JRRD Volume 48, Number 1, 2011

Pages 83–88

Journal of Rehabilitation Research & Development

Measurement of lower-limb muscle spasticity: Intrarater reliability

of Modified Modified Ashworth Scale

Nastaran Ghotbi, PhD, PT;1 Noureddin Nakhostin Ansari, PhD, PT;1* Soofia Naghdi, PhD, PT;1 Scott Hasson,
EdD, PT2
1Faculty of Rehabilitation, Tehran University of Medical Sciences, Tehran, Iran; 2Physical Therapy Department,
Angelo State University, San Angelo, TX

Abstract—The Modified Modified Ashworth Scale (MMAS) is cerebral palsy and can cause spasticity, which is com-
a clinical instrument for measuring spasticity. Few studies have mon, complex, and disabling in many patients with these
been performed on the reliability of the MMAS. The aim of the conditions. Spasticity affects at least 38 percent of people
present study was to investigate the intrarater reliability of the 12 months poststroke [1], and approximately 90 percent
MMAS for the assessment of spasticity in the lower limb. We of people with MS experience spasticity at some point
conducted a test-retest study on spasticity in the hip adductors,
[2]. The definition provided by Lance describes spasticity
knee extensors, and ankle plantar flexors. Each patient was mea-
sured by a hospital-based clinical physiotherapist. Twenty-three
as a velocity-dependent disorder of the stretch reflex that
patients with stroke or multiple sclerosis (fourteen women, nine results in increased muscle tone as one component of
men) and a mean +/– standard deviation age of 37.3 +/– UMN syndrome [3]. Measuring spasticity using reliable
14.1 years participated. The weighted kappa was moderate for the and valid tools is important for the evaluation of treat-
hip adductors (weighted kappa = 0.45, standard error [SE] = 0.16, ment efficacy. The Ashworth scale [4], Bohannon-Smith
p = 0.007), good for the knee extensors (weighted kappa = 0.62, modified Ashworth scale [5], and Tardieu scales [6–9]
SE = 0.12, p < 0.001), and very good for the ankle plantar flexors are currently used to measure spasticity.
(weighted kappa = 0.85, SE = 0.05, p < 0.001). The kappa value Currently, the Ashworth scales [4–5] are the most
for overall agreement was very good (weighted kappa = 0.87, commonly used measures of spasticity in clinical practice
SE = 0.03, p < 0.001). The reliability for the ankle plantar flexors and research. However, the reliability and validity of these
was significantly higher than that for the hip adductors. The scales has been recently challenged [10–16]. In an attempt
intrarater reliability of the MMAS in patients with lower-limb
to improve the quality of the Ashworth scales, Ansari et al.
muscle spasticity was very good, and it can be used as a measure
of spasticity over time.
modified the Bohannon-Smith modified Ashworth scale,

Key words: ankle plantar flexors, hip adductors, intrarater,

Abbreviations: MMAS = Modified Modified Ashworth Scale,
knee extensors, lower limb, MMAS, Modified Modified Ash-
MS = multiple sclerosis, SD = standard deviation, SE = stan-
worth Scale, muscle spasticity, rehabilitation, reliability.
dard error, UMN = upper motor neuron.
*Address all correspondence to Dr. Noureddin Nakhostin

Ansari, PhD, PT; Faculty of Rehabilitation, Tehran Univer-

INTRODUCTION sity of Medical Sciences, Enghelab Ave, Pitch-e-shemiran,
11489, PO Box 11155-1683, Tehran, Iran; +98-21-7753-3939;
Upper motor neuron (UMN) syndrome occurs in fax: +98-21-7788-2009. Email: [email protected]
multiple sclerosis (MS), stroke, spinal cord injuries, and DOI:10.1682/JRRD.2010.02.0020

83
84

JRRD, Volume 48, Number 1, 2011

titling the new scale the Modified Modified Ashworth group and the HSlope/MSlope and HMax/MMax ratios was
Scale (MMAS) [10]. In this modified version of the Ash- statistically significant (r = 0.39, p = 0.04) [22].
worth scale, the grade “1+” is omitted and the grade “2” is Few studies have investigated the reliability of the
redefined. In the MMAS, spasticity is scored on an ordinal MMAS. Only one study to date has evaluated intrarater
scale from 0 to 4 as follows: 0 = No increase in muscle reliability. Therefore, the goal of the present study was to
tone; 1 = Slight increase in muscle tone, manifested by a evaluate the intrarater reliability of the MMAS when sev-
catch and release or by minimal resistance at the end of the eral lower-limb muscle groups were assessed in patients
range of motion when the affected part(s) is moved in flex- with spasticity.
ion or extension; 2 = Marked increase in muscle tone,
manifested by a catch in the middle range and resistance
throughout the remainder of the range of motion, but METHODS
affected part(s) easily moved; 3 = Considerable increase in
muscle tone, passive movement difficult; and 4 = Affected Study Design
part(s) rigid in flexion or extension. This study employed a test-retest design to evaluate
The few studies performed on the reliability and the intrarater reliability of the MMAS in patients with
validity of the MMAS have been encouraging. For the lower limb spasticity.
measurement of knee extensor spasticity in 15 patients
after stroke, the MMAS demonstrated reliable inter- and Population
intrarater reliability measurements. The kappa values Patients with MS or stroke with spasticity were
were good between raters ( = 0.72, standard error [SE] = recruited from inpatient wards of Sina University Hospi-
0.14, p < 0.001) and very good within one rater ( = 0.82, tal, Tehran, Iran. To be included in the study, patients had
SE = 0.12, p < 0.001) [17]. The interrater reliability for to have no previous pathology of the affected lower limb
the MMAS was very good when evaluating wrist flexor and had to be able to understand simple commands. The
spasticity in 30 patients with hemiplegia (weighted kappa exclusion criteria were muscle contracture and severe
[w] = 0.92, SE = 0.03, p < 0.001) [18]. In a study to joint pain.
investigate the interrater reliability of the MMAS in the
assessment of elbow flexor spasticity, inexperienced rat- Rater
ers tested 21 adult patients with stroke. The w was 0.81 The rater was a physiotherapist with more than
(SE = 0.097, p = 0.0002), thus interrater reliability for 16 years experience in managing patients with muscle
two inexperienced raters was very good [19]. spasticity. She was familiar with the MMAS but was not
experienced using it. The descriptions of the rating crite-
In a cross-sectional study to assess the interrater reli-
ria were included on the recording form.
ability of the MMAS in the upper limb of adult patients
with hemiplegia, two physiotherapists rated two common
Scale and Testing Position
spastic muscle groups (elbow flexors and wrist flexors)
The MMAS was the scale used. Three muscle groups
of 15 patients. The w was 0.61 for elbow flexors and
in the affected lower limb of each patient were assessed:
0.78 for wrist flexors. The interrater reliability of the
hip adductors, knee extensors, and ankle plantar flexors.
MMAS was good [20]. The standardized test positions and movements are
To assess the interrater reliability of the MMAS in described in Table 1 and have been described and used in
lower-limb muscle spasticity, Ghotbi et al. tested 22 adults other studies [12,21].
with stroke or MS. Hip adductors, knee extensors, and
ankle plantar flexors were assessed. Interrater reliability Procedure
was very good for the hip adductors and the knee extensors The procedure used in previous investigations was
w = 0.82, p < 0.001) and good for the ankle plantar flex- followed [12,21]. Demographic data including age, sex,
ors (w = 0.74, p < 0.001) [21]. etiology, affected side, and disease duration were
One further study explored the validity of the MMAS recorded. The subjects were tested while in the hospital.
in 27 patients with stroke. The relationship between the The assessment order of limbs in cases of bilateral
MMAS scores obtained from the wrist flexor muscle involvement in patients with MS was randomized; in all
 85

GHOTBI et al. Intrarater reliability of MMAS

Table 1.
Standardized positions and movements for rating with Modified Modified Ashworth Scale.
Muscle Group Patient Rater
Hip Adductors Supine, head in midline, and lower limbs On side being tested, rater placed one hand underneath limb
in extended position. close to knee and other hand supported limb close to ankle. Limb
was moved into full abduction (without rotation).
Knee Extensors Side-lying, with hips and knees in extension. Behind patient, rater placed one hand just proximal to knee, on
Head and trunk aligned in straight line. Pillow lateral surface of thigh, to stabilize femur and other hand just
can be used behind hips, if necessary, to stabi- proximal to ankle. Knee was moved from maximum extension to
lize patient. maximum flexion.
Ankle Plantar Supine, with head in midline, and arms On side being tested, rater placed one hand under ball of foot,
Flexors alongside trunk. Lower limbs in extended while other hand stabilized limb around ankle joint. Ankle was
position. moved from maximum plantarflexion into maximum dorsiflexion.

patients, the assessment order of the muscles was ran- ined for intrarater reliability because eleven patients with
domized. The same physiotherapist performed a second MS were bilaterally involved. However, 101 muscle
assessment 2 days after the first. Patients were instructed spasticity assessments were obtained because one patient
to relax during the test and to not resist the passive move- found it difficult to stay in the side-lying position for the
ments applied by the physiotherapist. The joints were knee extensor muscle test. In this study, all MMAS
moved with a fast-stretching velocity by counting “one- grades from 0 to 4 were scored, but the patients were
thousand-and-one” as suggested by Bohannon and Smith most often assigned the scores of 0 and 1 across all mus-
[5]. The passive movement was repeated three times at cle groups. Most agreement was obtained for grade 0 fol-
each joint. lowed by grade 1 (Table 2).
For hip adductor and knee extensor muscle groups,
Data Analysis
no subjects were graded 3 or 4. Patients were most often
Descriptive statistics were calculated for subjects and assigned grades of 0 and 1. For ankle plantar flexor mus-
variables with SPSS, version 11.5 (SPSS Inc; Chicago, cles, the patients were mostly given grades of 2 and 1, but
Illinois). The quadratic weighted kappa statistic was used
patients did attain scores of 3 and 4. Table 3 shows the
to calculate reliability. The reliability results were inter-
weighted agreement for each of the muscles. The
preted as follows [23]: very good 0.81–1.00, good 0.61–
weighted kappa was moderate for the hip adductors, good
0.80, moderate 0.41–0.60, fair 0.21–0.41, and poor <0.21.
for the knee extensors, and very good for the ankle plan-
The chi-square test was performed to analyze the differ-
ence between the weighted kappa values for each muscle tar flexors. The kappa value for overall agreement was
group [24]. Statistical significance was set at p < 0.05. very good.

Table 2.
RESULTS Total assignments by rater for three lower-limb muscle groups using
Modified Modified Ashworth Scale (MMAS).
Twenty-three patients (fourteen women, nine men) with MMAS Scores: MMAS Scores: Assessment 2
Total
a mean ± standard deviation (SD) age of 37.3 ± 14.1 years Assessment 1 0 1 2 3 4
were examined in this study. Eighteen patients had MS 0 42 5 1 0 0 48
and five patients had hemiplegia due to a single stroke. 1 10 14 0 0 0 24
The mean ± SD ages of patients with stroke and MS were 2 0 9 7 1 0 17
61.6 ± 12.1 years and 33.1 ± 9.4 years, respectively. The 3 0 0 2 8 0 10
mean ± SD time poststroke was 517.6 ± 1,138.9 days. 4 0 0 0 0 2 2
The patients with MS had a mean ± SD history of MS of Total 52 28 10 9 2 101
2,615.7 ± 1,658.3 days. A total of 34 limbs were exam- Note: Overall w = 0.87 (standard error = 0.03), p < 0.001.
86

JRRD, Volume 48, Number 1, 2011

Table 3.
Weighted agreement within rater (intrarater) in Modified Modified Ashworth Scale scores for three muscle groups of lower limbs.
% Weighted
Muscle Group w 95% CI SE p-Value Interpretation
Agreement
Hip Adductors 89 0.45 0.13–0.77 0.16 0.007 Moderate
Knee Extensors 96 0.62 0.39–0.85 0.12 <0.001 Good
Ankle Plantar Flexors 98 0.85 0.77–0.94 0.05 <0.001 Very Good
Overall 98 0.87 0.81–0.93 0.03 <0.001 Very Good
CI = confidence interval, SE = standard error.

The chi-square test showed a significantly higher treatment. This implies that the patients were not clini-
weighted kappa value for the ankle plantar flexors than cally stable. The unstable condition of the patients com-
for the hip adductors (p < 0.05). The difference between bined with medication usage could have affected the tone
the weighted kappa values of the knee extensors and of the muscles.
those of either hip adductors or ankle plantar flexors was Although these factors could have affected the con-
not statistically significant.
sistency of the results on the reliability of the MMAS, the
following points should also be considered. First, the
DISCUSSION intrarater reliability in this study was good and very good
for the knee extensors and ankle plantar flexors. Second,
The results of the present study demonstrated the the interrater reliability of the MMAS had been previ-
intrarater reliability of the MMAS in measuring lower- ously investigated by two inexperienced physiotherapy
limb muscle spasticity in patients with neurological con- students in a hospital setting and very good agreement
ditions. The overall reliability for lower-limb muscle was found for each muscle and the overall weighted
spasticity was very good. kappa [21]. In this investigation of intrarater reliability,
The rater in this study was not trained in the use of the overall weighted kappa value for the three muscle
the MMAS. In actual clinical situations, very little time groups was very good.
exists for training clinicians. Therefore, we offered no
In patients with UMN lesions, multiple muscles in a
training to the rater. We also expected the training to be
limb are involved. Establishing reliability for individual
unnecessary because she was experienced in handling
neurological patients and had previous experience using muscle groups is important. However, reliable assessment
the Bohannon-Smith modified Ashworth scale [5]. of the entire affected limb is desirable, especially for clini-
So far, the findings on the reliability of the MMAS cal purposes. The overall agreement for the muscles in a
have been good and very good [17–21]. In this study, limb gives a general idea about the reliability of the assess-
moderate reliability was observed for the hip adductors ment. Therefore, the reliability of a scale in individual
and good reliability for knee extensors; however, these joints needs to be considered in light of the overall reli-
are lower than previous measures of intrarater reliability ability obtained in the limb. In this study, the overall very
[17]. One possible reason could be the clinical environ- good agreement for the muscles indicates the reliable
ment in the hospital. At times the hospital was not quiet. assessment of spasticity for the limb has been achieved.
Muscle spasticity in patients with UMN lesions fluctu-
Accordingly, the MMAS showed a high degree of
ates because of environmental stimuli [25].
intrarater reliability for lower-limb muscle spasticity
Another explanation for the lower reliability of the
hip adductors could be the larger mass of the lower limb. among patients with neurological conditions that makes it
The rater reported that the difference in the reliability for suitable for initial evaluation and reassessment. However,
the hip adductors and ankle plantar flexors was because we must note that the time interval between the two assess-
she had more control over the ankle. ments was short and the rater could remember her original
Another possible reason for the lower reliability in scores. The rater might have been influenced by her mem-
this study was that the patients were undergoing medical ory of the result of her first assessment.
 87

GHOTBI et al. Intrarater reliability of MMAS

CONCLUSIONS 5. Bohannon RW, Smith MB. Interrater reliability of a modi-

fied Ashworth scale of muscle spasticity. Phys Ther. 1987;
The present study demonstrated that the MMAS has 67(2):206–7. [PMID: 3809245]
overall very good intrarater reliability in patients with lower- 6. Ansari NN, Naghdi S, Hasson S, Azarsa MH, Azarnia S.
limb muscle spasticity. A statistically higher reliability for The Modified Tardieu Scale for the measurement of elbow
distal ankle plantar flexors than for proximal hip adductors flexor spasticity in adult patients with hemiplegia. Brain
was noted. The MMAS showed adequate reliability for the Inj. 2008;22(13–14):1007–12. [PMID: 19117179]
DOI:10.1080/02699050802530557
measurement of lower-limb muscle spasticity over time.
7. Haugh AB, Pandyan AD, Johnson GR. A systematic review
of the Tardieu Scale for the measurement of spasticity. Dis-
abil Rehabil. 2006;28(15):899–907. [PMID: 16861197]
ACKNOWLEDGMENTS DOI:10.1080/09638280500404305
8. Morris S. Ashworth and Tardieu scales: Their clinical rele-
Author Contributions:
Study concept and design: N. Nakhostin Ansari, N. Ghotbi.
vance for measuring spasticity in adult and paediatric neu-
Analysis and interpretation of data: N. Nakhostin Ansari, N. Ghotbi, rological populations. Phys Ther Rev. 2002;7(1):53–62.
S. Naghdi, S. Hasson. DOI:10.1179/108331902125001770
Drafting of manuscript: N. Ghotbi, N. Nakhostin Ansari, S. Naghdi. 9. Gracies JM, Marosszeky JE, Renton R, Sandanam J, Gan-
Study supervision: N. Ghotbi. devia SC, Burke D. Short-term effects of dynamic lycra
Statistical analysis: N. Nakhostin Ansari. splints on upper limb in hemiplegic patients. Arch Phys
Critical revision of manuscript for important intellectual content: Med Rehabil. 2000;81(12):1547–55. [PMID: 11128888]
N. Nakhostin Ansari, S. Naghdi, S. Hasson. DOI:10.1053/apmr.2000.16346
Financial Disclosures: The authors have declared that no competing
interests exist. 10. Ansari NN, Naghdi S, Moammeri H, Jalaie S. Ashworth
Funding/Support: This material was unfunded at the time of manu- Scales are unreliable for the assessment of muscle spasticity.
script preparation. Physiother Theory Pract. 2006;22(3):119–25.
Institutional Review: The project was approved by the Research [PMID: 16848350]
Council of Rehabilitation Faculty of Tehran University of Medical DOI:10.1080/09593980600724188
Sciences. A general written informed consent was obtained from the 11. Tederko P, Krasuski M, Czech J, Dargiel A, Garwacka-
patients for any test and examination for medical or research pur- Jodzis I, Wojciechowska A. Reliability of clinical spasticity
poses. We additionally obtained verbal consent from the patients or
measurements in patients with cervical spinal cord injury.
caregivers once the study had been explained in detail.
Ortop Traumatol Rehabil. 2007;9(5):467–83.
Participant Follow-Up: The authors do not plan to inform partici-
pants of the publication of this study. [PMID: 18026067]
12. Ansari NN, Naghdi S, Arab TK, Jalaie S. The interrater and
intrarater reliability of the Modified Ashworth Scale in the
REFERENCES assessment of muscle spasticity: Limb and muscle group
effect. NeuroRehabilitation. 2008;23(3):231–37.
[PMID: 18560139]
1. Watkins CL, Leathley MJ, Gregson JM, Moore AP, Smith
TL, Sharma AK. Prevalence of spasticity post stroke. Clin 13. Naghdi S, Ansari NN, Mansouri K, Olyaei GR, Asgari A,
Rehabil. 2002;16(5):515–22. [PMID: 12194622] Kazemnejad A. The correlation between Modified Ashworth
DOI:10.1191/0269215502cr512oa Scale scores and the new index of alpha motoneurones excit-
2. Hobart JC, Riazi A, Thompson AJ, Styles IM, Ingram W, ability in post-stroke patients. Electromyogr Clin Neuro-
Vickery PJ, Warner M, Fox PJ, Zajicek JP. Getting the mea- physiol. 2008;48(2):109–15. [PMID: 18435215]
sure of spasticity in multiple sclerosis: The Multiple Sclero- 14. Alibiglou L, Rymer WZ, Harvey RL, Mirbagheri MM. The
sis Spasticity Scale (MSSS-88). Brain. 2006;129(Pt 1):224– relation between Ashworth scores and neuromechanical
34. measurements of spasticity following stroke. J Neuroeng
[PMID: 16280352] Rehabil. 2008;5:18. [PMID: 18627628]
DOI:10.1093/brain/awh675 DOI:10.1186/1743-0003-5-18
3. Lance JW. The control of muscle tone, reflexes, and move- 15. Craven BC, Morris AR. Modified Ashworth scale reliability
ment: Robert Wartenberg Lecture. Neurology. 1980;30(12): for measurement of lower extremity spasticity among patients
1303–13. [PMID: 7192811] with SCI. Spinal Cord. 2010;48(3):207–13.
4. Ashworth B. Preliminary trial of carisoprodol in multiple [PMID: 19786977]
sclerosis. Practitioner. 1964;192:540–42. [PMID: 14143329] DOI:10.1038/sc.2009.107
88

JRRD, Volume 48, Number 1, 2011

16. Fleuren JF, Voerman GE, Erren-Wolters CV, Snoek GJ, Brain Inj. 2009;23(10):815–19. [PMID: 19697170]
Rietman JS, Hermens HJ, Nene AV. Stop using the Ash- DOI:10.1080/02699050903200548
worth Scale for the assessment of spasticity. J Neurol Neu- 22. Naghdi S, Ansari NN, Mansouri K, Asgari A, Olyaei GR,
rosurg Psychiatry. 2010;81(1):46–52. [PMID: 19770162] Kazemnejad A. Neurophysiological examination of the
DOI:10.1136/jnnp.2009.177071 Modified Modified Ashworth Scale (MMAS) in patients
17. Ansari NN, Naghdi S, Younesian P, Shayeghan M. Inter- with wrist flexor spasticity after stroke. Electromyogr Clin
and intrarater reliability of the Modified Modified Ashworth Neurophysiol. 2008;48(1):35–41. [PMID: 18338533]
Scale in patients with knee extensor poststroke spasticity. 23. Brennan P, Silman A. Statistical methods for assessing
Physiother Theory Pract. 2008;24(3):205–13. observer variability in clinical measures. BMJ. 1992;
[PMID: 18569857] 304(6840):1491–94. [PMID: 1611375]
DOI:10.1080/09593980701523802 24. Fleiss JL. The measurement of interrater agreement. In:
Fleiss JL, editor. Statistical methods for rates and propor-
18. Naghdi S, Ansari NN, Azarnia S, Kazemnejad A. Interrater
tions. New York (NY): Wiley; 1981. p. 222.
reliability of the Modified Modified Ashworth Scale (MMAS)
25. Blackburn M, Van Vliet P, Mockett SP. Reliability of mea-
for patients with wrist flexor muscle spasticity. Physiother
surements obtained with the modified Ashworth scale in
Theory Pract. 2008;24(5):372–79. [PMID: 18821443]
the lower extremities of people with stroke. Phys Ther.
DOI:10.1080/09593980802278959
2002;82(1):25–34. [PMID:11784275]
19. Ansari NN, Naghdi S, Hasson S, Mousakhani A, Nouriyan
A, Omidvar Z. Inter-rater reliability of the Modified Modi-
Submitted for publication February 26, 2010. Accepted
fied Ashworth Scale as a clinical tool in measurements of
in revised form August 29, 2010.
post-stroke elbow flexor spasticity. NeuroRehabilitation.
2009;24(3):225–29. [PMID: 19458429]
DOI:10.3233/NRE-2009-0472 This article and any supplementary material should be
20. Ansari NN, Naghdi S, Hasson S, Fakhari Z, Mashayekhi cited as follows:
M, Herasi M. Assessing the reliability of the Modified Ghotbi N, Nakhostin Ansari N, Naghdi S, Hasson S.
Modified Ashworth Scale between two physiotherapists in Measurement of lower-limb muscle spasticity: Intrarater
adult patients with hemiplegia. NeuroRehabilitation. 2009; reliability of Modified Modified Ashworth Scale. J Reha-
25(4):235–40. [PMID: 20037215] bil Res Dev. 2011;48(1):83–88.
DOI:10.3233/NRE-2009-0520 DOI:10.1682/JRRD.2010.02.0020
21. Ghotbi N, Ansari NN, Naghdi S, Hasson S, Jamshidpour B,
Amiri S. Inter-rater reliability of the Modified Modified
Ashworth Scale in assessing lower limb muscle spasticity.