中央健保局將採併發症/合併症結合申報的診斷關係群制度作為支付標準，不同的併發症/合併症醫療費用之給付將不同。目前醫療院所受限於自動化審核系統之功能，必須以人工方式逐一審核健保申報資料的次診斷與病摘，再將併發症/合併症編碼完整的登錄至申報資料。然而，從大量資料中找出併發症/合併症是一項勞力與知識密集的工作，且容易遺漏。因此，本研究針對全膝關節置換術（Total Knee Replacement, TKR）提出一個電腦化的併發症/合併症篩選程式，以提升該工作之準確性與效率。研究顯示，應用該程式從219位個案中找出11位疑似併發症/合併症個案之結果與疾病分類專家逐一檢視每本病歷後之結果相同。本研究成果不但可支援醫院內部TKR併發症/合併症之醫療照護品質評估外，也可提供送交申報資料前之先行稽核，以彌補應申報而未申報之收益損失。

The Bureau of National Health Insurance is going to implement a new payment mechanism which integrates complication/comorbidity with diagnosis related group system. Different complication/comorbidity cases will receive different payment rate. Due to the function limitations of the current screening program, hospitals have to identify complication/comorbidity cases from the discharge abstract manually. However, performing this work from large amount of discharge abstracts manually is a labor and knowledge intensive work and easy to make mistake. This paper presents a complication/comorbidity screen program (CSP) for total knee replacement (TKR) to perform this work electronically. Two hundreds and nineteen patient records are used to test the feasibility and validity of the program. The CSP provides the same results as those from the chart review by health care professionals. With this program, the hospital can identify TKR complication/comorbidity cases to enhance the health care quality easily. In addition, it can help hospitals audit the TKR complication/comorbidity data prior to declaring the data to the Bureau of National Health Insurance to prevent the payment loss