- Email: [email protected]
Inter- and intrarater reliability of the Ashworth Scale and the Disability Assessment Scale in patients with upper-limb poststroke spasticity
1349
Inter- and Intrarater Reliability of the Ashworth Scale and the Disability Assessment Scale in Patients With Upper-Limb Poststroke Spasticity Allison Brashear, MD, Ross Zafonte, DO, Michael Corcoran, MD, Nestor Galvez-Jimenez, MD, Jean-Michel Gracies, MD, PhD, Mark Forrest Gordon, MD, Anita Mcafee, OT, Kyle Ruffing, MD, Barbara Thompson, OT, Michael Williams, PT, Chia-Ho Lee, MS, Catherine Turkel, PharmD ABSTRACT. Brashear A, Zafonte R, Corcoran M, GalvezJimenez N, Gracies J-M, Gordon MF, Mcafee A, Ruffing K, Thompson B, Williams M, Lee C-H, Turkel C. Inter- and intrarater reliability of the Ashworth Scale and the Disability Assessment Scale in patients with upper-limb poststroke spasticity. Arch Phys Med Rehabil 2002;83:1349-54. Objective: To evaluate the reliability of the Ashworth Scale and the Disability Assessment Scale (DAS) in poststroke patients with upper-limb spasticity and functional disability. Design: Single-center trial. Setting: University medical center. Participants: Nine patients ⱖ6 months poststroke with upper-limb spasticity and impairment in the areas of hygiene, dressing, limb posture, or pain were included in the analysis. Interventions: Ten experienced medical professionals rated each patient in randomized order twice on the same day (results based on mean of evaluations at times 1 and 2). Elbow, wrist, finger, and thumb flexion tones were assessed by using the Ashworth score (range, 0 – 4), and functional disability was assessed using the DAS (range, 0 –3). Main Outcome Measures: Intra- and interrater reliability of the Ashworth Scale and DAS. Results: For the Ashworth parameters, 38 of 40 evaluations indicated excellent (weighted ⱖ.75) or good (weighted ⱖ.4) intrarater reliability. For DAS parameters, 31 of 40 evaluations indicated excellent or good intrarater reliability. The interrater reliability was also good for both the Ashworth Scale (Kendall W⫽.598 –.792) and DAS (Kendall W⫽.494 –.772) with statistically significant agreement found among raters (all P⬍.001). Conclusions: In patients with upper-limb spasticity after stroke, the Ashworth Scale and DAS had good intra- and interrater reliability when used by trained medical professions. Key Words: Cerebrovascular accident; Muscle spasticity; Outcome assessment (health care); Rehabilitation. © 2002 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
From the Indiana University School of Medicine, Indianapolis, IN (Brashear, Mcafee); University of Pittsburgh, PA (Zafonte); University of Maryland Medical System, Kernan Hospital, Baltimore, MD (Corcoran); Cleveland Clinic Florida, Weston, FL (Galvez-Jimenez); Mount Sinai School of Medicine, New York, NY (Gracies); Long Island Jewish Medical Center, New Hyde Park, NY (Gordon); Indiana University Hospital, Indianapolis, IN (Mcafee, Thompson, Williams); Neurological Research Institute of Sarasota, FL (Ruffing); and Allergan Inc, Irvine, CA (Lee, Turkel). Supported by Allergan Inc (grant no. 45-843-14). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Allison Brashear, MD, Indiana University Medical Ctr, 550 University Blvd, Rm 6620, Indianapolis, IN 46202-5167, e-mail: [email protected]. 0003-9993/02/8310-7068$35.00/0 doi:10.1053/apmr.2002.35474
LINICAL ASSESSMENTS ARE essential to guide treatment decisions and to measure progress in patients with C spasticity because of spinal cord injury (SCI), stroke, multiple sclerosis, or other neurologic causes. Although biomechanic and electrophysiologic methods have been developed that provide quantitative and objective assessment of muscle tone and reflex activity,1-4 these techniques require sophisticated equipment and are time-consuming to administer. In the early 1960s, Bryan Ashworth devised a simple 5-point scale that enabled clinicians to quantify the amount of muscle resistance encountered during passive movements.5 Although the Ashworth Scale has been widely used as a method for measuring the effects of therapeutic interventions on spastic hypertonia in clinical trials,2,6 its reliability remains questionable. Despite its accepted use in clinical practice, the Ashworth Scale provides no information about the degree of the patient’s functional impairment. To address the need for a simple, reliable, clinical method for objectively evaluating functional disability in patients with spasticity, a Disability Assessment Scale (DAS) was developed to assess functional impairment commonly seen in patients with poststroke upper-limb spasticity (ie, dressing, hygiene, limb position, pain). The current study was conducted to evaluate the inter- and intrarater reliability of the Ashworth Scale and the DAS in the assessment of upper-limb tone and functional disability, respectively, in poststroke patients. METHODS Participants Ten patients with upper-limb spasticity were recruited for this single-center trial. Patients were required to be at least 21 years of age; to have a documented history of stroke that resulted in spasticity of the elbow, wrist, finger, and/or thumb flexors in at least 1 limb; and to have some degree of impairment pertaining to hygiene, dressing, limb position, or pain. The protocol excluded patients who had experienced a stroke within 6 months of study enrollment, had a cast on the affected upper limb, had an intrathecal baclofen pump, or had received oral antispasticity medications (eg, baclofen, tizanidine, benzodiazepines, carisoprodol, cyclobenzaprine, methocarbamol) within 30 days of enrollment. This study was conducted in compliance with the ethical standards of good clinical practices, institutional review board regulation, informed consent regulations, and the current Helsinki Declaration.7 Raters The 10 independent raters were health care professionals from a variety of disciplines (physiatry, neurology, physical therapy, occupational therapy) who had specific experience in the treatment of patients with upper-limb spasticity poststroke. Before study initiation, all participating raters received simultaneous, standardized training on the use of the Ashworth Scale Arch Phys Med Rehabil Vol 83, October 2002
1350
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear
and the DAS. On the morning of the study, the raters attended a training session during which they were given detailed instructions regarding the evaluations and the overall study. This training provided guidance to investigators on assessment of muscle tone (Ashworth) and functional ability (DAS). Each rater was provided with a laminated card on which the scales were printed. Design Over a 1-day period, each rater assessed each subject twice while subjects were seated comfortably. The first assessment (evaluation 1) took place in the morning and the second assessment by the same rater (evaluation 2) took place in the afternoon. The 2 evaluations of the same subject took place at least 1 hour apart. All raters assessed each patient, and there was a 15-minute rest interval between every evaluation. To minimize the possibility of investigator bias during evaluations, patients were randomly assigned to 1 of 10 clinical examination rooms on the morning of the study day for evaluation 1 and rerandomized to a different room in the afternoon for evaluation 2. The raters were issued randomized lists indicating the order (by examination room number) in which they were to assess the patients for each evaluation. Thus, the patients were assessed in a different order for evaluation 1 and evaluation 2. The patients and raters were instructed not to discuss the results of the evaluations with each other or with other patients or raters during the study. Outcome Measures The primary outcome measures were the extent of agreement among all raters (interrater reliability) and the extent of agreement between each rater’s 2 evaluations (intrarater reliability) for the Ashworth Scale and the DAS. Ashworth Scale. The Ashworth Scale measures resistance to passive movement according to the following scale5: 0 means no increase in tone (none); 1 means slight increase in tone, giving a catch when the limb is moved in flexion or extension (mild); 2 means more marked increase in tone but limb is easily flexed (moderate); 3 means considerable increase in muscle tone (passive movement difficult [severe]); and 4 means limb rigid in flexion or extension (very severe). Ashworth scores were obtained during flexion of the elbow, wrist, fingers, and thumb, in that order, according to the techniques described later. Raters were instructed to perform each assessment as similarly as possible for each patient. Elbow. With the patient’s arm extended as much as possible and with the palm of the hand facing inward (neutral supination), the rater extended the forearm from maximum possible flexion to maximum possible extension no more than 3 consecutive times and rated flexion muscle tone according to the Ashworth Scale. Wrist. With the patient’s elbow as straight as possible and the forearm pronated so that the palm of the hand faced downward, the rater moved the wrist from maximum possible flexion to maximum possible extension no more than 3 consecutive times and rated flexion muscle tone according to the Ashworth Scale. Fingers. With the patient’s elbow as straight as possible, the forearm pronated so that the palm of the hand was facing inward, the patient’s wrist in neutral position, the rater opened and closed the fingers no more than 3 consecutive times and rated flexion muscle tone according to the Ashworth Scale. Thumb. With the patient’s elbow as straight as possible, the forearm pronated so that the palm of the hand was facing inward and the patient’s wrist in neutral position at 0°, the rater
Arch Phys Med Rehabil Vol 83, October 2002
raised and lowered the thumb from maximum possible flexion to maximum possible extension no more than 3 consecutive times and rated flexion muscle tone according to the Ashworth Scale. A mean Ashworth value for thumb flexion was recorded. Disability Assessment Scale. The raters interviewed each patient to determine the extent of functional impairment for patient hygiene, dressing, limb position, and pain according to the following scale: 0 means no disability, 1 means mild disability (noticeable but does not interfere significantly with normal activities), 2 means moderate disability (normal activities require increased effort and/or assistance), and 3 means severe disability (normal activities limited). Assessment of the 4 functional domains was performed according to the following guidelines. Hygiene. The rater assessed the extent of maceration, ulceration, and/or palmar infection; palm and hand cleanliness; ease of cleanliness; ease of nail trimming; and the degree of interference caused by hygiene-related disability in the patient’s daily life. Dressing. The rater assessed the difficulty or ease with which the patient could put on clothing (eg, shirts, jackets, gloves) and the degree of interference caused by dressingrelated disability in the patient’s daily life. Limb position. The rater assessed the amount of abnormal position of the upper limb. Pain. The rater assessed the intensity of pain or discomfort related to upper-limb spasticity. Statistical Analysis Interrater agreement analyses were performed for all raters. The extent of agreement was analyzed by using the Kendall W statistic, or coefficient of concordance,8 for each parameter rated on each of the 2 scales.a The null hypothesis was that there would be no agreement between raters’ rankings of the patients. If the P value was ⱕ.05, then the null hypothesis was rejected, and the level of agreement was determined by the magnitude of the Kendall W statistic. A 95% confidence interval (CI) was constructed to confirm the hypothesis testing of association. Intrarater agreement analyses were performed for all raters who evaluated patients at both evaluations. A weighted statistic was calculated assessing the agreement between the 2 ratings for each rater, for each parameter measured on each scale. For each parameter, an overall weighted statistic across all raters was estimated with a 95% CI. The null hypothesis was to test that each individual rater’s weighted value for any given parameter was equal to zero. The weighted statistic was tested using a chi-square test.9 The level of intrarater agreement was determined by the magnitude of the overall weighted statistic. Subgroup analyses were also performed according to the raters’ medical specialty or profession. RESULTS Patient Characteristics Ten patients were screened, entered the study, and completed both evaluations. One patient was reported as having a major protocol deviation (ie, he had taken a dose of antispasticity medication on the morning of the study day). Data for this patient were not included in the analysis. Demographic characteristics of the study population are provided in table 1. Of the 9 patients included in the analysis, 5 were men and 4 were women. The mean age was 60 years
1351
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear Table 3: Mean Ashworth Scores (Nⴝ9)
Table 1: Patients’ Demographic Characteristics (Nⴝ9) Age (y) Mean ⫾ SD Median Range Gender Men Women Race White Black Most common events in medical history Stroke/CVA Spasticity Allergies Depression Hypertension Myocardial infarction
59.9⫾16.17 69 37–81 5 (55.6%) 4 (44.4%) 6 (66.7%) 3 (33.3%) 9 (100%) 9 (100%) 5 (55.6%) 4 (44.4%) 4 (44.4%) 3 (33.3%)
Abbreviation: SD, standard deviation; CVA, cerebrovascular accident.
(range, 37– 81y); 6 patients (67%) were white and 3 (33%) were black. Rater Characteristics Demographic and professional characteristics of the raters are presented in table 2. Of the 10 raters, 4 were neurologists, 2 were physiatrists, 2 were occupational therapists, and 2 were physical therapists. The mean overall experience in their respective professions was 6.6 years (range, 1–18y). Evaluations The mean time between evaluation 1 and evaluation 2 was 3 hours 42 minutes (range, 1h 29min to 6h 15min). The median time was 3 hours 45 minutes. Ashworth Scale A summary of the statistics for each parameter assessed by the Ashworth is presented in table 3. Interrater reliability. The Kendall W statistic and 95% CI for interrater agreement were determined by each parameter for evaluation 1, evaluation 2, and the mean of both evaluations. The results showed significant (P⬍.001) agreement among raters for all parameters of the Ashworth Scale (table 4).
Evaluation 1 Evaluation 2 Mean of evaluations 1 and 2
Elbow
Wrist
Fingers
Thumb
1.8⫾.67 1.9⫾.67
1.7⫾.70 1.6⫾.64
2.2⫾.82 2.1⫾.83
1.3⫾.69 1.4⫾.75
1.9⫾.66
1.6⫾.66
2.1⫾.82
1.4⫾.71
NOTE. Values are mean ⫾ SD.
Intrarater reliability. Weighted statistics for intrarater agreement (table 5) indicated either good (weighted ⱖ.4) or excellent reliability (weighted ⱖ.75) for 95% (38 of 40) of the ratings on the Ashworth Scale. The overall weighted values for elbow, wrist, fingers, and thumb ranged from .67 to .74. The null hypothesis was not rejected for any measured parameter. Therefore, the overall weighted values did not differ across raters (Pⱖ.972). Disability Assessment Scale The summary of the statistics for each parameter assessed by the DAS is presented in table 6. Interrater reliability. For all of the DAS parameters (hygiene, dressing, limb position, pain), there was good agreement among raters for the measurements at evaluation 1, evaluation 2, and the mean of both evaluations (table 7). The null hypothesis that there was no agreement among raters on the ratings of the patients was rejected. Therefore, the results showed significant (P⬍.001) agreement among raters for all parameters. Intrarater reliability. For the DAS parameters, 78% (31/ 40) of the evaluations indicated either good intrarater reliability (weighted ⱖ.4) or excellent intrarater reliability (weighted ⱖ.75) (table 8). For all domains measured (hygiene, dressing, limb position, pain), the overall weighted values ranged from .52 to .78. Weighted values did not differ across raters (Pⱖ.478). The chi-square test, applied to test the null hypothesis, yielded P values greater than .05. Therefore, the null hypothesis that the weighted values by the 10 raters were equal for each of parameters rated was not rejected for any parameter. Subgroup Analysis Analyses of inter- and intrarater agreement were performed in subgroups defined by the profession of the rater (ie, neurolTable 4: Ashworth Interrater Agreement Elbow
Table 2: Raters’ Characteristics (Nⴝ10) Age (y) Mean ⫾ SD Median Range Profession Neurologist Physiatrist Occupational therapist Physical therapist Experience in profession (y) Mean ⫾ SD Median Range
37.5⫾4.53 39 29–42 4 (40%) 2 (20%) 2 (20%) 2 (20%) 6.6⫾4.99 6 1–18
Wrist
Fingers
Thumb
Evaluation 1 Kendall W .659 .551 .752 .490 95% CI* .312–1.000 .261–1.000 .356–1.000 .232–1.000 P† ⬍.001 ⬍.001 ⬍.001 ⬍.001 Evaluation 2 Kendall W .616 .507 .692 .625 95% CI .292–1.000 .240–1.000 .328–1.000 .296–1.000 P ⬍.001 ⬍.001 ⬍.001 ⬍.001 Mean of evaluations 1 and 2 Kendall W .765 .598 .792 .611 95% CI .363–1.000 .283–1.000 .375–1.000 .289–1.000 P ⬍.001 ⬍.001 ⬍.001 ⬍.001 * Based on chi-square approximation. Monte Carlo estimate of exact P value testing the Kendall W equal to zero.
†
Arch Phys Med Rehabil Vol 83, October 2002
1352
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear Table 7: DAS Interrater Agreement
Table 5: Ashworth Scale Intrarater Agreement Overall Weighted Statistic for Parameter Rater No.
Elbow
Wrist
Fingers*
Thumb*
1 2 3 4 5 6 7 8 9 10 Overall weighted 95% CI† P‡
.447 .526 .729 .651 .805 .462 .792 .458 .649 .843 .668 .436–.901 .998
.548 .300 .868 .550 .907 .906 .551 .599 .840 .651 .740 .530–.949 .972
.658 .770 .722 .897 .673 .615 .772 .758 .676 1.000 .740 .516–.964 1.000
.500 .509 .883 .401 .419 .653 .813 .654 .674 .000 .680 .435–.924 .985
* statistic for 1 rater with standard error (SE) equal to zero is not included in the computation for overall weighted , 95% CI, and chi-square test. † Based on normal approximation. ‡ From chi-square test for equality of statistics across raters.
ogist, physiatrist, occupational therapist, physical therapist) for each of the scales. The results of these subgroup analyses were similar to those of all raters combined, with good inter- and intrarater agreement for most parameters. DISCUSSION Quantification of muscle tone and associated impairment remains a clinical challenge due to the intricate relationship of this feature with other complications of the upper motoneuron syndrome (eg, muscle contracture, spastic dystonia).10 Over the past several years, numerous methods have been developed to provide information about the resistance of the spastic limb to passive joint movement.3,11 Although some of these methods have proven useful for assessing functional outcomes of patients with spasticity,2,12 none was designed specifically to evaluate function in patients with spasticity after stroke. The results of this study confirm the value of the DAS and the Ashworth Scale, which, when used together, provide complementary information regarding functional outcomes and muscle tone in patients with functional disability caused by upperlimb spasticity poststroke. Although recent efforts have been aimed at developing clinical assessment tools with wider applications than existing ones (eg, the Tone Assessment Scale),13 the Ashworth Scale and the Modified Ashworth Scale (MAS) remain the primary clinical measures of tone.14 Of the 2, the original Ashworth Scale has superior interrater reliability and has been validated as an ordinal-level measurement of resistance to passive movement.14 Only 1 other study15 has assessed the reliability of the
Hygiene
Evaluation 1 Evaluation 2 Mean of evaluations 1 and 2
Dressing
Limb Position
Pain
0.9⫾.52 0.8⫾.53
2.2⫾.53 2.3⫾.50
0.7⫾.71 0.6⫾.59
0.5⫾.69 0.4⫾.63
2.2⫾.48
NOTE. Values are mean ⫾ SD.
Arch Phys Med Rehabil Vol 83, October 2002
0.7⫾.64
0.5⫾.65
Pain
* Based on chi-square approximation. Monte Carlo estimate of exact P value testing the Kendall W equal to zero.
†
Ashworth Scale for use in the upper limbs (excluding studies that assessed the MAS).15 This study examined the intra- and interrater reliability of summated Ashworth scores for a variety of different upper-limb muscles. Similar to the present study, the authors found a good correlation among the 4 raters (Kendall W, .92) and concluded that the Ashworth Scale was a reliable method for evaluating upper-limb muscle tone. To assess the reliability of the Ashworth Scale and DAS across various health care professionals, the present study included 10 independent raters from different professions who were experienced in treating poststroke patients. This relatively large number of raters is an improvement over several previous studies13,16,17 that assessed the reliability of the Ashworth Scale and/or MAS using only 2 raters. In the present study, evaluations at times 1 and 2 were separated by at least an hour (avg time, 3h 42min; range, 1h 29min to 6h 15min). This interval was designed to counteract Table 8: DAS Intrarater Agreement Overall Weighted Statistic for Parameter Rater No.
Hygiene
Dressing*
Limb Position
Pain†
1 2 3 4 5 6 7 8 9 10 Overall weighted 95% CI‡ P㛳
.550 .400 .182 .781 .182 .526 .682 .400 .444 .727 .520 .239–.802 .994
.897 .716 .769 .307 .800 ⫺.052 .781 .379 .079 1.000 .530 .278–.782 .478
.625 .719 .352 .763 .809 .888 .769 .800 .526 .781 .775 .560–991 .998
.000 .727 .794 .308 .880 1.000 .880 .824 .700 .817 .776 .533–1.000 .992
* statistic for 1 rater with SE equal to zero is not included in the computation for overall weighted , 95% CI, and chi-square test. statistics for 2 raters with SE equal to zero are not included in the computation for overall weighted , 95% CI, and chi-square test. ‡ Based on normal approximation. 㛳 From chi-square test for equality of statistics across raters. †
0.9⫾.51
Limb Position
Evaluation 1 Kendall W .527 .477 .587 .695 95% CI* .250–1.000 .226–1.000 .278–1.000 .329–1.000 P† ⬍.001 ⬍.001 ⬍.001 ⬍.001 Evaluation 2 Kendall W .544 .557 .395 .726 95% CI .258–1.000 .264–1.000 .187–1.000 .344–1.000 P ⬍.001 ⬍.001 ⬍.001 ⬍.001 Mean of evaluations 1 and 2 Kendall W .626 .494 .557 .772 95% CI .297–1.000 .234–1.000 .264–1.000 .366–1.000 P ⬍.001 ⬍.001 ⬍.001 ⬍.001
Table 6: Mean Scores on the DAS (Nⴝ9) Hygiene
Dressing
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear
the effects of memory on the ratings (ie, at evaluation 2, it would be more difficult for a rater to remember how he/she scored a patient at evaluation 1). This interval was also meant to counteract the effects of repeated muscular stretching of patients’ limbs, although such effects should have been minimal because raters stretched each patient a maximum of 3 times in each joint position. If anything, this interval has contributed to an underestimation of Ashworth intrarater reliability in this study because fluctuations of tone during the day are well established.18 A recent review14 of the properties and limitations of the Ashworth Scale noted that documentation of changes in function are a relevant complement to the information provided by this scale. The use of functional outcome measures in tandem with measurements of muscle tone reflect the fact that the chief criterion for therapeutic intervention is functional loss caused by spasticity-induced pain, resistance to voluntary muscle control, and/or uncomfortable or undesired limb positions.12 Moreover, in patients with spasticity after stroke, function has been shown to correlate significantly (P⬍.01) with clinically assessed muscle strength.19 Therefore, functional outcome measurements are an essential component of clinicians’ efforts to meet the treatment goal of improved function in patients with spasticity. Together with measurements of muscle tone, these methods can provide complementary information for the management of individual patients with spasticity and for the evaluation of therapeutic interventions. Several methods have been developed to evaluate the functional impact of spasticity, but none of these correlate functional measures with those of increased tone as the Ashworth Scale does. One of the most prominent among these is the Fugl-Meyer Scale, which has been shown to be an accurate and objective method of describing the return of function after hemiplegia.2 In addition to evaluating movement within and independent of synergistic patterns, this instrument includes critical sensory observations of factors that contribute to motor function, such as position sense, pain-free movement, and joint movement.2 The Barthel Index has been shown to be a useful method of measuring activities of daily living (including feeding, grooming, using stairs, dressing) in the general population of people with disabilities.20 In disabled children and infants, the Pediatric Evaluation of Disability Inventory is a reliable, validated method of measuring both capability and performance of functional activities in 3 domains (self-care, mobility, social function).12,21 The Patient Evaluation Conference System represents an interdisciplinary approach to the evaluation of functional disabilities related to spasticity in patients with brain injury; it has been shown to be a sensitive, reliable, and valid method of assessment.22 Recently, a new functional assessment scale, the Valutazione Funzionale Mielolesi, has been validated for patients with SCI.23 Although all of these scales focus on measuring functional disability, none has been studied in patients with focal spasticity after stroke. In contrast, the DAS was specifically designed for use with this population. It is important to note that the present study addressed only the reliability of the Ashworth Scale and the DAS; we did not evaluate their validity. However, in a separate study,24 we assessed the relationship of several efficacy variables with DAS scores for the domain identified as the primary area of impairment (ie, patients and physicians selected either hygiene, dressing, limb position, or pain as their primary therapeutic intervention target). Results showed that DAS scores for this domain correlated with the Ashworth composite muscle tone score (sum of wrist and finger Ashworth scores) (r⫽.608), physician’s global assessment score (r⫽⫺.456), and patient’s
1353
global assessment score (r⫽⫺.506) at week 6 after injection with botulinum toxin type A or placebo. Additional studies will be useful to define the relationship between the DAS and burden of care, and also the utility of the DAS in patients with spasticity due to noncerebrovascular etiologies. CONCLUSION This study showed that the Ashworth Scale and the DAS have good intra- and interrater reliability when used by trained medical professionals to rate excessive tone and functional impairment in patients with upper-limb spasticity after stroke. These scales can be used readily in the clinical setting. References 1. Young RR, Wiegner AW. Spasticity. Clin Orthop 1987;Jun(219): 50-62. 2. Katz RT, Rymer WZ. Spastic hypertonia: mechanisms and measurement. Arch Phys Med Rehabil 1989;70:144-55. 3. Bajd T, Bowman B. Testing and modeling of spasticity. J Biomed Eng 1982;4:90-6. 4. Bajd T, Bowman B. Pendulum testing of spasticity. J Biomed Eng 1984;6:9-16. 5. Ashworth B. Preliminary trial of carisoprodol in multiple sclerosis. Practitioner 1964;192:540-2. 6. Alfieri V. Electrical treatment of spasticity: reflex tonic activity in hemiplegic patients and selected specific electrostimulation. Scand J Rehabil Med 1982;14:177-82. 7. World Medical Assembly. The Declaration of Helsinki: recommendations guiding physicians in biomedical research involving human patients. World Med J 1982;29:86-8. 8. Siegel S, Castellan N. Nonparametric statistics for the behavioral sciences. 2nd ed. New York: McGraw-Hill; 1988. p 284-91. 9. Fleiss JL. Statistical methods of rates and proportions. 2nd ed. New York: Wiley & Sons; 1981. p 38-46. 10. Lance JW. The control of muscle tone, reflexes, and movement: Robert Wartenberg Lecture. Neurology 1980;30:1303-13. 11. Gracies JM, Marosszeky JE, Renton R, Sandanam J, Gandevia SC, Burke D. Short-term effects of dynamic lycra splints on upper limb in hemiplegic patients. Arch Phys Med Rehabil 2000;81: 1547-55. 12. Hinderer SR, Gupta S. Functional outcome measures to assess interventions for spasticity. Arch Phys Med Rehabil 1996;77: 1083-9. 13. Gregson JM, Leathley M, Moore AP, Sharma AK, Smith TL, Watkins CL. Reliability of the Tone Assessment Scale and the modified Ashworth Scale as clinical tools for assessing poststroke spasticity. Arch Phys Med Rehabil 1999;80:1013-6. 14. Pandyan AD, Price CIM, Barnes MP, Rodgers H. A review of the properties and limitations of the Ashworth and modified Ashworth Scales as measures of spasticity. Clin Rehabil 1999;13:373-83. 15. Lee K-C, Carson L, Kinnin E, Patterson V. The Ashworth Scale: a reliable and reproducible method of measuring spasticity. J Neuro Rehabil 1989;3:205-9. 16. Bohannon RW, Smith MB. Interrater reliability of a modified Ashworth scale of muscle spasticity. Phys Ther 1987;67:206-7. 17. Haas BM, Bergstom E, Jamous A, Bennie A. The interrater reliability of the original and of the modified Ashworth scale for the assessment of spasticity in patients with spinal cord injury. Spinal Cord 1996;34:560-4. 18. Skold C. Spasticity in spinal cord injury: self- and clinically rated intrinsic fluctuations and intervention-induced changes. Arch Phys Med Rehabil 2000;81:144-9. 19. Bohannon RW, Warren ME, Cogman KA. Motor variables correlated with the hand-to-mouth maneuver in stroke patients. Arch Phys Med Rehabil 1991;72:682-4. 20. Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J 1965;14:61-5. 21. Feldman AB, Haley SM, Coryell J. Concurrent and construct validity of the Pediatric Evaluation of Disability Inventory. Phys Ther 1990;70:602-10.
Arch Phys Med Rehabil Vol 83, October 2002
1354
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear
22. Harvey RF, Jellinek HM. Functional performance assessment: a program approach. Arch Phys Med Rehabil 1981;62:456-60. 23. Taricco M, Apolone G, Colombo C, Filardo G, Telaro E, Liberati A. Functional status in patients with spinal cord injury: a new standardized measurement scale. Gruppo Interdisciplinare Valutazione Interventi Riabilitativi. Arch Phys Med Rehabil 2000;81: 1173-80. 24. Brashear A, Gordon MF, Elovic E, et al. A multicenter, doubleblind, randomized, placebo-controlled, parallel study of the
Arch Phys Med Rehabil Vol 83, October 2002
safety and efficacy of Botox (botulinum toxin type A) purified neurotoxin complex in the treatment of focal upper limb spasticity post-stroke [abstract]. Neurology 2001;56 Suppl 3:A78. Supplier a. StatXact associated with SAS; Cytel Software Corp, 675 Massachusetts Ave, Cambridge, MA 02139.
Inter- and Intrarater Reliability of the Ashworth Scale and the Disability Assessment Scale in Patients With Upper-Limb Poststroke Spasticity Allison Brashear, MD, Ross Zafonte, DO, Michael Corcoran, MD, Nestor Galvez-Jimenez, MD, Jean-Michel Gracies, MD, PhD, Mark Forrest Gordon, MD, Anita Mcafee, OT, Kyle Ruffing, MD, Barbara Thompson, OT, Michael Williams, PT, Chia-Ho Lee, MS, Catherine Turkel, PharmD ABSTRACT. Brashear A, Zafonte R, Corcoran M, GalvezJimenez N, Gracies J-M, Gordon MF, Mcafee A, Ruffing K, Thompson B, Williams M, Lee C-H, Turkel C. Inter- and intrarater reliability of the Ashworth Scale and the Disability Assessment Scale in patients with upper-limb poststroke spasticity. Arch Phys Med Rehabil 2002;83:1349-54. Objective: To evaluate the reliability of the Ashworth Scale and the Disability Assessment Scale (DAS) in poststroke patients with upper-limb spasticity and functional disability. Design: Single-center trial. Setting: University medical center. Participants: Nine patients ⱖ6 months poststroke with upper-limb spasticity and impairment in the areas of hygiene, dressing, limb posture, or pain were included in the analysis. Interventions: Ten experienced medical professionals rated each patient in randomized order twice on the same day (results based on mean of evaluations at times 1 and 2). Elbow, wrist, finger, and thumb flexion tones were assessed by using the Ashworth score (range, 0 – 4), and functional disability was assessed using the DAS (range, 0 –3). Main Outcome Measures: Intra- and interrater reliability of the Ashworth Scale and DAS. Results: For the Ashworth parameters, 38 of 40 evaluations indicated excellent (weighted ⱖ.75) or good (weighted ⱖ.4) intrarater reliability. For DAS parameters, 31 of 40 evaluations indicated excellent or good intrarater reliability. The interrater reliability was also good for both the Ashworth Scale (Kendall W⫽.598 –.792) and DAS (Kendall W⫽.494 –.772) with statistically significant agreement found among raters (all P⬍.001). Conclusions: In patients with upper-limb spasticity after stroke, the Ashworth Scale and DAS had good intra- and interrater reliability when used by trained medical professions. Key Words: Cerebrovascular accident; Muscle spasticity; Outcome assessment (health care); Rehabilitation. © 2002 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
From the Indiana University School of Medicine, Indianapolis, IN (Brashear, Mcafee); University of Pittsburgh, PA (Zafonte); University of Maryland Medical System, Kernan Hospital, Baltimore, MD (Corcoran); Cleveland Clinic Florida, Weston, FL (Galvez-Jimenez); Mount Sinai School of Medicine, New York, NY (Gracies); Long Island Jewish Medical Center, New Hyde Park, NY (Gordon); Indiana University Hospital, Indianapolis, IN (Mcafee, Thompson, Williams); Neurological Research Institute of Sarasota, FL (Ruffing); and Allergan Inc, Irvine, CA (Lee, Turkel). Supported by Allergan Inc (grant no. 45-843-14). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Allison Brashear, MD, Indiana University Medical Ctr, 550 University Blvd, Rm 6620, Indianapolis, IN 46202-5167, e-mail: [email protected]. 0003-9993/02/8310-7068$35.00/0 doi:10.1053/apmr.2002.35474
LINICAL ASSESSMENTS ARE essential to guide treatment decisions and to measure progress in patients with C spasticity because of spinal cord injury (SCI), stroke, multiple sclerosis, or other neurologic causes. Although biomechanic and electrophysiologic methods have been developed that provide quantitative and objective assessment of muscle tone and reflex activity,1-4 these techniques require sophisticated equipment and are time-consuming to administer. In the early 1960s, Bryan Ashworth devised a simple 5-point scale that enabled clinicians to quantify the amount of muscle resistance encountered during passive movements.5 Although the Ashworth Scale has been widely used as a method for measuring the effects of therapeutic interventions on spastic hypertonia in clinical trials,2,6 its reliability remains questionable. Despite its accepted use in clinical practice, the Ashworth Scale provides no information about the degree of the patient’s functional impairment. To address the need for a simple, reliable, clinical method for objectively evaluating functional disability in patients with spasticity, a Disability Assessment Scale (DAS) was developed to assess functional impairment commonly seen in patients with poststroke upper-limb spasticity (ie, dressing, hygiene, limb position, pain). The current study was conducted to evaluate the inter- and intrarater reliability of the Ashworth Scale and the DAS in the assessment of upper-limb tone and functional disability, respectively, in poststroke patients. METHODS Participants Ten patients with upper-limb spasticity were recruited for this single-center trial. Patients were required to be at least 21 years of age; to have a documented history of stroke that resulted in spasticity of the elbow, wrist, finger, and/or thumb flexors in at least 1 limb; and to have some degree of impairment pertaining to hygiene, dressing, limb position, or pain. The protocol excluded patients who had experienced a stroke within 6 months of study enrollment, had a cast on the affected upper limb, had an intrathecal baclofen pump, or had received oral antispasticity medications (eg, baclofen, tizanidine, benzodiazepines, carisoprodol, cyclobenzaprine, methocarbamol) within 30 days of enrollment. This study was conducted in compliance with the ethical standards of good clinical practices, institutional review board regulation, informed consent regulations, and the current Helsinki Declaration.7 Raters The 10 independent raters were health care professionals from a variety of disciplines (physiatry, neurology, physical therapy, occupational therapy) who had specific experience in the treatment of patients with upper-limb spasticity poststroke. Before study initiation, all participating raters received simultaneous, standardized training on the use of the Ashworth Scale Arch Phys Med Rehabil Vol 83, October 2002
1350
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear
and the DAS. On the morning of the study, the raters attended a training session during which they were given detailed instructions regarding the evaluations and the overall study. This training provided guidance to investigators on assessment of muscle tone (Ashworth) and functional ability (DAS). Each rater was provided with a laminated card on which the scales were printed. Design Over a 1-day period, each rater assessed each subject twice while subjects were seated comfortably. The first assessment (evaluation 1) took place in the morning and the second assessment by the same rater (evaluation 2) took place in the afternoon. The 2 evaluations of the same subject took place at least 1 hour apart. All raters assessed each patient, and there was a 15-minute rest interval between every evaluation. To minimize the possibility of investigator bias during evaluations, patients were randomly assigned to 1 of 10 clinical examination rooms on the morning of the study day for evaluation 1 and rerandomized to a different room in the afternoon for evaluation 2. The raters were issued randomized lists indicating the order (by examination room number) in which they were to assess the patients for each evaluation. Thus, the patients were assessed in a different order for evaluation 1 and evaluation 2. The patients and raters were instructed not to discuss the results of the evaluations with each other or with other patients or raters during the study. Outcome Measures The primary outcome measures were the extent of agreement among all raters (interrater reliability) and the extent of agreement between each rater’s 2 evaluations (intrarater reliability) for the Ashworth Scale and the DAS. Ashworth Scale. The Ashworth Scale measures resistance to passive movement according to the following scale5: 0 means no increase in tone (none); 1 means slight increase in tone, giving a catch when the limb is moved in flexion or extension (mild); 2 means more marked increase in tone but limb is easily flexed (moderate); 3 means considerable increase in muscle tone (passive movement difficult [severe]); and 4 means limb rigid in flexion or extension (very severe). Ashworth scores were obtained during flexion of the elbow, wrist, fingers, and thumb, in that order, according to the techniques described later. Raters were instructed to perform each assessment as similarly as possible for each patient. Elbow. With the patient’s arm extended as much as possible and with the palm of the hand facing inward (neutral supination), the rater extended the forearm from maximum possible flexion to maximum possible extension no more than 3 consecutive times and rated flexion muscle tone according to the Ashworth Scale. Wrist. With the patient’s elbow as straight as possible and the forearm pronated so that the palm of the hand faced downward, the rater moved the wrist from maximum possible flexion to maximum possible extension no more than 3 consecutive times and rated flexion muscle tone according to the Ashworth Scale. Fingers. With the patient’s elbow as straight as possible, the forearm pronated so that the palm of the hand was facing inward, the patient’s wrist in neutral position, the rater opened and closed the fingers no more than 3 consecutive times and rated flexion muscle tone according to the Ashworth Scale. Thumb. With the patient’s elbow as straight as possible, the forearm pronated so that the palm of the hand was facing inward and the patient’s wrist in neutral position at 0°, the rater
Arch Phys Med Rehabil Vol 83, October 2002
raised and lowered the thumb from maximum possible flexion to maximum possible extension no more than 3 consecutive times and rated flexion muscle tone according to the Ashworth Scale. A mean Ashworth value for thumb flexion was recorded. Disability Assessment Scale. The raters interviewed each patient to determine the extent of functional impairment for patient hygiene, dressing, limb position, and pain according to the following scale: 0 means no disability, 1 means mild disability (noticeable but does not interfere significantly with normal activities), 2 means moderate disability (normal activities require increased effort and/or assistance), and 3 means severe disability (normal activities limited). Assessment of the 4 functional domains was performed according to the following guidelines. Hygiene. The rater assessed the extent of maceration, ulceration, and/or palmar infection; palm and hand cleanliness; ease of cleanliness; ease of nail trimming; and the degree of interference caused by hygiene-related disability in the patient’s daily life. Dressing. The rater assessed the difficulty or ease with which the patient could put on clothing (eg, shirts, jackets, gloves) and the degree of interference caused by dressingrelated disability in the patient’s daily life. Limb position. The rater assessed the amount of abnormal position of the upper limb. Pain. The rater assessed the intensity of pain or discomfort related to upper-limb spasticity. Statistical Analysis Interrater agreement analyses were performed for all raters. The extent of agreement was analyzed by using the Kendall W statistic, or coefficient of concordance,8 for each parameter rated on each of the 2 scales.a The null hypothesis was that there would be no agreement between raters’ rankings of the patients. If the P value was ⱕ.05, then the null hypothesis was rejected, and the level of agreement was determined by the magnitude of the Kendall W statistic. A 95% confidence interval (CI) was constructed to confirm the hypothesis testing of association. Intrarater agreement analyses were performed for all raters who evaluated patients at both evaluations. A weighted statistic was calculated assessing the agreement between the 2 ratings for each rater, for each parameter measured on each scale. For each parameter, an overall weighted statistic across all raters was estimated with a 95% CI. The null hypothesis was to test that each individual rater’s weighted value for any given parameter was equal to zero. The weighted statistic was tested using a chi-square test.9 The level of intrarater agreement was determined by the magnitude of the overall weighted statistic. Subgroup analyses were also performed according to the raters’ medical specialty or profession. RESULTS Patient Characteristics Ten patients were screened, entered the study, and completed both evaluations. One patient was reported as having a major protocol deviation (ie, he had taken a dose of antispasticity medication on the morning of the study day). Data for this patient were not included in the analysis. Demographic characteristics of the study population are provided in table 1. Of the 9 patients included in the analysis, 5 were men and 4 were women. The mean age was 60 years
1351
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear Table 3: Mean Ashworth Scores (Nⴝ9)
Table 1: Patients’ Demographic Characteristics (Nⴝ9) Age (y) Mean ⫾ SD Median Range Gender Men Women Race White Black Most common events in medical history Stroke/CVA Spasticity Allergies Depression Hypertension Myocardial infarction
59.9⫾16.17 69 37–81 5 (55.6%) 4 (44.4%) 6 (66.7%) 3 (33.3%) 9 (100%) 9 (100%) 5 (55.6%) 4 (44.4%) 4 (44.4%) 3 (33.3%)
Abbreviation: SD, standard deviation; CVA, cerebrovascular accident.
(range, 37– 81y); 6 patients (67%) were white and 3 (33%) were black. Rater Characteristics Demographic and professional characteristics of the raters are presented in table 2. Of the 10 raters, 4 were neurologists, 2 were physiatrists, 2 were occupational therapists, and 2 were physical therapists. The mean overall experience in their respective professions was 6.6 years (range, 1–18y). Evaluations The mean time between evaluation 1 and evaluation 2 was 3 hours 42 minutes (range, 1h 29min to 6h 15min). The median time was 3 hours 45 minutes. Ashworth Scale A summary of the statistics for each parameter assessed by the Ashworth is presented in table 3. Interrater reliability. The Kendall W statistic and 95% CI for interrater agreement were determined by each parameter for evaluation 1, evaluation 2, and the mean of both evaluations. The results showed significant (P⬍.001) agreement among raters for all parameters of the Ashworth Scale (table 4).
Evaluation 1 Evaluation 2 Mean of evaluations 1 and 2
Elbow
Wrist
Fingers
Thumb
1.8⫾.67 1.9⫾.67
1.7⫾.70 1.6⫾.64
2.2⫾.82 2.1⫾.83
1.3⫾.69 1.4⫾.75
1.9⫾.66
1.6⫾.66
2.1⫾.82
1.4⫾.71
NOTE. Values are mean ⫾ SD.
Intrarater reliability. Weighted statistics for intrarater agreement (table 5) indicated either good (weighted ⱖ.4) or excellent reliability (weighted ⱖ.75) for 95% (38 of 40) of the ratings on the Ashworth Scale. The overall weighted values for elbow, wrist, fingers, and thumb ranged from .67 to .74. The null hypothesis was not rejected for any measured parameter. Therefore, the overall weighted values did not differ across raters (Pⱖ.972). Disability Assessment Scale The summary of the statistics for each parameter assessed by the DAS is presented in table 6. Interrater reliability. For all of the DAS parameters (hygiene, dressing, limb position, pain), there was good agreement among raters for the measurements at evaluation 1, evaluation 2, and the mean of both evaluations (table 7). The null hypothesis that there was no agreement among raters on the ratings of the patients was rejected. Therefore, the results showed significant (P⬍.001) agreement among raters for all parameters. Intrarater reliability. For the DAS parameters, 78% (31/ 40) of the evaluations indicated either good intrarater reliability (weighted ⱖ.4) or excellent intrarater reliability (weighted ⱖ.75) (table 8). For all domains measured (hygiene, dressing, limb position, pain), the overall weighted values ranged from .52 to .78. Weighted values did not differ across raters (Pⱖ.478). The chi-square test, applied to test the null hypothesis, yielded P values greater than .05. Therefore, the null hypothesis that the weighted values by the 10 raters were equal for each of parameters rated was not rejected for any parameter. Subgroup Analysis Analyses of inter- and intrarater agreement were performed in subgroups defined by the profession of the rater (ie, neurolTable 4: Ashworth Interrater Agreement Elbow
Table 2: Raters’ Characteristics (Nⴝ10) Age (y) Mean ⫾ SD Median Range Profession Neurologist Physiatrist Occupational therapist Physical therapist Experience in profession (y) Mean ⫾ SD Median Range
37.5⫾4.53 39 29–42 4 (40%) 2 (20%) 2 (20%) 2 (20%) 6.6⫾4.99 6 1–18
Wrist
Fingers
Thumb
Evaluation 1 Kendall W .659 .551 .752 .490 95% CI* .312–1.000 .261–1.000 .356–1.000 .232–1.000 P† ⬍.001 ⬍.001 ⬍.001 ⬍.001 Evaluation 2 Kendall W .616 .507 .692 .625 95% CI .292–1.000 .240–1.000 .328–1.000 .296–1.000 P ⬍.001 ⬍.001 ⬍.001 ⬍.001 Mean of evaluations 1 and 2 Kendall W .765 .598 .792 .611 95% CI .363–1.000 .283–1.000 .375–1.000 .289–1.000 P ⬍.001 ⬍.001 ⬍.001 ⬍.001 * Based on chi-square approximation. Monte Carlo estimate of exact P value testing the Kendall W equal to zero.
†
Arch Phys Med Rehabil Vol 83, October 2002
1352
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear Table 7: DAS Interrater Agreement
Table 5: Ashworth Scale Intrarater Agreement Overall Weighted Statistic for Parameter Rater No.
Elbow
Wrist
Fingers*
Thumb*
1 2 3 4 5 6 7 8 9 10 Overall weighted 95% CI† P‡
.447 .526 .729 .651 .805 .462 .792 .458 .649 .843 .668 .436–.901 .998
.548 .300 .868 .550 .907 .906 .551 .599 .840 .651 .740 .530–.949 .972
.658 .770 .722 .897 .673 .615 .772 .758 .676 1.000 .740 .516–.964 1.000
.500 .509 .883 .401 .419 .653 .813 .654 .674 .000 .680 .435–.924 .985
* statistic for 1 rater with standard error (SE) equal to zero is not included in the computation for overall weighted , 95% CI, and chi-square test. † Based on normal approximation. ‡ From chi-square test for equality of statistics across raters.
ogist, physiatrist, occupational therapist, physical therapist) for each of the scales. The results of these subgroup analyses were similar to those of all raters combined, with good inter- and intrarater agreement for most parameters. DISCUSSION Quantification of muscle tone and associated impairment remains a clinical challenge due to the intricate relationship of this feature with other complications of the upper motoneuron syndrome (eg, muscle contracture, spastic dystonia).10 Over the past several years, numerous methods have been developed to provide information about the resistance of the spastic limb to passive joint movement.3,11 Although some of these methods have proven useful for assessing functional outcomes of patients with spasticity,2,12 none was designed specifically to evaluate function in patients with spasticity after stroke. The results of this study confirm the value of the DAS and the Ashworth Scale, which, when used together, provide complementary information regarding functional outcomes and muscle tone in patients with functional disability caused by upperlimb spasticity poststroke. Although recent efforts have been aimed at developing clinical assessment tools with wider applications than existing ones (eg, the Tone Assessment Scale),13 the Ashworth Scale and the Modified Ashworth Scale (MAS) remain the primary clinical measures of tone.14 Of the 2, the original Ashworth Scale has superior interrater reliability and has been validated as an ordinal-level measurement of resistance to passive movement.14 Only 1 other study15 has assessed the reliability of the
Hygiene
Evaluation 1 Evaluation 2 Mean of evaluations 1 and 2
Dressing
Limb Position
Pain
0.9⫾.52 0.8⫾.53
2.2⫾.53 2.3⫾.50
0.7⫾.71 0.6⫾.59
0.5⫾.69 0.4⫾.63
2.2⫾.48
NOTE. Values are mean ⫾ SD.
Arch Phys Med Rehabil Vol 83, October 2002
0.7⫾.64
0.5⫾.65
Pain
* Based on chi-square approximation. Monte Carlo estimate of exact P value testing the Kendall W equal to zero.
†
Ashworth Scale for use in the upper limbs (excluding studies that assessed the MAS).15 This study examined the intra- and interrater reliability of summated Ashworth scores for a variety of different upper-limb muscles. Similar to the present study, the authors found a good correlation among the 4 raters (Kendall W, .92) and concluded that the Ashworth Scale was a reliable method for evaluating upper-limb muscle tone. To assess the reliability of the Ashworth Scale and DAS across various health care professionals, the present study included 10 independent raters from different professions who were experienced in treating poststroke patients. This relatively large number of raters is an improvement over several previous studies13,16,17 that assessed the reliability of the Ashworth Scale and/or MAS using only 2 raters. In the present study, evaluations at times 1 and 2 were separated by at least an hour (avg time, 3h 42min; range, 1h 29min to 6h 15min). This interval was designed to counteract Table 8: DAS Intrarater Agreement Overall Weighted Statistic for Parameter Rater No.
Hygiene
Dressing*
Limb Position
Pain†
1 2 3 4 5 6 7 8 9 10 Overall weighted 95% CI‡ P㛳
.550 .400 .182 .781 .182 .526 .682 .400 .444 .727 .520 .239–.802 .994
.897 .716 .769 .307 .800 ⫺.052 .781 .379 .079 1.000 .530 .278–.782 .478
.625 .719 .352 .763 .809 .888 .769 .800 .526 .781 .775 .560–991 .998
.000 .727 .794 .308 .880 1.000 .880 .824 .700 .817 .776 .533–1.000 .992
* statistic for 1 rater with SE equal to zero is not included in the computation for overall weighted , 95% CI, and chi-square test. statistics for 2 raters with SE equal to zero are not included in the computation for overall weighted , 95% CI, and chi-square test. ‡ Based on normal approximation. 㛳 From chi-square test for equality of statistics across raters. †
0.9⫾.51
Limb Position
Evaluation 1 Kendall W .527 .477 .587 .695 95% CI* .250–1.000 .226–1.000 .278–1.000 .329–1.000 P† ⬍.001 ⬍.001 ⬍.001 ⬍.001 Evaluation 2 Kendall W .544 .557 .395 .726 95% CI .258–1.000 .264–1.000 .187–1.000 .344–1.000 P ⬍.001 ⬍.001 ⬍.001 ⬍.001 Mean of evaluations 1 and 2 Kendall W .626 .494 .557 .772 95% CI .297–1.000 .234–1.000 .264–1.000 .366–1.000 P ⬍.001 ⬍.001 ⬍.001 ⬍.001
Table 6: Mean Scores on the DAS (Nⴝ9) Hygiene
Dressing
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear
the effects of memory on the ratings (ie, at evaluation 2, it would be more difficult for a rater to remember how he/she scored a patient at evaluation 1). This interval was also meant to counteract the effects of repeated muscular stretching of patients’ limbs, although such effects should have been minimal because raters stretched each patient a maximum of 3 times in each joint position. If anything, this interval has contributed to an underestimation of Ashworth intrarater reliability in this study because fluctuations of tone during the day are well established.18 A recent review14 of the properties and limitations of the Ashworth Scale noted that documentation of changes in function are a relevant complement to the information provided by this scale. The use of functional outcome measures in tandem with measurements of muscle tone reflect the fact that the chief criterion for therapeutic intervention is functional loss caused by spasticity-induced pain, resistance to voluntary muscle control, and/or uncomfortable or undesired limb positions.12 Moreover, in patients with spasticity after stroke, function has been shown to correlate significantly (P⬍.01) with clinically assessed muscle strength.19 Therefore, functional outcome measurements are an essential component of clinicians’ efforts to meet the treatment goal of improved function in patients with spasticity. Together with measurements of muscle tone, these methods can provide complementary information for the management of individual patients with spasticity and for the evaluation of therapeutic interventions. Several methods have been developed to evaluate the functional impact of spasticity, but none of these correlate functional measures with those of increased tone as the Ashworth Scale does. One of the most prominent among these is the Fugl-Meyer Scale, which has been shown to be an accurate and objective method of describing the return of function after hemiplegia.2 In addition to evaluating movement within and independent of synergistic patterns, this instrument includes critical sensory observations of factors that contribute to motor function, such as position sense, pain-free movement, and joint movement.2 The Barthel Index has been shown to be a useful method of measuring activities of daily living (including feeding, grooming, using stairs, dressing) in the general population of people with disabilities.20 In disabled children and infants, the Pediatric Evaluation of Disability Inventory is a reliable, validated method of measuring both capability and performance of functional activities in 3 domains (self-care, mobility, social function).12,21 The Patient Evaluation Conference System represents an interdisciplinary approach to the evaluation of functional disabilities related to spasticity in patients with brain injury; it has been shown to be a sensitive, reliable, and valid method of assessment.22 Recently, a new functional assessment scale, the Valutazione Funzionale Mielolesi, has been validated for patients with SCI.23 Although all of these scales focus on measuring functional disability, none has been studied in patients with focal spasticity after stroke. In contrast, the DAS was specifically designed for use with this population. It is important to note that the present study addressed only the reliability of the Ashworth Scale and the DAS; we did not evaluate their validity. However, in a separate study,24 we assessed the relationship of several efficacy variables with DAS scores for the domain identified as the primary area of impairment (ie, patients and physicians selected either hygiene, dressing, limb position, or pain as their primary therapeutic intervention target). Results showed that DAS scores for this domain correlated with the Ashworth composite muscle tone score (sum of wrist and finger Ashworth scores) (r⫽.608), physician’s global assessment score (r⫽⫺.456), and patient’s
1353
global assessment score (r⫽⫺.506) at week 6 after injection with botulinum toxin type A or placebo. Additional studies will be useful to define the relationship between the DAS and burden of care, and also the utility of the DAS in patients with spasticity due to noncerebrovascular etiologies. CONCLUSION This study showed that the Ashworth Scale and the DAS have good intra- and interrater reliability when used by trained medical professionals to rate excessive tone and functional impairment in patients with upper-limb spasticity after stroke. These scales can be used readily in the clinical setting. References 1. Young RR, Wiegner AW. Spasticity. Clin Orthop 1987;Jun(219): 50-62. 2. Katz RT, Rymer WZ. Spastic hypertonia: mechanisms and measurement. Arch Phys Med Rehabil 1989;70:144-55. 3. Bajd T, Bowman B. Testing and modeling of spasticity. J Biomed Eng 1982;4:90-6. 4. Bajd T, Bowman B. Pendulum testing of spasticity. J Biomed Eng 1984;6:9-16. 5. Ashworth B. Preliminary trial of carisoprodol in multiple sclerosis. Practitioner 1964;192:540-2. 6. Alfieri V. Electrical treatment of spasticity: reflex tonic activity in hemiplegic patients and selected specific electrostimulation. Scand J Rehabil Med 1982;14:177-82. 7. World Medical Assembly. The Declaration of Helsinki: recommendations guiding physicians in biomedical research involving human patients. World Med J 1982;29:86-8. 8. Siegel S, Castellan N. Nonparametric statistics for the behavioral sciences. 2nd ed. New York: McGraw-Hill; 1988. p 284-91. 9. Fleiss JL. Statistical methods of rates and proportions. 2nd ed. New York: Wiley & Sons; 1981. p 38-46. 10. Lance JW. The control of muscle tone, reflexes, and movement: Robert Wartenberg Lecture. Neurology 1980;30:1303-13. 11. Gracies JM, Marosszeky JE, Renton R, Sandanam J, Gandevia SC, Burke D. Short-term effects of dynamic lycra splints on upper limb in hemiplegic patients. Arch Phys Med Rehabil 2000;81: 1547-55. 12. Hinderer SR, Gupta S. Functional outcome measures to assess interventions for spasticity. Arch Phys Med Rehabil 1996;77: 1083-9. 13. Gregson JM, Leathley M, Moore AP, Sharma AK, Smith TL, Watkins CL. Reliability of the Tone Assessment Scale and the modified Ashworth Scale as clinical tools for assessing poststroke spasticity. Arch Phys Med Rehabil 1999;80:1013-6. 14. Pandyan AD, Price CIM, Barnes MP, Rodgers H. A review of the properties and limitations of the Ashworth and modified Ashworth Scales as measures of spasticity. Clin Rehabil 1999;13:373-83. 15. Lee K-C, Carson L, Kinnin E, Patterson V. The Ashworth Scale: a reliable and reproducible method of measuring spasticity. J Neuro Rehabil 1989;3:205-9. 16. Bohannon RW, Smith MB. Interrater reliability of a modified Ashworth scale of muscle spasticity. Phys Ther 1987;67:206-7. 17. Haas BM, Bergstom E, Jamous A, Bennie A. The interrater reliability of the original and of the modified Ashworth scale for the assessment of spasticity in patients with spinal cord injury. Spinal Cord 1996;34:560-4. 18. Skold C. Spasticity in spinal cord injury: self- and clinically rated intrinsic fluctuations and intervention-induced changes. Arch Phys Med Rehabil 2000;81:144-9. 19. Bohannon RW, Warren ME, Cogman KA. Motor variables correlated with the hand-to-mouth maneuver in stroke patients. Arch Phys Med Rehabil 1991;72:682-4. 20. Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J 1965;14:61-5. 21. Feldman AB, Haley SM, Coryell J. Concurrent and construct validity of the Pediatric Evaluation of Disability Inventory. Phys Ther 1990;70:602-10.
Arch Phys Med Rehabil Vol 83, October 2002
1354
RELIABILITY OF ASHWORTH AND DISABILITY SCALES, Brashear
22. Harvey RF, Jellinek HM. Functional performance assessment: a program approach. Arch Phys Med Rehabil 1981;62:456-60. 23. Taricco M, Apolone G, Colombo C, Filardo G, Telaro E, Liberati A. Functional status in patients with spinal cord injury: a new standardized measurement scale. Gruppo Interdisciplinare Valutazione Interventi Riabilitativi. Arch Phys Med Rehabil 2000;81: 1173-80. 24. Brashear A, Gordon MF, Elovic E, et al. A multicenter, doubleblind, randomized, placebo-controlled, parallel study of the
Arch Phys Med Rehabil Vol 83, October 2002
safety and efficacy of Botox (botulinum toxin type A) purified neurotoxin complex in the treatment of focal upper limb spasticity post-stroke [abstract]. Neurology 2001;56 Suppl 3:A78. Supplier a. StatXact associated with SAS; Cytel Software Corp, 675 Massachusetts Ave, Cambridge, MA 02139.