Introduction: When we evaluate the degree of myelopathy, Japanese Orthopaedic Association (JOA) score and the 10‐second test are usually used. However, the minimum clinically important difference (MCID) for these measures is rarely reported. The purpose of this study was to evaluate MCIDs for JOA score and 10‐second test in cervical myelopathy disorders.
Patients and Method: The JOA decided to revise the JOA score for patients with cervical myelopathy and to develop a new outcome measure. In part of the project, a total of 304 patients with cervical myelopathy disorders, whose symptoms were supposed to be stable, completed the questionnaire (including JOA score and the 10‐second test) twice to verify the reliability. Of those 304 patients, 205 patients who did not have other joint disease were included in the current study. We evaluated: 1) Pearson's correlation coefficient (ρ) for first and second JOA scores and 10‐second test values; 2) distribution of differences between the first and second values; 3) standard error of the mean (SEM) for differences between the first and second values; and 4) the 95% confidence interval for minimum detectable change (MDC95). SEM was calculated using the formula: SEM = s✓(1 ‐ r), where s = standard deviation (SD) of the difference between the first and second values, and r = Pearson's correlation coefficient between the first and second values. MDC95 was calculated using the formula: MDC95 = SEM× ✓2 × 1.96.
Results: 1) Correlation coefficients were ρ = 0.89 for JOA score, ρ = 0. 93 for the 10‐second test of the right hand, ρ = 0.93 for the 10‐second test of the left hand, and ρ = 0.92 for the 10‐second test of the more severely affected side. 2) Mean and SD of the difference between the first and second values was ‐0.20 ± 1.43 for JOA score (Figure 1), ‐0.57 ± 2.86 for the 10‐second test of the right hand, ‐0.49 ± 2.80 for the 10‐second test of the left hand and ‐0.59 ± 2.93 for the 10‐second test of the more severely affected side (Figure 2).
Discussion and Conclusions: We must consider MCID when interpreting the results of clinical measures. This study verified the MCID of JOA score and the 10‐second test by statistically evaluating differences between the two evaluations of patients with stable symptoms. A change less than MDC95 is taken statistically as representing a measurement error. In conclusion, we could judge myelopathy as improved (deteriorated) in cervical myelopathy disorders only when the JOA score showed a change of more than 2 points (MDC95 = 1.3) and the 10‐second test showed a change of more than 3 cycles (MDC95 = 2. 0˜2.2).