For the purposes of this review, we conducted a literature search to identify all relevant publications on the psychometric properties of the MAS.
Sabari et al. (2005) used Rasch analysis to examine the validity of the scoring hierarchies of the Upper Limb subscales of the MAS and found large floor and ceiling effects for subscales 6 and 7 (31 and 28 percent for both respectively) and large floor effects for subscale 8 (38 percent). Participants were a mean of 104 days post-stroke, within a range of 3 days to 6.4 years.
In a later study, Miller, Slade, Pallant and Galea (2010) used Rasch analysis to evaluate the psychometric properties of the Upper Limb subscales of the MAS (UL-MAS) in post-stroke patients and found adequate floor and ceiling effects (14 and 9 percent respectively). Participants were a mean of 64.8 days post stroke, within a range of 4 to 193 days. Miller at al. questioned the clinical applicability of results from Sabari et al. because some participants were as long as 6 years post-stroke and in clinic the MAS is generally administered in the acute/subacute phase.
No studies have examined the internal consistency of the MAS.
Carr et al. (1985) evaluated the test-retest reliability of the MAS in 14 patients with stroke who were examined by the same rater on two occasions, with a 4-week interval between assessments. Test-retest correlations were excellent, ranging from r=0.87 to r=1.00 (the average correlation was r=0.98).
No studies have examined the intra-rater reliability of the MAS.
Carr et al. (1985) selected 5 patients at various stages of recovery for inter-rater reliability testing. Twenty physical therapists and physical therapy undergraduate students were raters (the general tonus item was excluded from the evaluation). Overall, the MAS was found to have excellent inter-rater reliability, with a mean correlation of r = 0.95. The greatest agreement was achieved on the balanced sitting item (r = 0.99), and the least agreement was on the sitting to standing item (r = 0.89).
Poole and Whitney (1988) examined the inter-rater reliability of the MAS in 24 patients with stroke. Two examiners observed and scored each subject independently. The inter-rater reliability for the total MAS (r = 0.99), and for the individual items (ranging from r = 0.92 to r = 1.00) was excellent, with the exception of the general tonus item, which demonstrated poor reliability (r = 0.29).
Carr et al. (1985) based items and scoring options of the MAS on observations of the improvement of a large number of patients. Thus, no formal content validation is available (Carr et al., 1985).
Malouin, Pichard, Bonneau, Durand and Corriveau (1994) assessed the concurrent validity of the MAS in comparison to the Fugl-Meyer Assessment early after stroke and reported excellent Spearman’s correlations for total scores (r = 0.96, excluding the general tonus item). Correlations between MAS items and corresponding Fugl-Meyer Assessment items were excellent (ranging from r = 0.65 to r = 0.93). The MAS Balance score correlations with the Fugl-Meyer Assessment-Sensation scores of light touch and position sense were excellent (r = 0.64 and r = 0.67, respectively), but correlations with the Fugl-Meyer Assessment Balance items were poor (r = 0.12 and r = -0.10, respectively).
Poole and Whitney (1988) assessed the concurrent validity of the MAS in comparison to the Fugl-Meyer Assessment in a more chronic population and found similar results to Malouin et al. (1994).Excellent Spearman’s correlations were found for total score (r = 0.88), and for individual items (ranging from r = 0.64 to r = 0.92), with the exception of sitting balance, which correlated poorly (r = 0.28).
Miller at al. (2010) examined the construct validity of test item 72 (radial deviation of the wrist). Radial deviation is no longer thought to be an isolated movement occurring at the wrist and is now believed to occur as a part of a coordinated synergy (Mason, Gomez & Ebner, 2001 as cited in Miller, Slade, Pallant & Galae, 2010) or program of movement (Marotta, Medendorp & Crawford, 2003 as cited in Miller et al., 2010). In addition, radial deviation is often reduced in individuals over 65 years, thus impacting results for item 72. The evidence suggests that item 72 adds little meaning to the assessment of motor recovery in patients with stroke.
Simondson et al. (2003) examined the convergent validity of a new scale, the Mobility Scale for Acute Stroke Patients with other established scales (MAS, Functional Ambulation Classification system, Functional Independence Measure, Barthel Index). Mobility items from the Mobility Scale for Acute Stroke Patients had an excellent correlation with corresponding items on the MAS (r = 0.89), demonstrating the convergent validity of the MAS.
Tyson and DeSouza (2004) examined the convergent validity of the MAS in 48 patients post-stroke. It was found that a simple measure of functional sitting balance (sitting arm raise and forward reach tests) correlated adequately with the sitting item of the MAS (r = 0.33 and r = 0.54).
No studies have examined the known groups validity of the MAS.
Dean and Mackey (1992) reported significant differences between mean scores for each item on the MAS from admission to discharge from stroke rehabilitation after an average of 71 days of rehabilitation.
Nugent, Schurr and Adams (1994) found an adequate correlation (r = 0.45) between the number of repetitions of a weight-bearing exercise (designed to strengthen the leg extensor muscles) and the change in the MAS score for the walking item among 25 patients receiving inpatient rehabilitation. To be included in the study, subjects had to have a score greater than 0 but less than 6 on the walking item of the MAS. All of the patients who practiced the exercise achieved independent walking for at least a 3m distance, which gave a final MAS score of three or greater.
English and Hillier (2006) examined the responsiveness of the MAS in 61 rehabilitation inpatients with stroke. The ability of the MAS to detect change was limited in this sample. The MAS item 5 (walking) showed a large effect size (ES) (ES = 1.02) and was able to detect change amongst lower functioning subjects (12 patients showed no change). The other items of the MAS were less responsive, in particular, the effect sizes for the arm items (items 6 to 8) change scores were small (ES ranged from 0.36 to 0.5) and between 44.3 and 63.9% of subjects did not change on these measures. In addition, over 80% of subjects were rated at the extremes of the scales on all three of the arm items. These findings suggest that clinicians should be cautious in choosing the MAS to measure change in patients as for some subgroups and for certain items, clinical change is unlikely to be detected by this tool.
Dean, C. M., Mackey, F. M. (1992). Motor assessment scale scores as a measure of rehabilitation outcome following stroke. Aust J Physiother, 38, 31-35.
English, C. K., Hillier, S. L. (2006). The sensitivity of three commonly used outcome measures to detect change amongst patients receiving inpatient rehabilitation following stroke. Clinical Rehabilitation, 20, 52-55.
Hsueh, I-P., Hsieh, C-L. (2002).Responsiveness of two upper extremity function instruments for stroke inpatients receiving rehabilitation. Clinical Rehabilitation, 16(6), 617-624.
Kjendahl, A., Jahnsen, R., Aamodt, G. (2005). Motor Assessment Scale in Norway: Translation and inter-rater reliability. Advances in Physiotherapy, 7(1), 7-12.
Lannin, N. A. (2004). Reliability, validity and factor structure of the upper limb subscale of the Motor Assessment Scale (UL-MAS) in adults following stroke. Disability & Rehabilitation, 26(2), 109-116.
Loewen, S. C., Anderson, B. A. (1988). Reliability of the Modified Motor Assessment Scale and the Barthel Index. Phys Ther, 68, 1077-1081.
Loewen, S. C., Anderson, B. A. (1990). Predictors of stroke outcome using objective measurement scales. Stroke, 21, 78-81.
Malouin, F., Pichard, L., Bonneau, C., Durand, A., Corriveau, D. (1994). Evaluating motor recovery early after stroke: comparison of the Fugl-Meyer Assessment and the Motor Assessment Scale. Arch Phys Med Rehabil, 75(11), 1206-1212.
Miller, K.J., Slade, A.L., Pallant, J.F., Galea, M.P. (2010). Evaluation of the psychometric properties of the upper limb subscales of the Motor Assessment Scale using a Rasch analysis model. J Rehabil Med, 42, 315-322.
Poole, J. L., Whitney, S. L. (1988). Motor assessment scale for stroke patients: concurrent validity and interrater reliability. Arch Phys Med Rehabil, 69(3), 195-197.
Sabari, J. S., Lim, A. L., Velozo, C. A., Lehman, L., Kieran, O., Lai, J. S. (2005). Assessing arm and hand function after stroke: a validity test of the hierarchical scoring system used in the motor assessment scale for stroke. Arch Phys Med Rehabil, 86(8), 1609-1615.
Simondson, J. A., Goldie, P., Greenwood, K. M. (2003). The mobilityscaleforacute stroke patients: Concurrent validity. Clinical Rehabilitation, 17(5), 558-564.