背景與目的：體戚覺功能表現冋以以溫度戚覺量化測試（quantitative sensory testing)做較精確的量化評 估，此測試方式被用來協助診斷或做為療效評估的工具，而好的評估工具必須具有良好信度。本研究目的檢 驗溫度咸覺量化測試（溫度與疼痛偵測能力及疼痛耐受力）之再測信度。方法：健康成人受試者（n = 28)接受 六種測試，包括冷與熱溫度咸覺閾值（CDT和WDT)、冷痛與熱痛戚覺閾值（CPT和HPT)及冷痛與熱痛耐受 力（CPTol和HPTol)測試，受試者在一週後接受第二次施測。統計分析相對信度（relative reliability)與絕對信度(absolute reliability)結果：相對信度中有五種测試其組内相關係數（inh'aclass con'eladon coefficient, ICC) 2 OW 其信度極佳。其中HPTd和CPTol信度最佳，ICC高達0.94和0.91，而WDT之ICC最低（0.65)。絕對信度 只有CDT、WDT和HPTol有較佳的信度，其測量標準誤（standard error of measurement，SEM)很小，介於0.38 至0.64°C，最小真正變化量（smallest real difference, SRD)也很小，介於1.05至1.77°C。而CPT和CPTol有大 的SEM值（3.56和1.24°C)及SRD值（9.87和5.57°C)。由布藺德--奥特曼圖（Bland-Al丨man plols)也顯示CPT 和CPTol有較大的平均差異。結論：雖然所有溫度戚覺量化测試有好到極佳的相對信度，只有CDT、WDT和 HPTol同時具有較佳的絕對信度，此三種溫度咸覺量化測試相當穩定，適合臨床評估體戚覺功能使用。

Background and purpose： Somatosensory functions can be assessed by thermal quantitative sensory resting (QST). QST is recommended in diagnosis of somatosensory dysfunction or as an outcome measure. Tlie purpose of this study was to evaluate the tcst-rctest reliabilities of experimental thermal and pain assessment of thermal QSTs. Methods： Tlic test-retest reliability for thermal detection, pain detection, and pain tolerance were analyzed. Six testing parameters (cold detection threshold, CDT; warm detection threshold, WDT; cold pain threshold, CPT; heat pain threshold, HPT; cold pain tolerance, CPTol; and heat pain tolerance, HPTol) were tested twice on 28 participants with an interval of 1 week. Both relative and absolute reliability statistics were estimated. Results： Tlie relative reliabilities were excellent for CDX CPT, HPT, CPTol and HPTol, with intraclass correlation coefficient (ICC) > 0.75. Both heat and cold tolerance assessment had the best reliability (ICC = 0.94 and 0.91, respectively), while WDT had the lowest ICC (0.65). The absolute reliabilities were better for CDT, WDT, and HPTol, with small standard error of measurement (SEM) (0.38 to 0.64°C). Hie values of smallest real difference (SRD) for CDT, WDT, HPTol were also small (1.05 to 1.77°C). On the other hand, CPT and CPTol had large SEM values (3.56 and 1.24°C) and SRD values (9.87 and 5.57°C). Bland-Altman plots also showed a large mean difference in CPT and CPTol. Conclusion: All tests have good to excellent relative reliability; however, only CDT, WDT, HPTol have small SEM and SRD. CDT, WDT, and HPTol are useful for clinical practice because they have good relative and absolute reliability.