- 學位論文
探討癌症病患中文版心裡社會普查量表之信效度
Psychometric properties of thePsychosocial Screen for Cancer-Revised(PSSCAN-R)in Taiwan
摘要
本研究主要目的為建立中文版癌症病患心理社會普查量表的信效度。採橫斷式研究設計，在台灣北部某區域教學醫院體制下三所醫院的癌症門診和社區醫學中心為研究場所，採方便取樣方式共招募癌症患者151位，社區民眾共90位。研究工具為結構式問卷，包括 (1) 個人基本資料表 (2) 中文版癌症病患心理社會普查量表(3) 中文版醫院焦慮憂鬱量表 (4) 中文版社會支持量表 (5) 中文版困擾溫度計。資料分析包括描述性統計、相關分析、獨立樣本t檢定、卡方檢定、因素分析及信效度檢驗等統計法來檢定中文版癌症病患心理社會普查量表的信效度。項目分析結果顯示此量表15個題目都達顯著水準，予以保留。因素分析在社會支持層面可以解釋的累積變異量為32%。心理社會層面的題目採用5分式的likert-scale計分，萃取出兩個因素，其中兩題有跨因素負荷量，可以解釋的累積變量為67%。信效度的部分，社會支持次量表的庫李信度係數為0.711，心理情緒層面的次量表內在一致性Cronbach'sα為0.899，再測信度採採組內相關係數，ICC值為0.554。專家效度整體CVI值為0.95，聚歛效度則和中文版醫院焦慮量表(r=0.808，p=.000)、中文版社會支持量表(r=0.339，p=.000)達顯著正相關，眾知群體效度中癌症患者和社區一般民眾的焦慮分數和憂鬱分數也有顯著差異(p=0.003，p=.000)。在疾病變項中，憂鬱分數和焦慮分數則是和合併症總數以及罹癌時間有關。合併症總數越多者憂鬱分數和焦慮分數明顯較高，罹癌時間則以半年以上到兩年之間的患者憂鬱分數最高。社會支持分數較高者也反應出較低的憂鬱和焦慮症狀(r= -0.203，p=.013)。 相較於原始量表的Cronbach'sα為0.83，再測信度r值為0.64。中文版癌症病患心理社會困擾量表亦具有良好客觀的信效度，且能夠有效辨別出具有臨床意義的心理痛苦的癌症患者，簡短易測，又涵蓋3個面向，可做為臨床評估癌症患者心理社會層面困擾的工具。
並列摘要
Aim of this research is to understand the reliability and validity of the Mandarin Version of Psychosocial Screen for Cancer Patient-Revised. In Taiwan, most studies used instruments to assess psychological distress for cancer patients such as Brief Rating Scale (BSR-5) or Beck Depression Inventory (BDI). However, most these scales were not purposively designed for cancer patients. In addition, those scales measure single dimension and lack of comprehensive dimension to screen the psychological distress for cancer patients in clinical settings. The 15-item psychosocial screen for cancer (PSSCAN) was to be brief, at a relatively low reading level, capture psychological variables relevant to distress (anxiety and depression) and social support in cancer patients. We designd a cross-sectional study, using convienient sampling as a research method in 3 teaching hospitasl in Northern Taiwan. Totally 151 cancer patietns and 90 participants from community health centres were recruited in this study. The Hospital Anxiety and Depression Scale, Interpersonal Support Evaluation List-16, and Distress Thermometer were used to test the convergent and criterion related validty of PSSCAN-R. It has good psychometrics including high internal consistency (alpha averaging .8, and acceptable test-retest stability over 2~3 weeks (averaging ICC = .719). Validity has been established for content, construct and concurrent validity. However, the result of factor analysis revealed two items of the psychological subscale in PSSCAN_R were cross-loding in two factors. Considering the culture issue and phrasing, we suggested modifying the description of the content to make it more suitable for its psychometric property, depression. In addition, more case number (> 200) would be better for factor analysis. Compare to its original english version, the internal consistency averging .83 and test-retest stablity over 2 monthes (averaging r= .64), the Mandarin Version of PSSCAN_R in Taiwan also has good reliability and validiy, and it might be a good instrument which contains more comporehensive aspest than others, to be a brief and effective screening tool in clinical use.