摘 要 以六標準差(Six Sigma, 6σ) 來改善夜間住診檢體檢驗時效，減少醫師診斷與處置之延 遲。以南部某醫學中心檢驗醫學部門為研究場所，採用D-M-A-I-C (界定－衡量－分析－改善 －管制) 推動模式，(一) 界定階段，大夜檢體量 (占當日17%) 遠小於日間，然生化和血液作 業時間相對較高，從護士站採集檢體到實驗室完成檢驗分析平均花費約114 分鐘，由內部顧 客 (醫師) 聲音轉換為品質特性，將目標設定為90 分鐘，逾時則視為一次失效；(二) 衡量階 段，根據柏拉圖分析找出時效變異潛在要因，顯示檢體實作分析占47.24%、檢體輸送占 21.91%、檢體分類占19.63%，三項作業造成流程主要變異來源；(三) 分析階段，透過特性要 因分析訂定矯正措施，包括調整實驗室人力配置與班別、修正實驗室資訊系統 (LIS) 簽收程 式、落實條碼黏貼品質、以及檢體採集交付等規範；(四) 改善階段，執行矯正措施並驗證措 施之有效性，於2009 年6 月～8 月蒐集資料，結果顯示醫師平均等候時間由114 分降至78~83 分鐘，平均差異31.69 分達顯著 (p < .001)，總流程 < 90分達標率由35.66%提升為62.51% ~ 67.89%，以六標準差評估服務失效之變異，其Sigma值由2.14σ提升為4.07σ；(五) 管制階段，獨 立樣本T 檢定顯示介入矯正措施後平均作業時間顯著縮短，卡方檢定也顯示作業逾時發生率 與介入矯正措施顯著相關。本專案未達6σ 水準，顯示流程時效尚有改善空間。

Abstract This study presented a model for improving examination efficiency for inpatients’ specimen at night by six sigma principles, adopting the D-M-A-I-C (define, measure, analyze, improve and control) approach. The laboratory medicine department of one medical center as research setting, and the efficiency was monitored and recorded with the laboratory information system (LIS) and the physician's order system (POIS) during 2009 June to August. (1) Define, the total examination flow time over 90 minutes was regarded as a failure. (2) Measure, Pareto Analysis were used to identify three operations resulting in major variation, specimen analysis, transportation and classification accounting for 47.24%, 21.91%, 19.63%, separately. (3) Analyze, corrective interventions to be created such as adjustment of laboratory staff shift, reporting quality of bar code and modified LIS program. (4) Improve, the sigma level for the failure of total process time upgraded from 2.14 σ to 4.07 σ after interventions implemented. (5) Control, the effect of intervention to be verified, and then modified the regulation of specimen delivery. The evidence suggested that our actions can stabilize the efficiency of examination process and then reduce the probability of postponing physicians’ diagnosis or treatment.