本研究旨在以「迷你國際神經精神會談量表」(MINI)檢測「情緒溫度計」做對照，判定內科病人情緒困擾的篩檢準確度及最佳切分點，以建立全院住院病人皆適用的篩檢標準值。研究期間自2014年8月1日至10月31日，為期3個月，採橫斷性研究設計，集群抽樣，選取綜合科、心臟科、腸胃科三個內科病房共194位病人為收案對象。研究分析使用SPSS套裝統計軟體，首先採描述性統計，接著使用迷你國際神經精神會談量表做疾病篩檢，最後計算出接收者操作特徵(ROC)分析所得的「ROC曲線下面積」大小，檢測情緒溫度計的準確度及切分點。結果發現運用情緒溫度計於內科病人情緒困擾篩檢，其ROC曲線下面積高達.809；情緒溫度計最佳的切分點設為4分，對照敏感度72.7%，特異性82.4%。本研究顯示「情緒溫度計」是一個有效且方便的測量工具，可協助臨床醫護人員及早發現病人目前情緒困擾症狀並介入護理措施，進而提升醫療照護品質，達優質全人照護。

This study compared the results obtained from the Mini-International Neuropsychiatric Interview (MINI) to examine screening accuracy and the best cut-off score for patients with emotional disturbance, thereby establishing a standard screening value applicable to all inpatients. This cross-sectional study was conducted between August 1, 2014, and October 31, 2014, and adopted cluster sampling to select 194 patients as participants from 3 medical wards: the general department, department of cardiology, and department of gastroenterology. Statistical analysis was conducted using SPSS. First, descriptive statistics were adopted and MINIs were conducted to carry out disease screening. Then, the area under the receiver operating characteristic curve (ROC) obtained from ROC analysis was calculated to detect the MINIs accuracy and cut-off score. Findings revealed that when MINIs were used for the screening of medical patient's emotional disturbance, the area under the ROC curve reached up to .809; the best cut-off score of the MINIs was set to be 4 scores, and its corresponding sensitivity and specificity were 72.7% and 82.4%, respectively. This study demonstrated that MINIs are an effective and convenient measurement tool that can help clinical medical personnel identify emotional disturbance symptoms and take appropriate interventional measures, thereby improving quality of care for all inpatients undergoing emotional disturbance.