The Fugl-Meyer Upper Extremity (FMUE) Scale1 is a widely used and highly recommended stroke-specific, performance-based measure of impairment.2, 3 It is designed to assess reflex activity, movement control and muscle strength in the upper extremity of people with post-stroke hemiplegia. It has been extensively used as an outcome measure in rehabilitation trials and to record post-stroke recovery, particularly in the USA.4
The FMUE Scale comprises 33 items, each scored on a scale of 0 to 2, where 0 = cannot perform, 1 = performs partially and 2 = performs fully. It is free, requires only household items for testing, and takes up to 30 minutes to administer. Two illustrated manuals outlining the assessed components of the scale and scoring criteria have been published to address shortcomings of the original description.5, 6
The time taken to complete the full FMUE Scale has led researchers to develop variants, including a distal upper extremity sub-scale comprising 12 wrist/hand items7 and a ‘short form’ six-item scale of the whole FMUE Scale, which was developed using Rasch analysis to determine the easiest and most difficult items.8 In the development of the short form of the scale, care was taken to preserve the original content representativeness, which is based on sequential stages of post-stroke motor recovery, first documented by Brunnstrom in 1966.9 However, the short form version has been shown to be less sensitive to change at an individual level, which limits its clinical utility.10
Extensive assessment of the psychometric properties of the FMUE Scale has been undertaken.3 The longitudinal stability of assessment items (apart from reflex evaluation) over 6 months has been established, supporting the validity of the measure over time.4 Excellent internal consistency (alpha = 0.94 to 0.98 across four administrations over 6 months) has been demonstrated.11 Satisfactory concurrent validity has been shown in comparison with three other commonly used measures of upper extremity motor recovery.11 In this study, the FMUE Scale was the only tool that did not have significant floor and ceiling effects, and intra-rater (ICC 0.99, 95% CI 0.99 to 1.00) and inter-rater (ICC 0.96, 95% CI 0.92 to 0.98) agreement were shown to be excellent. These authors reported a minimal detectable change for intra-rater assessments of 5.2 on the 66-point scale (8% of the total measure) and 12.9 (20% of the total measure) for inter-rater assessments.
A range of data exists for minimal clinically important differences for the FMUE Scale. A change of between 4 and 7 points in chronic stroke,7 and 9 to 10 points in subacute stroke12 is considered to be clinically significant. A recent study by Hoonhorst et al13 aimed to determine the optimal cut-off scores for the FMUE Scale regarding predictions of upper limb capacity at 6 months post stroke. These authors reported that FMUE Scale scores < 31 corresponded with ‘no to poor’ upper extremity capacity, while 32 to 47 represented ‘limited capacity’, 48 to 52 represented ‘notable capacity’ and 53 to 66 represented ‘full’ upper extremity capacity. Shelton et al14 reported that a 10-point increase from admission to discharge on the FMUE Scale corresponded to a 1.5-point change on the Functional Independence Measure.
Persistent upper extremity deficits are common following stroke. Consequently, psychometrically sound outcome measures that are also easy to use are essential to document change in upper extremity function over time. Although the FMUE Scale has been extensively utilised, shortcomings include that individual finger movements are not assessed; consequently, deficits in distal fine motor functions may be under-reported. Given the importance of fractionated finger movement for many upper extremity functions, this scale may be insufficiently sensitive to document change in very high-functioning individuals. Conversely, the evaluation of upper extremity reflexes, although criticised by some authors,4 provides the ability to detect small changes in the sensorimotor system, which may be particularly useful in those with very limited volitional movement in the acute phase of stroke. Another advantage is the availability of detailed illustrated descriptions of the scale, 5, 6 which may facilitate use by researchers with a limited clinical background.
- Fugl-Meyer, A. et al. Scand J Rehabil Med. 1975; 7: 13–31
- van Wijck, F.M. et al. Neurorehabil Neural Repair. 2001; 15: 23–30
- Gladstone, D. et al. Neurorehabil Neural Repair. 2002; 13: 232–240
- Woodbury, M.L. et al. Arch Phys Med Rehabil. 2008; 89: 1563–1569
- Deakin, A. et al. Physiotherapy. 2003; 89: 751–763
- Sullivan, K.J. et al. Stroke. 2011; 42: 427–432
- Page, S.J. et al. Arch Phys Med Rehabil. 2012; 93: 2373–2376
- Hsieh, Y.W. et al. Stroke. 2007; 38: 3052–3054
- Brunnstrom, S. Phys Ther. 1966; 46: 357–375
- Chen, K.L. et al. Arch Phys Med Rehabil. 2014; 95: 941–949
- Lin, J.-H. et al. Phys Ther. 2009; 89: 840–850
- Arya, K. et al. Top Stroke Rehabil. 2011; 18: 599–610
- Hoonhorst, M.H. et al. Arch Phys Med Rehabil. 2015; 96: 1845–1849
- Shelton, F.D. et al. Neurorehabil Neural Repair. 2001; 15: 229–237