於醫學中心之整合性用藥管理模式的發展與評估

Abstract

背景：藥物治療是現代醫療的關鍵，但藥品相關問題也是病人健康的威脅。已有許多研究支持藥師提供的臨床藥事服務能夠減少藥品相關問題。然而，受限於現實情境，研究中有效的服務並不能直接實行到所有醫療環境。因此，發展臨床藥事服務模式時，必須考慮在地脈絡與條件，並運用科學方法來評估其效益。 目的：在臺大醫院規劃一項臨床藥事服務模式並評估其效果，本研究包含下列三個研究目的： 1. 根據實證發展整合性用藥管理模式，以減少住院病人的藥品相關問題，並建立評估計畫。 2. 評估該整合性用藥管理模式對於病人照護的臨床效果 3. 評估該整合性用藥管理模式對醫療人員在臨床藥事服務的看法以及與藥師合作意願的影響，並探討這些影響背後的機轉。 方法：為了系統性發展與評估一項臨床藥事服務模式，本研究採取下列研究方法： 1. 成立研究團隊，並採取以下步驟來建立整合性用藥管理模式和評估計畫：文獻回顧、定義問題、發展介入模型、建立標準流程、選擇評估指標、建立評估計畫。此模式將在臺大醫院內科部整合醫學科中的一病房進行試驗性實施，以評估其對於病人照護以及專業間合作關係的影響。 2. 採中斷時間序列研究方法。將研究病房每月入院病人的臨床指標數據統合為一個觀察值，以評估介入前後每月觀察值的變化。研究的首要結果是平均用藥疏失通報數，亦對下列臨床指標進行評估：平均每日院內處方使用數、平均每日自備藥使用數、每日藥費中位數、平均抗生素使用天數、平均住院天數、超長住院天數率、進入加護病房率、及死亡率。使用分段迴歸分析來評估整合性用藥管理模式實施後各項結果指標的水平和斜率變化，並且進行了敏感性和次族群分析。 3. 採解釋性序列混合式研究。首先，透過問卷調查，評估整合性用藥管理模式對醫療人員的看法以及其與藥師合作意向的影響。接著，透過半結構式訪談與主題式分析，整理醫療人員對整合性用藥管理模式的經驗。最後，整合量性與質性研究的發現，以探討整合性用藥管理模式是透過那些機制來影響醫療人員的合作關係。 結果：本研究發展了一整合性用藥管理模式，並透過研究支持其效果，主要結果如下： 1. 基於文獻回顧的結果，我們選定了三項經實證有效的臨床藥事服務：用藥連貫性照護、用藥審查與以病房為基礎的藥師服務，將這三項服務整合成整合性用藥管理模式。此模式在病人的入院、住院中與出院前階段皆提供相應的臨床藥事服務。此外，我們透過改變理論建立此模式的邏輯模型，並依此制定評估指標與評估計畫。 2. 在實施整合性用藥管理模式後，測試病房的用藥疏失平均通報數顯著增加（水平變化：1.02，95%信賴區間：0.68至1.35，p< 0.001）。並且，下列用藥疏失的通報數增加較為明顯：漏開藥或用藥落差、不適當的用藥選擇、不適當的途徑或劑型、不符合健保給付規定。對於年齡≥75歲的病人，模式實施後，平均每日院內處方數顯著減少（水平變化：-1.78，95%信賴區間：-3.06至-0.50，p= 0.009）。不過，其餘指標則沒有統計上顯著變化。 3. 共有58位醫療人員完成了問卷，其結果顯示，相比於一般藥事照護模式的病房，來自實施整合性用藥管理模式的病房的醫療人員有更高的意圖與藥師討論病人的用藥問題。接著，我們對11位醫療人員進行訪談，他們表示，在整合性用藥管理模式實施後，他們與藥師的合作關係得到提升，體驗到更有效的臨床藥事服務，並注意到與藥師的溝通有所改善。在量化與質化資料的整合結果發現，整合性用藥管理模式在促進合作關係的關鍵機制是常規地讓藥師整合到醫療團隊中，這使醫療人員能熟悉藥師的角色，改善溝通，並使藥師能夠更好地識別醫療人員的需求。 結論：本研究計畫基於實證和臺大醫院的環境條件，發展了整合性用藥管理模式。這個模式旨在有效減少病人的藥品相關問題，同時提升整體照護品質。在評估研究中顯示，此整合性用藥管理模式能有效提高了用藥疏失的發現率，亦能減少老年族群的用藥數量。此外，實施此模式後，醫療人員與藥師之間的溝通與合作顯著加強，藥師積極參與並與醫療團隊直接溝通，提高了臨床現場的服務品質，並增強了醫療人員對藥師角色的理解和認可。這些研究成果可為其他醫療機構提供參考。不過，本研究在參與人數和病人族群方面的範圍相對較小。未來研究應評估不同設置對臨床藥事服務模式的影響，也可進行更嚴謹的經濟研究，以確定此模式的經濟效益。

Background: Pharmacotherapy is crucial in current medical practices, yet drug-related problems threaten patients'' health. Pharmacist-led clinical pharmaceutical services (CPS) have been proven to reduce these problems. However, evidence-proven services in research settings cannot be directly implemented in real-world settings due to contextual differences. Therefore, developing service models tailored to the local context and scientifically evaluating their effectiveness is essential to expand local CPSs. Objectives: This study aims to develop and evaluate a CPS model at the National Taiwan University Hospital, and it includes the following three research objectives: 1. Develop an integrated medication management model (IMM) based on current evidence to reduce drug-related problems in hospitalized patients and establish an evaluation plan. 2. Evaluate the clinical effectiveness of the IMM model for patient care. 3. Evaluate the impact of the IMM model on healthcare professionals'' (HP) perceptions of CPSs and their intention to collaborate with pharmacists and explore the mechanisms behind the impact. Methods: To systematically develop and evaluate a CPS model, this study adopts the following research methods: 1. We formed a research team and adopted the following steps to develop the IMM model and its evaluation plan: literature review, problem definition, intervention model development, standard process establishment, and setting evaluation indicators. The model was piloted in one medical center ward to evaluate its impact on patient care and interprofessional collaboration. 2. We conducted an interrupted time-series study, aggregating clinical data of monthly admissions from the study wards into observational units and comparing the changes in monthly observations before and after implementing the IMM model. The primary outcome was the mean number of medication error (ME) reports. The following clinical outcomes were also evaluated: mean number of daily inpatient prescriptions, mean number of daily self-prepared medications, median daily medication costs, mean antibiotic use days, mean length of stay, extended stay rate, ICU admission rate, and mortality rate. Segmented regression analysis was used to assess the level and slope changes in outcome indicators after implementing the IMM model. We also conducted sensitivity and subgroup analyses. 3. We employed a sequential explanatory mixed methods design. First, through questionnaire surveys, we assessed the impact of the IMM on HPs'' intentions to collaborate with pharmacists. Subsequently, semi-structured interviews and thematic analysis captured the HPs'' experiences with the model. Finally, we integrated the quantitative and qualitative findings to interpret the mechanism of the impact of the IMM. Results: This study developed an IMM model and evaluated its effectiveness, with the main findings as follows: 1. Based on the literature review, we have selected three evidence-based CPSs, medication reconciliation, medication review, and ward-based pharmacist services, encompassed in the IMM model. The model provides service at all stages (admission, hospitalization, and pre-discharge) of a patient''s hospital stay. Additionally, we established a logic model of the IMM model based on the Theory of Change to formulate evaluation indicators. 2. After implementing the IMM model, there was a significant increase in the mean number of ME reports (level change: 1.02, 95% confidence interval [CI]: 0.68 to 1.35, p< 0.001), with a notable rise in reports of omission or medication discrepancy, inappropriate drug choices, inappropriate routes or formulation, and not-in-benefit package. Additionally, there was a significant reduction in the mean number of daily inpatient prescriptions for patients aged ≥75 years (level change: -1.78, 95% CI: -3.06 to -0.50, p= 0.009). However, other indications did not show statistically significant changes. 3. Fifty-eight HPs completed the questionnaire, indicating that HPs from the intervention ward had a higher intention to discuss medication-related issues with pharmacists. Eleven HPs were interviewed, and they stated having better working relationships with pharmacists, experiencing more effective CPSs, and noting improved communication with pharmacists. The integration of quantitative and qualitative findings revealed that the critical mechanism of the IMM in promoting collaborative relationships is the integration of regular pharmacists into medical practice, which familiarizes HPs with pharmacists'' roles, improves communication, and enables pharmacists to identify HPs'' needs. Conclusion: We developed an IMM model grounded in evidence-based CPSs tailored to our specific context. This model is designed to minimize drug-related problems in hospitalized patients and improve overall care quality. Evaluation studies have shown that the IMM model notably increases the detection of medication errors and decreases medication usage in elderly patients. Following its implementation, there has been an enhancement in communication and collaboration between pharmacists and other HPs. Pharmacists are actively involved and engage directly with the medical team, improving service quality and deepening HPs'' appreciation of pharmacists'' roles. These results can guide other healthcare facilities. However, the study was limited by the small number of participants and the range of patient groups involved. Future research should assess the model''s impact across various settings and conduct thorough pharmacoeconomic research to determine its financial effectiveness.