比較同意或拒絕由醫學中心急診轉入區域醫院住院之病患特性與相關預後

前言：自全民健保實施以來，伴隨而來重要的影響是國人的就醫習慣，導致輕症病患至全國各大醫學中心急診就診；持續失衡的結果，逐漸使得真正「急」和「重」的病患無法適當的得到醫療資源和照護。為改善醫學中心急診壅塞、病患待床時間過長問題，衛生福利部協同台北市政府衛生局於104年2月開始實行「台北市醫學中心急診待床轉院計畫」，其初步的滿意度雖已達正面的效果，但國內在接受轉院和拒絕轉院的病患特色分佈、病患接受適當的下轉至區域醫院層級之醫院接受後續的治療成效及其與續留醫學中心接受治療的病患之住院天數及死亡率是否有差異皆未有研究探討。研究目的：探討北部某家醫學中心：(1)醫學中心符合急診待床轉院計畫病患之基本屬性與疾病特性；(2)醫學中心急診病患同意轉至區域醫院住院過程中，轉入的時間、轉入醫院及等待轉入時間對預後的影響；(3)比較轉往區域醫院接受治療的病患和拒絕轉院續留醫學中心接受治療的病患，兩者間候床天數、動向死亡率及住院天數的差異。研究方法：此研究為回溯式世代(retrospective cohort)設計，以緊急傷病患電子轉診系統(Mars)、北區某醫學中心醫院電子病歷檔案和區域醫院紙本病歷回覆為資料之來源，收案期間為2016/01/01~2016/12/31，經北區某醫學中心醫療團隊評估後可由急診轉台北市立聯合醫院住院之所有的成人病患為研究對象，排除直接轉入加護病房、安寧病房及手術病患等，以描述性統計分析各變項，以相關性統計、傾向分數配對及多變項邏輯斯迴歸分析兩群體間預後（死亡率、住院天數）的差異。研究結果：總轉出率26.9%，其中有效個案數為858人，拒絕轉院續留醫學中心治療個案共438人，轉往聯合醫院為420人，聯合醫院病歷回覆數為283人。相較拒絕轉至區域醫院續留醫學中心接受治療的病患，影響轉院重要因素為年齡(OR=0.984, p<0.001)、檢傷平均動脈壓(OR=1.008, p = 0.047)及檢傷到詢問時間(OR=0.998, p <0.001)，意指當病患年紀越年輕、生命徵象越趨於穩定和及早介入詢問轉院意願其接受轉院的程度越高。與預後相關的死亡分析結果指出：當共病指數越高(OR=1.789, p <0.001)、有不實施心肺復甦/安寧緩和註記者(OR=45.408, p <0.001)和疾病診斷有改變者(OR=10.212, p= 0.005)其較容易有死亡發生，有無轉院則未發現統計上顯著影響。整體總住院平均天數為10.7天，研究結果中發現當共病指數越高(β=0.148 , p <0.001)、出院診斷是肺炎(β= 0.106 , p＝0.003)和蜂窩性組織炎者(β=0.075 ,p＝0.03)、後續醫療處置越多(β=0.32, p <0.001)、等待轉入時間較長(β=0.243, p <0.001)，總住院天數將隨之增加；反之，急診出院(β= -0.57, p <0.001)與轉院(β= -0.123, p＝0.005)則總住院天數會減少。結論：本研究為臨床醫療照護模式提供重要資訊，在醫學中心符合急診待床轉院計畫中年齡、生命徵象與檢傷到詢問時間為影響病患轉院意願的相關因子，更從而探討出轉院至區域醫院住院過程中，轉入的時間、轉院和等待轉入時間與總住院天數的減少有顯著的相關，但與死亡預後無關。

關鍵字

急診壅塞；急診暫留時間；轉院；預後

並列摘要

Background: Patients’ preferences concerning medical utilities in Taiwan have shifted greatly owing to implementation of National Health Insurance. People now tend to visit emergency departments in medical centers even for trivial problems, invariably causing overcrowding and prolonged waiting times in emergency departments. Thus, the “Taipei Medical Center Emergency Department Referral Program” was implemented in February 2014. Patients waiting for admission were suggested to transfer from the emergency department in a medical center to an admission ward in a regional hospital following their evaluation. This program has received positive feedback. However, the characteristics of patients accepting or refusing referrals, the duration of hospitalization and outcomes remain unclear. Purpose: The aims of the study are to (1) characterize patients accepting or refusing referral in the “Taipei Medical Center Emergency Department Referral Program” ; (2) examine the influence of admission timing, emergency boarding time and the receiving regional hospital on clinical outcomes; and (3) determine the differences in mortality and duration of hospitalization between patients who accept or refuse referral from a medical center to a regional hospital. Methods: This was a retrospective cohort study, utilizing data from the urgent patient electronic referral system “Mars”, electronic medical records from one medical center in Taipei and hard copies of referral sheets from receiving regional hospitals. The study included adult patients who were evaluated and considered appropriate for transferring during 1 January to 31 December 2016. We analyzed the prognosis (mortality rate and duration of hospitalization) between patients accepting and refusing referral from the medical center to a regional hospital using multiple logistic regression analysis after propensity score 1:1 matching. The factors related to patient transfers and their influences on outcomes were studied. Results: The rate of transfer to regional hospitals was 26.9%. There were 858 patients evaluated and suggested to transfer to a regional hospital. Of these, 420 patients accepted the referral. Medical records were completed and sent back from regional hospitals for 283 patients. The factors associated with refusing referral included age more than 65 years [odds ratio OR: 0.984, p < 0.001], lower triage mean arterial blood pressure (OR: 1.008, 95%, p = 0.047) and longer duration from triage to request for transferring time (OR: 0.998, p < 0.001). After propensity score matching, inter-hospital transfer was not associated with a difference in in-hospital mortality (p = 0.541). The factors related to in-hospital mortality were higher Charlson comorbidity index (OR: 1.789, p < 0.001), having a “do not resuscitate” order (OR: 45.408, p < 0.001) and having a diagnosis change (OR: 10.212, p = 0.005). The factors related to longer hospital length of stay included higher Charlson comorbidity index (β = 0.148, p < 0.001), discharge diagnosis of pneumonia (β = 0.106, p = 0.003) and cellulitis (β = 0.075, p = 0.03), and more subsequent management after suggesting the transfer (β = 0.32, p < 0.001). However, inter-hospital transfer was associated with shorter total hospital length of stay (β = −0.123, p = 0.005) as well as with direct discharge from emergency department (β = −0.57, p < 0.001). Conclusions: This retrospective propensity score-matching cohort study delineates the age, vital signs, duration from triage to request for transferring time are the major factors which affecting the patient’s willing of transfer. Transferring from an overcrowded emergency department in a medical center to a regional hospital is associated with better outcomes of shorter duration of hospitalization without effects on in-hospital mortality.

並列關鍵字

Emergency department overcrowding ； Emergency boarding time ； Inter-hospital transfer ； Outcomes

參考文獻

Charlson, M. E., Pompei, P., Ales, K. L., & MacKenzie, C. R. (1987). A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of Chronic Diseases, 40(5), 373-383.

中文部分