蒙特利爾認知評估及簡易心智狀態問卷調查表於中風病人之再測信度及最小可偵測變化值探討

前言：認知功能障礙是中風病人常見的損傷，評估病人的認知功能更能掌握病人認知功能之狀況。目前臨床單位常用蒙特利爾認知評估 (Montreal Cognitive Assessment, MoCA) 及簡易心智狀態問卷調查表 (Short Portable Mental State Questionnaire, SPMSQ)。此兩量表之再測信度 (test-retest reliability) 及最小可偵測變化值 (minimal detectable change, MDC) 驗證有限，影響對評量結果之解釋。故本研究目的為驗證MoCA及SPMSQ應用在中風病人之再測信度及MDC。方法：招募發病超過6個月以上，沒有嚴重的認知缺陷之中風病人參與研究。受試者接受二次MoCA及SPMSQ評估，間隔為二週。利用組內相關係數 (intraclass correlation coefficient, ICC) 驗證二量表之再測一致性，同時計算測量標準誤 (standard error of measurement, SEM) 及MDC，並以配對t檢定及效應值d驗證二量表之練習效應。結果：MoCA及SPMSQ之ICC分別為0.89及0.87。MoCA及SPMSQ之MDC (MDC%) 分別為5.9 (27.3%) 及1.5 (16.3%)。SPMSQ之練習效應極微小 (d = 0.0)，而MoCA具有低度之練習效應 (d = 0.3)。結論：研究結果顯示MoCA及SPMSQ應用於中風病人具備良好之再測信度及可接受之MDC值。惟MoCA具有低度之練習效應，建議在正式測驗前增加練習次數，以降低練習效應之影響。研究結果將有助於臨床及研究人員合理解讀個案前後測分數的變化，以及選擇適當的評估工具。

Background and Purpose: Cognitive dysfunction is a common disability of patients with stroke. It is necessary for clinicians to assess cognitive function. The Montreal Cognitive Assessment (MoCA) and the Short Portable Mental State Questionnaire (SPMSQ) are commonly used to indicate patients’ attention performance in clinical environments. However, the test-retest reliability and minimum detectable change (MDC) of these two instruments in patients with stroke remain unknown. The purpose of this study was to examine the test-retest reliability and MDC of the MoCA and SPMSQ in patients with stroke.Methods: We recruited 63 patients who had stroke for at least 6 months and had no serious cognitive deficits in the study. The patients were administered with the MoCA and SPMSQ twice, 2 weeks apart. The intra-class correlation coefficient (ICC) was employed to verify the test-retest reliability of the MoCA and SPMSQ, and was used to calculate the standard error of measurement (SEM) and the MDC. We used paired t-test and effect size d to examine the practice effect of these 2 measures.Results: The values of ICC were good for the MoCA and SPMSQ (ICC = 0.89 and 0.87, respectively). The MDC (MDC%) of the MoCA and SPMSQ were 5.93 (27.81%) and 1.76 (19.63%), respectively. The effect sized d of the SPMSQ was very small (d = 0.0), while the effect size of the MoCA was small (d = 0.3).Conclusion: The results of the study showed that the MoCA and SPMSQ have good test-retest reliability and acceptable MDC values for patients with stroke. However, the MoCA had small practice effect. More practice before the formal test should be helpful in reducing the practice effect. The results of the study can help clinicians and researcher interpret change scores and select appropriate cognition measure for the patients.