instrumentation, the thorough orientation of staff in the use of the instrumentation, continuing staff development (eg, through supervision, the positing of current articles, [31] and journal clubs), and informal reliability studies.

Much remains to be done in regard to research to foster the appropriate use of measurements. Through research, new objective measures need to be identified. Particularly valuable may be measures that bypass some of the variability inherent in patient effort. Such a measure using electrical stimulation was recently described by McDonnell et al. [32] Some commonly advocated measures need to be examined more closely. For example, it has been proposed that the statistical descriptions of the variability of measurements of muscle strength can be used to distinguish true maximal from feigned maximal efforts by patients. Recent research casts some doubt on this proposition. [33] Although theory and hypotheses can play a role in this process, their potentially obstructive capacity should also be realized. [34,35] There remain a large number of objective measures that realize some popularity, the reliability of which is yet unproven. Population-specific reliabilities are particularly lacking. Two aspects of criteria validity seem to me to be of particular importance and worthy of research investment. I believe function is the most important criterion. We need to know whether our various objective measures correlate with function. If they do not, perhaps their clinical use should be questioned. Fortunately, research is now beginning to be published that investigates the relationship between measures of supposedly important variables and function. [36] Objective measures can also serve as criteria for subjective measures. If subjective measures can be shown to be correlated significantly with objective measures, then the reduced time and cost associated with the subjective measures may help justify their use. Research correlating subjective and objective measures has already been performed on some variables (eg, shoulder subluxation, [4] muscle strength [37]).

Conclusion

I have provided little information in this article as to how objective measures can be used, specifically in making clinical decisions. The questions addressed, however, should provide a framework by which the problems associated with clinical decision making can be approached.

References

[1] Rothstein JM: Measurement and clinical practice: Theory and application. In Rothstein JM (ed): Measurement in Physical Therapy: Clinics in Physical Therapy. New York, NY, Churchill Livingstone Inc, 1985, vol 7, pp 41-42

[2] Edwards RHT, McDonnell MK: Hand-held dynamometer for evaluating voluntary muscle function. Lancet 2:757-758, 1974

[3] Bohannon RW: Relationship between shoulder pain and selected variables in patients with hemiplegia. Clinical Rehabilitation 2:111-117, 1988

[4] Prevost R, Arsenault AB, Dutil E, et al: Shoulder subluxation in hemiplegia: A radiologic correlational study. Arch Phys Med Rehabil 68:782-785, 1987

[5] Nakamura R, Hosokawa J, Tsuji I: Relationship of muscle strength of knee extension to walking capacity in patients with spastic hemiparesis. Tohoku J Exp Med 145:335-340, 1985

[6] Bohannon RW: Selected determinants of ambulatory capacity in patients with hemiplegia. Clinical Rehabilitation 3:47-53, 1989

[7] Bohannon RW: Rolling to the nonplegic side: Influence of teaching and limb strength in hemiplegic stroke patients. Clinical Rehabilitation 2:215-218, 1988

[8] Bohannon RW: Determinants of transfer capacity in patients with hemiparesis. Physiotherapy Canada 40:236-239, 1988

[9] Weiss CH, Bucuvalas MJ: Truth tests and utility tests: Decision-makers' frames of reference for social sciences research. American Sociological Review 45:302-313, 1980

[10] Bohannon RW: Statistical analysis of productivity in one physical therapy department. Phys Ther 67:1553-1557, 1987

[11] Beasley WC: Influence of method on estimates of normal knee extensor force among normal and postpolio children. Phys Ther Rev 36:21-41, 1956

[12] Griffin JW, McClure MH, Bertorini TE: Sequential isokinetic and manual muscle testing in patients with neuromuscular disease: A pilot study. Phys Ther 66:32-35, 1986

[13] Koller W, Kase S: Muscle strength testing in Parkinson's disease. Eur Neurol 25:135-136, 1986

[14] Miller LC, Michael AF, Baxter TL, et al: Quantitative muscle testing in childhood dermatomyositis. Arch Phys Med Rehabil 69:610-613, 1988

[15] Bobath B: Adult Hemiplegia: Evaluation and Treatment, rev ed 2. London, England, William Heinemann Medical Books Ltd, 1978, pp 93-98

[16] Colebatch JG, Gandevia SC, Spira PJ: Voluntary muscle strength in hemiparesis: Distribution of weakness at the elbow. J Neurol Neurosurg Psychiatry 49:1019-1024, 1986

[17] Norkin CC, White DJ: Measurement of Joint Motion: A Guide to Goniometry. Philadelphia, PA, F A Davis Co, 1985, p 76

[18] Bohannon RW, Gajdosik RL, LeVeau BF: Relationship of pelvic and thigh motions during unilateral and bilateral hip flexion. Phys Ther 65:1501-1504, 1985

[19] Goodkin R, Diller L: Reliability among physical therapists in diagnosis and treatment of gait deviations in hemiplegics. Percept Mot Skills 37:727-734, 1973

[20] Holden MK, Gill KM, Magliozzi MR, et al: Clinical gait assessment in the neurologically impaired: Reliability and meaningfulness. Phys Ther 64:35-40, 1984

[21] Bohannon RW: Velocity and cadence of gait in hemiparetic stroke patients. Int J Rehabil Res 10:73-75, 1987

[22] Lovell FW, Rothstein JM, Personius WJ: Reliability of clinical measurements of lumbar lordosis taken with a flexible rule. Phys Ther 69:96-102, 1989

[23] Bohannon RW: Simple clinical measures. Phys Ther 67:1845-1850, 1987

[24] Bohannon RW: Information-Accessing Behavior of Physical Therapists. Doctoral Dissertation. Raleigh, NC, North Carolina State University, 1988

[25] Bohannon RW: Core journals of physiotherapy. Physiotherapy Practice 3:126-128, 1987

[26] Bohannon RW, LeVeau BF: Clinicians' use of research findings: A review of literature with implications for physical therapists. Phys Ther 66:45-40, 1986

[27] Wolf SL, Edwards DI, Shutter LA: Concurrent assessment of muscle activity (CAMA): A procedural approach to assess treatment goals. Phys Ther 66:218-224, 1986

[28] Mizrahi J, Susak Z, Heller L, et al: Objective expression of gait improvement of hemiplegics during rehabilitation by time-distance parameters of stride. Med Biol Eng Comput 20:628-634, 1982

[29] Davies JB: Use of heart rate in assessment of orthoses. Physiotherapy 63:112-115, 1977

[30] Hamzeh MA, Bowker P, Sayegh A: The energy costs of ambulation using two types of walking frame. Clinical Rehabilitation 2:119-123, 1988

[31] Bohannon RW, Larkin PA: Current journal article provision for the physical therapy clinicians of one department. Phys Ther 66:689-690, 1986

[32] McDonnell MK, Delitto A, Sinacore DR, et al: Electrically elicited fatigue test of the quadriceps femoris muscle: Description and reliability. Phys Ther 67:941-945, 1987

[33] Bohannon RW, Smith MB: Differentiation of maximal and submaximal knee extension efforts by isokinetic testing. Clinical Biomechanics 3:215-218, 1988

[34] Greenwald AG, Leippe MR, Pratkanis AR, et al: Under what conditions does theory obstruct research progress? Psychol Rev 93:216-229, 1986

[35] Winter DA: Are hypotheses really necessary in motor control research? Journal of Motor Behavior 19:276-279, 1987

[36] Spaulding SJ, Stachota E, McPherson JJ, et al: Wrist muscle tone and self-care skill in persons with hemiparesis. Am J Occup Ther 43:11-16, 1989

[37] Bohannon RW: Manual muscle test scores and dynamometer test scores of knee extension strength. Arch Phys Med Rehabil 67:390-392, 1986

R Bohannon, EdD, PT, is Associate Professor, Program in Physical Therapy, School of Allied Health Professions, University of Connecticut, 358 Mansfield Rd, PO Box U-101, Storrs, CT 06269-2101 (USA).