Psychometric Performance of Clinician- and Patient-Reported Outcome Measures in Surgically Treated Thoracic Myelopathy: A Multicenter Prospective Study.

Abstract

Study DesignMulticenter prospective observational study.ObjectivesTo determine the responsiveness, reliability, and construct validity of commonly used clinician-reported and patient-reported outcome measures in surgically treated thoracic myelopathy, a condition for which standardized outcome assessment has not been established.MethodsAdults undergoing surgery for MRI-confirmed thoracic myelopathy were prospectively enrolled at nine tertiary centers. Clinician-reported measures included the Japanese Orthopaedic Association (JOA) score, a thoracic-adapted JOA subtotal (JOAt), and the modified JOA (mJOA). Patient-reported outcomes included the Japanese Orthopaedic Association Cervical Myelopathy Evaluation Questionnaire (JOACMEQ) and EuroQOL 5-dimension (EQ-5D). Assessments were performed preoperatively and at 3-6 months postoperatively. Internal responsiveness was evaluated using Wilcoxon's r, test-retest reliability using ICC(1,1), and construct validity using Spearman's correlation with EQ-5D.ResultsFifty-one patients were included (mean age 65.6 years; 59% male). Significant postoperative improvement was observed across major clinician-reported and patient-reported outcomes. Large internal responsiveness was demonstrated for JOA (r=0.64), JOAt (0.63), mJOA (0.68), and JOACMEQ lower-extremity function (LF) (0.59). Test-retest reliability was high for JOA, JOAt, mJOA, JOACMEQ-LF, and EQ-5D (ICC range 0.78-0.85). JOACMEQ-LF showed the strongest construct validity with EQ-5D at both baseline (ρ=0.62) and follow-up (ρ=0.72).ConclusionsJOAt, mJOA, and JOACMEQ lower-extremity function demonstrated robust psychometric performance in thoracic myelopathy. These measures represent suitable candidates for standardized outcome assessment and provide a methodological foundation for future clinical studies and guideline development in this under-studied condition.