Introduction
Cervical myelopathy is a common spinal disorder that is most often caused by age-related changes, including intervertebral disc degeneration, osteophyte formation, ligamentous hypertrophy, and the ossification of the posterior longitudinal ligament (OPLL) [1,2]. These degenerative processes tend to narrow the spinal canal, resulting in chronic spinal cord compression, progressive neurologic impairment, and reduced quality of life [3]. Prompt diagnosis and timely surgical intervention are critical for improving neurologic and overall outcomes and preventing irreversible disease progression [4–6].
The importance of early detection and management highlights the need for reliable and standardized functional assessment tools for cervical myelopathy [7]. Various scales have been developed to assess neurologic function, disability, and overall clinical status in affected patients. Among these, the Modified Japanese Orthopedic Association (mJOA) score, as introduced by Benzel et al. [8], remains one of the most widely used instruments for assessing functional status in cervical myelopathy [7,9]. Originally designed to assess disease severity and functional impairment, the mJOA score has also been used as an outcome measure, with score changes used in indices such as the Hirabayashi recovery rate [10].
For the mJOA score to be applied internationally, it must be translated and adapted to the language and context of the target population. Cultural adaptation ensures that each functional domain is interpreted as intended, allowing valid comparisons across different populations [11,12]. Moreover, the mJOA score itself was developed as a culturally adapted version of the original Japanese questionnaire. During this process, the chopstick-use component was replaced with an item assessing the ability to fasten shirt buttons to reflect cultural differences. This modification resulted in the widely adopted 18-point version [8]. Since then, the mJOA score has been translated and validated in several languages, including Dutch, Italian, Portuguese, and Arabic, with these versions consistently showing high reliability and validity [13–16].
However, no validated Korean version of this score currently exists, limiting standardized assessments in Korean-speaking patients. Thus, this study aimed to (1) culturally adapt and translate the mJOA score into Korean and (2) assess its reliability and validity, thereby providing a standardized assessment tool for clinical and research use in Korean-speaking patients with cervical myelopathy.
Materials and Methods
Study design and population
This study was a prospective, questionnaire-based study. The protocol was reviewed and approved by the Asan Medical Center (2025-0293). Written informed consent was obtained from all patients before enrollment.
Between March and June 2025, 122 native Korean-speaking patients with cervical myelopathy were consecutively recruited from a spine outpatient clinic. The inclusion criteria were as follows: (1) a diagnosis of cervical myelopathy, (2) age >18 years, and (3) completion of the required questionnaire at least once. The exclusion criteria were as follows: (1) refusal to participate; (2) any condition impairing effective communication; (3) neurologic symptoms attributable to causes other than myelopathy; or (4) history of surgery for tumor, infection, or fracture.
The sample size was determined based on a reliability analysis, assuming an expected intraclass correlation coefficient (ICC) of 0.80 (α=0.05, power=0.90). Moreover, a minimum of 62 paired retests were required to achieve acceptable precision (±0.10). Accounting for an anticipated retest completion rate of 70%, the target sample size was set at 90 participants, which was also sufficient for secondary validity analyses [17].
The baseline demographic and clinical characteristics, including age, sex, diagnosis, and treatment status, were obtained. The treatment status was classified as conservative, preoperative, or postoperative. During the outpatient visit, written informed consent was obtained, and the participants completed the first paper-based questionnaire. Subsequently, the same questionnaire was administered online via a mobile platform 3–7 days later to assess the test–retest reliability. The study comprised two phases: (1) the cross-cultural adaptation of the mJOA score and its translation into Korean and (2) the assessment of its reliability and validity.
Translation and cross-cultural adaptation
Permission to develop the Korean version of the mJOA score was obtained from the developer of the original English version (Dr. Edward Benzel). The questionnaire comprises four domains: the upper limb motor domain, the lower limb motor domain, the upper limb sensory domain, and the bladder function domain, all of which were retained in the Korean version. Furthermore, the translation and cross-cultural adaptation adhered to the guidelines proposed by Guillemin et al. [11] and Beaton et al. [12].
Forward translation was independently performed by one spine surgeon (S.P.) and one medical student (S.L.), both of whom were native Korean speakers fluent in English. The two translations were synthesized under the supervision of an independent spine surgeon to create the preliminary Korean version. Back-translation was performed by an independent bilingual professional translator who was blinded to the original English version. Subsequently, two spine surgeons (S.P. and S.L.) reviewed the back-translated version against the original to ensure linguistic accuracy and cultural relevance. The pre-final version was reviewed and refined by an expert committee comprising four spine surgeons (i.e., C.J.H., J.H.C., H.D.L., and J.P.) and two coordinators (i.e., M.Y.L. and S.J.Y.) specializing in spinal disease care. Furthermore, the finalized Korean mJOA score was approved and prepared for psychometric testing.
Analysis of reliability and validity
Descriptive statistics were calculated for each domain and the total Korean mJOA score. The values were presented as the mean±standard deviation and range. Floor and ceiling effects were defined as the proportion of participants achieving the lowest or highest possible scores, respectively, with ≥15% considered significant [18].
Internal consistency was assessed using Cronbach’s α and corrected item–total correlations for each domain. Cronbach’s α values between 0.70 and 0.95 and item–total correlations above 0.30 were considered acceptable [18]. Test–retest reliability was assessed in participants who completed both questionnaires (with an interval of 3–7 days) using the ICC and with 95% confidence intervals. The ICC values were interpreted as follows: 0.00–0.20, poor; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, good; and 0.81–1.00, excellent agreement [19].
Construct validity was assessed by correlating the Korean mJOA score with other patient-reported outcome measures, including the adapted Korean version of the Neck Disability Index (AK-NDI) and Visual Analog Scale (VAS) scores for neck and arm pain. Pearson’s correlation coefficients (r) were calculated between the total mJOA score and the AK-NDI, neck pain VAS, and arm pain VAS.
Statistical analysis
All statistical analyses, including descriptive statistics, internal consistency, item–total correlations, test–retest reliability, and construct validity, were performed using IBM SPSS Statistics for Mac, ver. 29.0 (IBM Corp., Armonk, NY, USA). Statistical significance was set at a p-value of <0.05.
Results
Study population
A total of 122 patients met the inclusion criteria. The cohort comprised 74 men (60.7%) and 48 women (39.3%), with a mean age of 62.5±11.5 years (range, 32–87 years). The diagnoses included cervical myelopathy secondary to ossification of the OPLL and cervical spondylotic myelopathy. Three patients (2.5%) were managed conservatively, 17 (13.9%) were managed preoperatively, and 102 (83.6%) were managed postoperatively (Table 1).
Descriptive statistics and floor/ceiling effects
Table 2 shows the descriptive statistics for each domain and the total Korean mJOA score. The mean total score was 14.83±2.88 (range, 6–18). No significant floor effects were observed in any domain or the total score. However, ceiling effects exceeding the 15% threshold were observed across all domains, most notably in the bladder function (72.1%) and upper limb motor (61.5%) domains [18].
Internal consistency
Cronbach’s α for the total Korean mJOA score was 0.666, with corrected item–total correlations ranging from 0.369 to 0.596. The deletion of any single item did not substantially alter Cronbach’s α (Table 3).
Test–retest reliability
Of the 122 participants, 92 (75.4%) completed the second questionnaire, with an interval of 3–7 days between administrations. The ICC for the total score was 0.896 (p<0.001), indicating almost perfect agreement. Substantial to almost perfect agreement was observed in the upper limb motor (ICC=0.854, p<0.001) and lower limb motor domains (ICC=0.833, p<0.001). Moderate agreement was observed for the upper limb sensory domain (ICC=0.651, p<0.001) and fair agreement for the bladder function domain (ICC=0.478, p<0.001) (Table 4) [19].
Construct validity
The total Korean mJOA score showed a moderate negative correlation with the AK-NDI (r=−0.497, p<0.001) and a weak negative correlation with the arm pain VAS (r=−0.339, p=0.003). However, no significant correlation was observed with neck pain VAS (p=0.107) (Table 5).
Discussion
This study successfully translated and validated the Korean version of the mJOA score through a cross-cultural adaptation process that adhered to established international guidelines. The developed questionnaire showed high test–retest reliability and significant associations with other patient-reported outcome measures, including the AK-NDI and the arm pain VAS. Moreover, the validated Korean mJOA score provides clinicians and researchers with a standardized instrument for assessing functional status in Korean-speaking patients with cervical myelopathy, allowing for consistent symptom evaluation and facilitating international comparisons.
The translation adhered to a standardized procedure previously described in the literature, incorporating both forward and backward translations conducted by healthcare professionals and certified translators [11,12]. A consensus meeting was convened to review the linguistic accuracy and cultural equivalence of the translated version. This rigorous process also ensured that the content, integrity, and conceptual meaning of the mJOA items were preserved and appropriately conveyed in Korean.
The Korean mJOA score showed acceptable internal consistency, with a Cronbach’s α of 0.666. Although slightly below the conventional threshold of 0.70, lower α values are common—and generally considered acceptable—for scales with few items assessing heterogeneous constructs [18,20]. Similar Cronbach’s α values have been reported in validation studies of other language versions of the mJOA, such as the Arabic (α=0.78) and Italian (α=0.60) versions [14,21]. Comparable results have also been observed in the psychometric analyses of the original English version (α=0.63), suggesting that the internal consistency of the Korean version is consistent with that of previously validated instruments [22].
The test–retest reliability, assessed in a subset of patients who completed the questionnaire twice at an interval of 3–7 days, was moderate to excellent for the total score and most domains [19]. These findings confirm that the questionnaire reliably measures functional impairment in patients with cervical myelopathy. However, a significant ceiling effect was observed, particularly in the bladder function and upper limb motor domains [18]. This suggests that while the instrument effectively assesses moderate to severe impairments, it may be less sensitive to subtle functional deficits or changes in patients with milder disease or postoperative recovery. Furthermore, since most patients (83.6%) were assessed during postoperative follow-up, their improved functional status likely contributed to the observed ceiling effects. Therefore, clinicians and researchers should exercise caution when interpreting the mJOA scores in patients with near-normal function or in postoperative follow-up settings in which functional recovery is already advanced. In such contexts, complementary assessment tools or more sensitive functional measures may be required to fully capture subtle but clinically significant changes.
Construct validity was assessed by correlating the Korean mJOA with established patient-reported outcome measures, which is consistent with recommended psychometric methods [23]. The Korean mJOA score showed a moderate negative correlation with the AK-NDI (r=−0.497, p<0.001) and a weak negative correlation with the arm pain VAS (r=−0.339, p=0.003). Given that the higher AK-NDI and VAS scores indicate greater disability, these negative correlations were expected. The AK-NDI has previously been validated in Korean-speaking patients with upper extremity disorders [24]. These findings suggest that the Korean mJOA score appropriately reflects the overall functional status in cervical myelopathy. Since the mJOA score primarily assesses neurologic function rather than pain, its correlations with pain-based measures such as the VAS were expectedly weaker [8]. This emphasizes the complementary value of combining the mJOA score with pain-specific instruments to capture the multidimensional impact of cervical myelopathy.
However, several limitations should be acknowledged. First, this study was conducted at a single tertiary care center, which may limit its generalizability to broader populations. Second, most participants were postoperative cases, which likely contributed to the high ceiling effects and limited assessment across the full spectrum of disease severity. Third, the bladder function scores may be influenced by comorbid conditions unrelated to cervical myelopathy—a known limitation of the mJOA format, in which domain-specific symptoms may not always be clearly distinguishable. Finally, although reliability and construct validity were established, responsiveness—the ability of the Korean mJOA score to detect clinically significant changes over time—was not assessed. Future multicenter studies involving larger, more diverse populations and longitudinal designs are needed to assess the responsiveness of the Korean mJOA score to clinical changes and to establish the minimum clinically important difference, which will be essential for interpreting treatment-related improvements and for use in clinical trials.
Conclusions
In conclusion, the Korean version of the mJOA score showed acceptable internal consistency, substantial test–retest reliability, and significant construct validity. Despite the notable ceiling effects, it represents a reliable instrument for assessing functional impairment in Korean-speaking patients with cervical myelopathy. Future multicenter studies involving larger, more diverse populations and longitudinal designs are needed to further assess its responsiveness and clinical applicability. Furthermore, its implementation will allow for the standardized assessment of Korean-speaking populations in clinical practice and research settings.
PDF Links
PubReader
XML Download
Full text via DOI
Download Citation
