- 學位論文
台灣版糖尿病生活品質測量工具Diabetes-39 的發展與驗證
Development and Validation of the Diabetes-39 Instrument for Measuring Diabetes Quality of Life in Taiwan
摘要
傳統上，我們採用「生物醫學」指標〈例如醣化血色素〉來測量糖尿病人的健康結果。然而，這些指標並沒有考慮疾病或治療過程對病人日常活動功能及生活品質所造成的衝擊。就生活品質的測量而言，台灣目前仍然缺乏具信效度的測量工具。Diabetes-39〈D-39〉為目前世界上使用率最高，且公認具信效度的工具之一。因此，本研究的目的有二，其一為進行D-39中文版的翻譯工作。其二為驗證中文版D-39之心理計量特質。本研究採用橫斷研究法，收集民國95年2月1日至民國95年4月31日間赴署立台南醫院糖尿病門診就診的所有糖尿病患280人。資料的收集方式分別為病人問卷與臨床病歷摘要等。問卷主要收集病患之社會人口學項與糖尿病生活品質等。病歷摘要主要收集不同臨床變項用以驗證D-39，包括：實驗室診斷指標〈醣化血色素、空腹血糖、飯後兩小時血糖、微白蛋白尿、肌酸酐、腎絲球過濾值等〉以及糖尿病併發症〈視網膜病變、腎臟病變、神經病變、足部病變、心血管疾病、及腦血管疾病等〉。翻譯個過程包括正向英文、反向中文翻譯、以及病人焦點團體討論。心理計量分析包括信度與效度。效度驗證則包括構念效度、聚合與鑑別效度〈以SF-36為效標〉、預知團體效度、及綜合鑑別效度等。翻譯結果顯示，中文版D-39在語意上與英文版相似，且能被病人理解與接受。構念效度分析發現D-39中文版的生活品質測量結構與英文版相吻合。聚合與鑑別效度分析發現，D-39與SF-36測量同質性概念的面向〈例如：SF-36之身體生理功能與D-39之活力與活動力〉其相關性比異質性的面向要強。D-39對於飯後兩小時血糖以及肌酸酐等指標分別呈現有測量意義的預知團體效度〈效果量大於0.2〉。D-39對於心血管及腦血管疾病呈現有測量意義的預知團體效度〈效果量大於0.2〉。最後，D-39對神經病變、腦血管疾病、視網膜病變、足部病變、及心血管疾病的綜合鑑別效度達到可接受的標準〈C-統計值大於0.7〉。本研究已成功地翻譯並驗證中文版D-39，D-39的開發將有利於台灣糖尿病治療計畫進行成效之評估，或臨床上對病人健康結果進行追蹤。
並列摘要
Biomedical indicators, such as hemoglobin A1c, are frequently used to assess diabetes patients’ health outcomes. However, these indicators do not account for the impact of illness or treatment on patients’ functioning or quality of life (QOL). To date QOL instruments are still not available for diabetic patients in Taiwan. The Diabetes-39 (D-39), a well-validated and widely used QOL instrument worldwide, can be useful for diabetes patients in Taiwan. The purpose of this study was to translate the D-39 into Chinese and test its psychometric properties. Data were collected from 280 patients treated in diabetes clinics of Tainan Hospital, Taiwan’s Department of Health, through patient survey and medical records. Patient survey focused on patient’s socio demographic characteristics and QOL. Information collected from medical records was consist of laboratory indicators (i.e., hemoglobin A1c (HbA1c), fasting plasma glucose, 2-hour postprandial plasma glucose (2-h PPG), creatinine (Cr), and glomerular filtration rate) and diabetes complications (retinopathy, nephropathy, neuropathy, diabetic foot complications, cardiovascular complications, and cerebrovascular complications). Clinical information was used to validate the D-39. We held forward/back translations and focus group session to translate the D-39. For psychometric evaluation, we analyzed the construct validity, convergent/discriminant validity (using the SF-36 as the criterion), known-group validity, and overall discriminant validity of the D-39. The findings suggest that the translated D-39 reserved equivalent semantics as the original D-39. Pearson’s correlations were weak (r <0.4; P<0.05) with subscales measuring heterogeneous constructs (e.g., mental and physical domains) between the D-39 and SF-36, but high (r >0.6; P<0.05) among dimensions measuring homogenous constructs, suggesting good convergent/discriminant validity. D-39 discriminated better for 2-h PPG, Cr, cardiovascular complications, and cerebrovascular complications, suggesting good known-group validity. The C-indexes suggest greater discrimination (>0.7) by the D39 for all complications except nephropathy, indicating good overall discriminate validity. This study has successfully translated the D39 and demonstrated good psychometric properties. The favorable discriminative capability of the D-39 for clinical indicators suggests the D-39 may be acceptable for use in assessing treatment effectiveness and monitoring patient outcomes in Taiwan.