背景：傳統死因統計使用單一原死因進行統計製表，許多死亡證明書上死因相關訊息因此無法被充分利用，因此許多學者主張用多重死因分析來解決此限制。台灣是高肝硬化死亡率國家，可惜相關研究不多。 目標：本研究希望瞭解：1.提及肝硬化死亡證明書填寫數目是否隨年代增加而增加？2.以多重死因方法所呈現的肝硬化死亡率改變趨勢是否不同於單一原死因的死亡率改變趨勢？3.肝硬化死亡伴隨併發症的型態是否隨年代改變而改變？4.肝硬化與其他疾病的相關性是否隨年代改變而改變？5.醫師在哪些情況下會將肝硬化併發症填到死亡證明書上？ 方法：抽樣1987, 1992, 1997三年5%死亡證明書，將死亡證明書上每個診斷都譯碼ICD-10代碼並輸入ACME/TRANSAX電腦軟體選擇原死因及產生多重死因統計。首先分析每張死亡明書開具病症數分布之年代別性別年齡別差異；其次比較單一原死因統計與多重死因統計推估死亡率趨勢改變之差異；第三部份為肝硬化與併發症與其他疾病間相關組合之跨時分析；最後以醫院樣本比較病歷與死亡證明書之填寫內容。 結果：肝硬化相關死亡證明書每張填寫死因診斷數隨著年代增加而增加，女性比男性平均診斷數多，年齡越高診斷數也越多。以所有提及肝硬化所計算出來的死亡率上升趨勢比以單一原死因死亡率趨勢陡峭，比值在女性及高齡者較大，表示兩群中肝硬化致命性較低。肝硬化伴隨肝癌比例在女性逐年上升，男性小於等於54歲者出現併發症比例最高。 結論：多重死因分析統計可以呈現許多不同於單一原死因統計的有用訊息，尤其對於肝硬化併發症負擔與疾病組合可以提供相當有用的參考，醫院病歷資料分析顯示，醫師填寫併發症與相關疾病還是有相當考慮致死機轉，因此可以反映嚴重度，但是無法反映盛行率。 Background: Traditionally, mortality data are tabulated according to single underlying cause of death (UCD). Many cause of death related information filled on the death certificates were not fully used. Multiple causes of death (MCD) approach was proposed by scholars to solve the above problem. The mortality of liver cirrhosis (LC) was high in Taiwan, nevertheless, related studies were few. Objectives: This study aimed 1) to assess if the number of conditions reported on LC related deaths increased across years; 2) to determine if the pattern of changes in LC mortality rate differed between those according to single UCD and to MCD approach; 3) to examine if the complications associated with LC changed across years; 4) to illustrate if the association of LC with main causes of death changed across years; 5) to explore in what kind of circumstances would the physicians reported LC complications on the death certificates. Method: Five percent of death certificates issued in the year 1987, 1992, and 1997 were extracted for analyses. All diagnoses filled on the death certificates were coded and entered into ACME/TRANSAX to select the UCD and produce MCD statistics. Percentage distributions of conditions on one death certificate by year, sex, age and diagnoses were calculated. Trends of death rate according to UCD and MCD were compared. Associations between diagnoses on death certificate were also analyzed by chi-square test. Hospital medical records were reviewed to explore the logic why physicians reported some complications on death certificates. Results: Average number of conditions reported on one death certificate increased with years, women had higher number than men. The number also increased while the age increase. The number and rate of death of which liver cirrhosis was selected as total mentioned increased obviously than as the underlying cause of death (UCD), especially in women and old-aged. That means that the rate of mortality due to liver cirrhosis is lower in these two groups (women and old-aged). The rate of liver cirrhosis accompanied by hepatocellular carcinoma (HCC) in women increased with years. The rate of complication in male less than 54 y/o is highest. Conclusions: As MCD statistics did provide more useful information than UCD statistics, especially in the estimating the burden of complications of liver cirrhosis and disease relationship. According to the database of chart of Chung Shan Hospital , Doctors have considered the mechanism of disease and wrote it down to certificate. Therefore, the analysis of death certificate can show the severity of disease, but can not reveal the prevalence.
