中文版睡眠困擾量表（神經疾患生活品質）的心理計量品質研究

Abstract

研究背景與目的：中文版神經疾患生活品質成效計劃之睡眠困擾量表（the Chinese version of the NEUROQOL Sleep Disturbance Scale, NEUROQOL-SD-C）為一項用以評估睡眠困擾情形的評估工具，整份量表共有八題試題，為五點量尺評分的量表。本研究目的為驗證NEUROQOL-SD-C應用於台灣社區精神疾病患者與社區健康人的心理計量品質，並檢視精神疾病患者與健康人的睡眠情形。 研究方法：本研究受試者年齡介於18至65歲，資料收集100位社區精神復健機構的精神疾病患者與100位社區健康人。受試者進行NEUROQOL-SD-C、中文版雅典失眠量表、中文版匹茲堡睡眠品質量表、簡式症狀量表－50、世界衛生組織生活品質問卷－臺灣簡明版、貝氏憂鬱量表第二版、貝氏焦慮量表及中文版職能自我評估之施測。以現代測驗理論之羅序測驗分析模式與傳統測驗理論檢驗NEUROQOL-SD-C之內在一致性、建構效度、再測信度、同時效度、收斂效度與區辨效度。 研究結果：本研究將NEUROQOL-SD-C調整成為七題試題，並調整評分量尺成為四點量尺評分的量表。NEUROQOL-SD-C具有可接受至良好的內在一致性，驗證性因素分析結果符合單一建構的標準，所有試題符合羅序測驗分析模式之試題合適度的假設，且在性別、教育程度與族群皆無試題差異功能的現象，具有良好的建構效度。NEUROQOL-SD-C也具有良好的再測信度、同時效度、收斂效度與區辨效度。本研究結果顯示，台灣社區健康人比精神疾病患者呈現顯著較好的睡眠品質（t= -6.469, p＜0.001），然而在失眠情形（t=-1.267, p=0.207）與睡眠困擾程度（t=0.367, p=0.714）上，則沒有顯著差異。 研究結論：本研究結果支持NEUROQOL-SD-C為一項具有信度與效度的評估工具，合適作為研究者及臨床工作人員評估精神疾病患者及一般健康人的睡眠困擾情形之用。

Introduction & Objects: The Chinese version of the NEUROQOL Sleep Disturbance Scale （NEUROQOL-SD-C）evaluates the self-reported sleep disturbance. NEUROQOL-SD-C contains 8 items which are rated by five-point scale. The aim of this study is to investigate the psychometric properties of NEUROQOL-SD-C for the individuals with mental illness and normal subjects. Method: 100 psychiatric patients from community rehabilitation institutes and 100 healthy subjects from community, age 18-65, were recruited. The NEUROQOL-SD-C, the Chinese version of the Athens Insomnia Scale, the Chinese version of the Pittsburgh Sleep Quality Index, Brief Symptom Rating Scale, the World Health Organization Quality of Life－BREF（Taiwan version）, Beck Depression Inventory II, Beck Anxiety Inventory and the Chinese version of the Occupational Self Assessment were administered. Rasch Analysis and Classic Test Theory were used to examine the internal consistency, construct validity, test-retest reliability, concurrent validity, convergent validity and discriminant validity. Healthy subjects show better sleep quality than psychiatric patients significant（t= -6.469, p＜0.001）, however, no different on insomnia（t=-1.267, p=0.207）and sleep disturbance（t=0.367, p=0.714）between two groups. Result: The final NEUROQOL-SD-C contains 7 items which are rated by four-point scale. The NEUROQOL-SD-C demonstrated good internal consistency, acceptable item-total correlations, and met unidimensionality assumption set by Confirmatory Factor Analysis. All items had acceptable fit statistics in Rasch analysis. No differential item functioning was found in gender, education level and group. NEUROQOL-SD-C also showed good test-retest reliability, concurrent validity, convergent validity and discriminate validity. Conclusion: NEUROQOL-SD-C demonstrated satisfactory reliability and validity, and can be a useful tool in clinical evaluation and clinical research. Key Words：