Hsueh I-Ping, Hsu Miao-Ju, Sheu Ching-Fan, Lee Su, Hsieh Ching-Lin, Lin Jau-Hong
School of Occupational Therapy, College of Medicine, National Taiwan University, Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, Taiwan.
Neurorehabil Neural Repair. 2008 Nov-Dec;22(6):737-44. doi: 10.1177/1545968308315999. Epub 2008 Jul 21.
To provide empirical justification for selecting motor scales for stroke patients, the authors compared the psychometric properties (validity, responsiveness, test-retest reliability, and smallest real difference [SRD]) of the Fugl-Meyer Motor Scale (FM), the simplified FM (S-FM), the Stroke Rehabilitation Assessment of Movement instrument (STREAM), and the simplified STREAM (S-STREAM).
For the validity and responsiveness study, 50 inpatients were assessed with the FM and the STREAM at admission and discharge to a rehabilitation department. The scores of the S-FM and the S-STREAM were retrieved from their corresponding scales. For the test-retest reliability study, a therapist administered both scales on a different sample of 60 chronic patients on 2 occasions.
Only the S-STREAM had no notable floor or ceiling effects at admission and discharge. The 4 motor scales had good concurrent validity (rho >or= .91) and satisfactory predictive validity (rho = .72-.77). The scales showed responsiveness (effect size d >or= 0.34; standardized response mean >or= 0.95; P < .0001), with the S-STREAM most responsive. The test-retest agreements of the scales were excellent (intraclass correlation coefficients >or= .96). The SRD of the 4 scales was 10% of their corresponding highest score, indicating acceptable level of measurement error. The upper extremity and the lower extremity subscales of the 4 showed similar results.
The 4 motor scales showed acceptable levels of reliability, validity, and responsiveness in stroke patients. The S-STREAM is recommended because it is short, responsive to change, and able to discriminate patients with severe or mild stroke.
为选择适用于中风患者的运动量表提供实证依据，作者比较了Fugl - Meyer运动量表（FM）、简化版FM（S - FM）、运动功能脑卒中康复评估工具（STREAM）和简化版STREAM（S - STREAM）的心理测量特性（效度、反应度、重测信度和最小真实差异[SRD]）。
在效度和反应度研究中，50名住院患者在康复科入院和出院时接受FM和STREAM评估。S - FM和S - STREAM的评分从相应量表中获取。在重测信度研究中，一名治疗师在60名慢性病患者的不同样本上两次使用这两种量表。
仅S - STREAM在入院和出院时没有明显的地板效应或天花板效应。这4种运动量表具有良好的同时效度（rho≥0.91）和满意的预测效度（rho = 0.72 - 0.77）。这些量表显示出反应度（效应量d≥0.34；标准化反应均值≥0.95；P < 0.0001），其中S - STREAM反应度最高。量表的重测一致性极佳（组内相关系数≥0.96）。这4种量表的SRD为其相应最高分的10%，表明测量误差处于可接受水平。这4种量表的上肢和下肢分量表结果相似。
这4种运动量表在中风患者中显示出可接受的信度、效度和反应度水平。推荐使用S - STREAM，因为它简短、对变化敏感，并且能够区分重度或轻度中风患者。
