目的：找出家暴受虐住院的高危險族群，瞭解家暴受虐住院的醫療利用情況，並探討影響家暴受虐住院死亡的因子。方法：使用2006-2007年全民健康保險資料庫，串連其中的「住院醫療費用清單明細檔(DD)」與「醫事機構基本資料檔(HOSB)」，選取ICD-9-CM中N Code爲995.5x、995.8x，E Code爲E967.x的家暴受虐個案，並以統計軟體SPSS 18.0進行分析。結果：2006-2007年台灣地區有659人因家暴受虐住院（男性37.03%、女性62.97%），總醫療花費爲新台幣33,331,171元，住院費用的中位數爲新台幣15,585元。整體的住院率爲每10萬人1.44，女性住院率（每10萬人1.83）高於男性（每10萬人1.05），0-4歲的嬰帅兒住院率遠高於其他年齡層，住院率前三名的縣市爲台南市、高雄縣、屏東縣。家暴受虐住院傷患中25.34%（167人）看外科、40.97%（270人）在區域醫院住院、平均每人的查爾森合併症嚴重度指標(Charlson Comorbidity Index, CCI)爲0.25分、57.51%的傷患接受手術處置，3%（21人）住院死亡。以多變項邏輯斯回歸(Multivariate Logistic Regression)分析，結果顯示影響家暴受虐傷患住院死亡的因子計有共伴傷病（查爾森合併症嚴重度指標）、就醫科別、手術處置，即共伴傷病越嚴重、在神經外科住院及有接受手術處置的傷患，住院死亡的風險越高，整體模式的解釋力爲33.2%。結論：未來研究可結合門急診資料、死亡統計及家暴通報系統，將能完整呈現家暴受虐傷患的傷害流行病學全貌，以提供相關單位擬定防制策略參考。

Objectives: Finding the high-risk groups of death caused by domestic violence inpatient and medical services utilization of hospitalized domestic violence patients, and the factors associated with hospital mortality caused by domestic violence injuries. Methods: This research brought "inpatient expenditures by admissions (DD) " and "registry for contracted medical facilities (HOSB)" from national health insurance research database in 2006-2007 into analysis by using SPSS 18.0 software. We defined domestic violence cases as ICD-9-CM N Code 995.5x, 995.8x and E967.x. Results: There were 659 inpatient cases caused by domestic violence injuries (males 37.03%, females 62.97%), and their total medical expenditures was NT 33,331,171 dollars, the median of medical expenditures was NT 15,585 dollars in Taiwan, 2006-2007. The total inpatient rate was 1.44 per 100,000, and females were higher than males (males 1.05 per 100,000, females 1.83 per 100,000). The inpatient rate of child from 0 through 4 years old was higher than any others. Tainan City had the highest inpatient rate, followed by Kaohsiung County, Pingtung County and so on. Among hospitalized domestic violence patients, 25.34% (167 people) were consulted at surgical department, 40.97% (270 people) were hospitalized in the regional hospital. Each person had 0.25 CCI (Charlson Comorbidity Index) scores in the average, and 57.51% of domestic violence patients received surgeries, operations or treatments. The hospital mortality caused by domestic violence was 21 people. The factors associated with hospital mortality caused by domestic violence were comorbidity (Charlson Comorbidity Index), division of medicine, and surgeries by multivariate logistic regression analysis, that is, inpatient cases due to domestic violence who had more serious comorbidity, consulted at department of neurosurgery, or received surgeries, operations or treatments had higher hospital mortality risk; the explanation of overall model was 33.2%. Conclusions: In order to make the programs intervening domestic violence effectively, we should describe the epidemiology of domestic violence completely by linking outpatient, emergency room registry, vital statistics system and national domestic violence database in future research.