復健功能量表之發展與效度研究

Development and Validation of a Rehabilitation Functional Scale

復健醫療的治療成效，取決於病患功能的獨立情形。因此評估病人的功能獨立程度，是各個專業一致的業務內容之一。台灣目前臨床所使用的量表仍以國外發展者為主。這類量表目前仍缺少本土實證的信、效度資料，因此臨床資料的流通與比較仍屬困難。台灣復健功能量表的發展即為了滿足此需求，此量表涵蓋了日常生活活動，社會心理層面等部分，屬於一份多向度的測驗。本文的目的是探討由國人自行發展的復健功能量表的信、效度品質。經由現代測驗理論中的羅序測量模式，分析量表的建構效度、項目順序效度以及內在一致性。本研究的個案共321名。來自於北部兩家醫學中心的復健科病房。診斷別包括中風、腦傷、脊髓損傷、骨科問題以及其他神經系統相關問題。本研究採用羅序模式之電腦分析程式BIGSTEPS以分析。研究結果顯示，台灣復健功能量表涵蓋三個建構內容：分別為日常生活功能量表，認知－社交量表以及主觀安適感量表。在項目合適度（item fit statistics）上，除了日常生活功能量表中的進食及爬樓梯兩項未能符合橫式之預測，其餘的項目均符合羅序模式之推估。項目效度上，台灣復健功能量表各個次量表之項目難易度順序，除了走路1項外，其餘與國外相關量表之項目反應類似，符合項目難易度效度。項目的內在一致性部分，各個次量表的係數均達0.95以上，顯示具有不錯的內在一致性。本研究結果支持合灣復健功能量表是一份具有信、效度的功能量表。針對其中部分項目的問題，可以考慮更改其評分方式與標準，以提昇其項目品質。臨床上本量表將有助於建立國人復健醫療成效指標之基準，作為健保給付之依據。

The purpose of the study is to investigate the reliability and validity of a functional measure developed in Taiwan. Since one of the outcome measures of the medical rehabilitation is the degree ot independence of the clients, the need to evaluate the functional independence ot the clients become a clinical task for each professional member Currently, clinicians still use functional measure, which was developed in foreign countries without much evidence of their psychometric qualities, Therefore, the comparison of clinical data hospital-wide is difficult. The development of Taiwanese Rehabilitation Functional Scale (TaRFS) is out of the needs from clinicians and program managers. The data of the study were from the Taiwan Rehabilitation Database System. There were 528 observations from 321 patients with a variety of diagnoses. Diagnoses include CVA, brain injury, spinal cord injury, orthopedic conditions, and other neurological conditions. They were from two university-affiliated hospitals on the north part of Taiwan. Rasch measurement model was used to analyze the psychometric qualities of the scale. The results showed that the TaRFS contains three constructs: Activities of Daily Living (ADL), Cognitive-Social Scale (CSS), and Subjective Well-Being (SWB) dimensions. With the exception of two items on ADL scale, the remaining items fit the model. The order of item difficulties for ADL scale is similar to the one for FIM used in foreign countries except for the mobility item on ADL scale. The internal consistency is high as demonstrated by high item separation reliability index. The results of the study indicated that the TaRFS is a psychometrically sound instrument for measuring the functional outcome of patients in medical rehabilitation in Taiwan.