ABSTRACT
OBJECTIVE: To develop a classification for neuromuscular disease patients in each of the three motor function domains (D1: standing and transfers; D2: axial and proximal function; D3: distal function). MATERIALS AND METHODS: A draft classification was developed by a study group and then improved by qualitative validation studies (according to the Delphi method) and quantitative validation studies (content validity, criterion validity and inter-rater reliability). A total of 448 patients with genetic neuromuscular diseases participated in the studies. RESULTS: On average, it took 6.3minutes to rate a patient. The inter-rater agreement was good when the classification was based on patient observation or an interview with the patient (Cohen's kappa=0.770, 0.690 and 0.642 for NM-Score D1, D2 and D3 domains, respectively). Stronger correlations (according to Spearman's coefficient) with the respective "gold standard" classifications were found for NM-Score D1 (0.86 vs. the Vignos Scale and -0.88 vs. the Motor Function Measure [MFM]-D1) and NM-Score D2 (-0.7 vs. the Brooke Scale and 0.64 vs. MFM D2) than for NM-Score D3 (0.49 vs. the Brooke scale and -0.49 vs. MFM D3). DISCUSSION/CONCLUSIONS: The NM-Score is a reliable, reproducible outcome measure with value in clinical practice and in clinical research for the description of patients and the constitution of uniform patient groups (in terms of motor function).