- 期刊
某醫學中心醫師和疾病分類人員ICD-9-CM編碼一致性之分析
The Consistency of ICD-9-CM Coding between Physicians and Coders in a Medical Center
摘要
我國於1995年3月1日起正式實施全民健康保險，為我國國民保健閉創了新的里程碑。然而實施全民健康保險面臨了醫療費用急遽高漲的困境，為解決此一問題，全民健康保險住院部份先採「論病例計酬制」取代原有的「論量計酬制」，再進而實施以診斷關聯群為塞準的前瞻性支付制度(Diagnostic Related Groups/Prospective Payment System; DRGs/PPS)，以作為未來住院醫療費用給付的塞礎，期能藉以舒緩醫療費用高漲之威脅。 然而台灣本土版DRG會對台灣的醫院管理帶來什麼樣的影響？需要醫療院所用心的去解決。首先醫院必須先解決院內疾病分類代碼的一致性？因為缺乏一致性的編碼，醫院將會因為編碼不實而可能產生巨大的損失。但是對於編碼的一致性卻一直沒有類似的研究來探討。因此本研究以某醫學中心為例，以2002年出院病患共82,312筆有效樣本數的疾病分類編碼主要診斷和醫師作的入院診斷編碼一致性，經Logistic Regression分析結果發現疾病分類編碼住要診斷不一致程度的影響因素有：有病人特質、疾病特質、醫師特質等變項。其中尤其是以「有無C/C」(Comobidity/Complication)對於編碼一致性的影響最大。 針對本研究的結果，建議各醫院應針對編碼不一致分析其原因並提出解決方案，同時要加強醫師或相關人員ICD-9-CM Coding教育，以建立病歷書寫規範以提昇病歷書寫品質，並制訂完善之審核制度，配合醫院各部科推動臨床路徑(clinical pathway)，提高疾病分類編碼品質。又可以利用資訊系統加強對申報案件的稽核，以提昇健保費用申報中疾病分類編碼之適確性。
並列摘要
The implementation of the National Health Insurance (NHI) on March, 1, 1995 has unveiled a new era for Taiwan's public health. However, like other countries Taiwan has faced dramatic increase in health expenditures. In order to solve this problem, the NHI Bureau has begun to replace traditional ”fee-for-service” with a ”case payment” method. Its final destination is a ”Perspective Payment System/Diagnostic Related Group” (PPS/DRGs) which is believed to release the pressure of soaring medical costs. However the impact of Taiwan version of DRG is still unknown to hospitals and need to be solved carefully by hospital managers. One urgent problem needed to be solved is the inconsistency of ICD coding It is mainly because that lack of consistency in ICD coding may result in tremendous financial loss. This study examines the consistency of ICD coding between physicians and ICD coders of a Medical center. Claim data of 82,306 discharged patients in 2002 were studied. The analysis is performed by applying Logistic Regression and factors that affect the consistency of coding including patient, disease and physician characteristics were studied. Results of the Logistic Regression found that all of these characteristics could significantly influence the inconsistency of ICD coding between physicians and ICD coders. Among these characteristics, with/without Co-morbidity/Complication is the variable which has the highest odds ratio. Results of this study suggest that hospital administrators have to enhance ICD coding training, or on-job continuously ICD coding training, for all physicians, nurses and ICD coders so that a hospital may not suffer the consequences from inconsistency of ICD coding. Besides, hospital administrators have to aware that too much inconsistency in ICD coding may have significantly negative impact on cost containment and implementation of clinic pathways, which are believed to be effective methods for improving quality and cost performance of a hospital.