- 學位論文
慢性阻塞性肺病(COPD)緩和醫療介入決定因素探討-照護連續性及介入基準
Determinants of Initiating Palliative Care in COPD- Continuity of Care and Initiating Criteria
摘要
背景 台灣正快速地進入高齡化社會，照護連續性與非癌末期緩和醫療照護決策，成為必須重視的議題。慢性阻塞性肺病(Chronic obstructive pulmonary disease, COPD)是慢性呼吸道發炎疾病，會導致肺功能下降、病人反覆住院及增加死亡率。根據世界衛生組織的統計，2030年COPD將躍升為全球第三大死亡原因。對於COPD病患提高照護連續性，可降低病患再住院及死亡率。此外，末期COPD病患緩和醫療介入過晚，也可能與照護連續性不佳有關，最終導致病患反覆插管、進展至長期呼吸器依賴。然而，由於缺乏明確的緩和醫療介入判斷基準共識，也會使COPD病患緩和醫療介入過晚。本研究將探討照護連續性對COPD死亡病患醫療資源及緩和醫療使用之關係、發展台灣緩和醫療切入基準專家共識，最後證驗專家共識之死亡預測力等，以做為推動慢性肺病緩和療護之參考。 研究方法 (一)探討照護連續性對COPD死亡病患醫療資源、緩和醫療使用以及簽屬拒絕心肺復甦(Do not resuscitate, DNR) 時點之關係：本研究使用中部某醫學中心臨床資料，針對COPD死亡病患，使用照護連續性指標（Continuity of Care index, COCI）、修正型照護連續性指標(Modified, Modified Continuity Index, MMCI)以及經常照護供給者指標（Usual Provider of Care index, UPC index）等三種指標，以探討門診照護連續性與死亡前一年醫療資源及緩和醫療使用之關係。 (二)以修正型德菲法發展台灣COPD緩和醫療切入基準專家共識：採用修正型德菲法與適當性方法RAND/University of California Los Angeles (RAND/UCLA)，建立台灣COPD緩和醫療介入時點之臨床指標。本研究使用兩回合的問卷，加上一次專家座談會產生共識。專家由胸腔暨重症加護醫學會及中華民國重症醫學會推薦，共14人，分布於北中南及宜蘭地區，包括公立醫學中心3人、私立醫學中心9人、區域教學醫院2人。 (三)比較不同緩和醫療切入基準對COPD病患死亡預測力是否有差異：使用中部某醫學中心資料，評估因COPD急性惡化而住院之病患，出院後一年之死亡預測效力。以本研究所發展的德菲法專家共識為驗證指標。本研究進一步比較不同預測模組對於COPD病患出院後一年內死亡之預測力，作為COPD病患出院前評估未來死亡風險及是否介入緩和醫療家庭會談、預立醫囑之決策共享依據。 結果 第一部份結果顯示，照護連續性高(MMCI)確實會下降COPD病患死亡前一年住院及急診次數，但在醫療支出方面，反而呈現增加的狀況。照護連續性與緩和醫療資源利用無相關性。與醫療花費最相關的因素是COPD末期DNR簽署的時點，在死亡當次住院才簽屬的定義為晚期DNR(Late DNR)，其一年醫療總花費是早期DNR(Early DNR)組的1.42倍。綜上，我們推論，這個現象與國內缺乏一個經專家共識產生的COPD末期判定標準，造成緩和醫療資源介入太晚有關。第二部分，以修正型德非法，共選定九類指標以及一類綜合型指標如下：(1) 年齡>80歲；(2) 喘促量表Modified Medical Research Council (mMRC) Dyspnea Scale ≧3；(3) 肺功能參數: 一秒吐氣量 (Forced Expiratory Volume in one second, FEV1)≦30% 預測值；(4) 動脈血氧參數；(5) 身體質量參數(Body Mass Index, BMI) <20或非計畫體重下降；(6) 嚴重或多重共病症；(7) 過去病史-因急性惡化住院；(8) 過去病史-因急性惡化使用過非侵襲性呼吸器或侵襲性呼吸器；(9) 生活起居能力：失能需要照護，需使用鼻胃管導尿管等由他人協助生活起居。綜合型指標ADO (Age、Dyspnea、Obstruction)，分數建議為≧8分。第三部分結果顯示，德菲法專家會議所發展出來的指標，對於一年死亡的預測，具統計顯著差異(p= 0.004, C index= 0.558)。若條件為以總分數作為死亡預測力，德菲法專家會議所發展出來的指標，分數越高，其對死亡的預測力越高，達統計學顯著差異(p<0.001, C index= 0.630)。值得注意的是，ADO指標在本研究中，也同樣證實其對COPD具有很好的死亡預測力。如果以ADO再加上德菲法專家會議指標中的急性惡化病史、BMI兩項指標，這個模型的COPD死亡預測力最高(ADO+ AE+ BMI)，達統計顯著差異 (p<0.001, C index= 0.662)。這些綜合指標可做為COPD末期緩和醫療介入評估工具。 結論 本研究對慢性阻塞性肺病進行照護連續性與醫療資源及緩和醫療資源利用的相關性探討，並經由德菲法專家會議形成COPD末期指標判斷基準，更經由實際醫院資料驗證其死亡預測能力。透過這些指標的逐一檢視，提供臨床端、病患本身及家屬端開始正視病患有可能在一年內進展至死亡的風險評估依據，並據此開始提供除了侵入性維生醫療之外的可能治療選項，讓病患及家屬有更多的時間去思考死亡議題、增進死亡識能，最後自己選擇希望接受的臨終照護模式。本研究結果可做為COPD末期緩和醫療照護介入評估工具及緩和醫療政策之參考，以提高專科醫師及COPD病患之死亡識能，從而提高COPD末期照護品質。
並列摘要
Background Taiwan is entering an aging society, therefore the continuity of care for chronic diseases and the decision-making for palliative care in non-cancer stages have become important issues. Chronic obstructive pulmonary disease (COPD) is a chronic respiratory tract disease that causes decreased lung function, repeated hospitalization, and increased mortality. According to the World Health Organization, COPD will jump into the 3rd leading cause of death in the world in 2030. Increasing care continuity for COPD patients can reduce patient re-hospitalization and mortality. Poor care continuity may cause COPD patients delay to receive palliative care that caused these patients eventually being repeated intubation and progress to ventilator dependent status. However, the initiate of palliative care in late stage COPD patients is often too late. It may due to lackeness of consensus or criteria to initiate palliative care for COPD patients. The current study will firstly explore the association among the care continuity, the medical resources utilizations and the use of palliative care in patients died in COPD. Secondly, we will develop the consensus and criteria of initiate the palliative care in COPD by Taiwan expertis. Finally, we will validate these criteria to predict the 1-year mortality of COPD after hospitalization. We want to establish the useful clinical criteria to be the reference for promoting the palliatve care treatment of chronic lung disease. Material and Method (1) To explore the association of care continuity among COPD patients for their medical resources and palliative care utilizations, and the timing of “Do not resuscitate” (DNR) decisions. This study used clinical data from a medical center in central Taiwan. We use three indexs of care continuity to conduct the current study. These indexs include continuity of Care Index (COCI), modified modified continuity index (MMCI), and usual provider of care index (UPC index). (2) To develop the expert consensus of palliative care intervention for late stage COPD patients by using modified Delphi method. The expert consensus adopts the RAND/UCLA Appropriateness Method developed by the American RAND Corporation and the University of Los Angeles. We enrolled the experts from public medical center, private medical center and regional hospital located in the north, central, south and easten of Taiwan. Through the three rounds of Delphi methos of 14 experts, the feedback and consensus meeting results are used to assess the appropriateness of those indicators. (3) To validate the power of modified Delphi criteria in the prediction of 1-year mortality in COPD This study used clinical data from a medical center in central Taiwan. Patients hospitalized due to acute exacerbation of COPD were enrolled. We validated the power of modified Delphi criteria in the prediction of 1-year mortality in COPD. Results The first part of the results showed that high care continuity (MMCI) did reduce the number of hospitalizations and emergency visits in the year before the death of COPD patients, but in terms of medical expenses, it showed an increase. There is no correlation between care continuity and palliatve care utilization. The most relevant factor for medical expenses is the time when the DNR is signed at the end of the COPD. Patients who had DNR requested in their last admission is defined as late DNR, and those who had a DNR directive prior to their last (terminal) admission to the hospital were classified as Early DNR. The total annual medical expenditure of Late DNR group was 1.42 times higher than the early DNR group. We suggested this phenomenon is related to the lackness of consensus for when to start the palliative care in COPD patients in Tawain. In the second part, 9 criteria of initiating palliative care in COPD were developed by modified Delphi method. Nine indicators were selected as follows: (1) age > 80 years; (2) Modified Medical Research Council (mMRC) Dyspnea Scale ≧3; (3) pulmonary function Parameters: Forced Expiratory Volume in one second (FEV1) ≦30% predicted value; (4) Arterial blood oxygen parameters; (5) Body Mass Index (BMI) <20 or unplanned weight (6) severe or multiple comorbidities; (7) Past medical history - hospitalization due to acute exacerbation; (8) Past medical history - use of non-invasive respirators or invasive respirators due to acute exacerbations; (9) Daily life ability: disability requires care, and it is necessary to use a nasogastric catheter to assist others in daily life. ADO (Age, Dyspnea & Obstruction) index was also thought to be a comprehensive indicator to predict COPD mortality. In the third part, we validated the power of modified Delphi criteria in the prediction of 1-year mortality in COPD. The Delphi criteria have statistically significant differences in the prediction of 1-year of death (p = 0.004, C index = 0.558). The higher of the total score, the higher the predictive power of death (p<0.001, C index= 0.630). The ADO index also be confirmed as a good predictor for death in COPD in this study. We found that ADO plus the medical hx of acute exacerbation and BMI (ADO+ AE+ BMI), this model has the highest predictive value of COPD death (p<0.001, C index= 0.662). Conclusions In this study, we evaluated the relationship between the continuity of care and the utilization of medical resources and palliative care in COPD patients. We not only developed the expert consensus of palliative care intervention for late stage COPD patients by using modified Delphi method, but also verified the mortality prediction ability by actual hospital data. Through the serial review of these indicators, we want to provide the physicians, the patient themselves and their family members the guidance that COPD patients may progress to death within one year. The results of this study can provide the indicators to initiate palliative care or non-invasive care for late stage COPD patients. We suggest these indicators can not only be used by physicians and COPD patients, but also as an assessment tool or criteria to reimburse the palliative care in national health care system, thereby improving the quality of COPD terminal care in the future.