*中心館藏網址:http://tulips.ntu.edu.tw/record=b2171096*cht  

[ 摘要 ]
近十年來花蓮地區的事故傷害一直是台灣事故傷害情況較為嚴重者之一， 而花蓮地區的原住民較多、飲酒問題亦較嚴重。本研究之主要目的在於探討事故傷害急診病患與其他急診病患的飲酒狀況。病患來自花蓮最大之綜合醫院急診室。以問卷調查及血液酒精濃度測試方式收集770名急診個案資料（事故傷害個案226名、非事故傷害個案544名）。本研究個案中平時偶爾喝者佔11.2%（85名），而經常喝或幾乎天天喝者有9.1%（69名）。事故傷害個案平時飲酒的比例較非事故傷害個案為高。在本研究個案中，有57名（7.4%）的個案於入院時有聞到酒味、79名個案（10.4%）自述於此次來院前有飲酒。事故傷害個案聞到酒味的比例及自述其來院前有飲酒的比例均較非事故傷害個案為高。而機動車事故傷害個案與其他事故傷害個案在聞到酒味的比例及自述其來院前有飲酒的比例上均無明顯差異。以邏輯斯迴歸分析就診原因為「事故傷害」的危險性，其結果發現：來院時血中酒精濃度≧50㎎/dL的個案為來院時血中酒精濃度＜50㎎/dL的個案的5.2679倍；平常偶爾飲酒者的危險性是平常不飲酒者的0.0829倍；平常常常喝酒及幾乎天天飲酒者的危險性是平常不飲酒個案的0.4676倍；「有非事故傷害急診經驗個案」為事故傷害個案的危險性是「無急診經驗個案」的0.2239倍；「有事故傷害急診經驗個案」的危險性是「無急診經驗個案」的3.5273倍。另外針對事故傷害個案，以就診原因為「機動車事故傷害」的危險性來看，非原住民的危險性為原住民的2.1843，學生的危險性為無業、家庭主婦者的3.4317倍、學生的危險性為有工作者的4.3860倍。如將學生排除於分析之外後發現，事故傷害個案中，非原住民其為機動車事故傷害個案 的危險性為原住民的2.6788倍。以「血中酒精濃度」為依變項進行邏輯斯迴歸分析則發現，事故傷害急診個案來院時血中酒精濃度≧50㎎/dL的機會為非事故傷害急診個案的4.9737倍。而急診個案中，平時常常飲酒者來院時血中酒精濃度≧50㎎/dL的機會為平時不飲酒者的17.9675倍，平時偶爾飲酒者來院時血中酒精濃度≧50㎎/dL的機會為平時不飲酒者的2.0265倍。

[ 英文摘要 ]
The purpose of this study is to investigate the differences of alcohol use between the injury patients and the non-injury patients at emergency room. Motor vehicle injured patients were also compared with other injury patients. Data were collected from 770 patients (including 226 injury patients and 544 non-injury patients) who were admitted at the emergency room of the largest general hospital in Hualien with questionnaires and blood alcohol tests. It was found that 9.1% of over all patients had daily drinking habits,11.2% of patients drank occasionally. Injury patients were more likely to have drinking experiences than non-injury patients .There was no difference in drinking habits between motor vehicle injury patients and other injury patients. When the patients came into the emergency room, 7.4% of them had alcohol smell, and 10.4% of them said that they had drunk before the incidents. Injury patients were more likely to have alcohol smell and to have being drinking before the incidents than non-injury patients. Motor vehicle injury patients were not different from other injury patients on this issue. Logistic regression model show that , the patients that had blood alcohol over 50㎎/dL were more likely to be injury patients than the patients that had blood alcohol less 50㎎/dL（OR = 5.2679 ）; the patients that has non-injury emergency experiment were less likely to be injury patients than those do not have emergency experiment （OR = 0.2239 ）; the patients that has injury emergency experiment were more likely to be injury patients than those do not have emergency experiment（OR = 3.5273）.Among injury patients, the non-aboriginals had an OR of 2.1843 for motor vehicle injury in comparison with aboriginals, the student had an OR of 3.4317 for motor vehicle injury in comparison with the patients without a job, and an OR of 4.3860 for motor vehicle injury in comparison with working patients. After exclude the students, the non-aboriginal patients had an OR of 2.6788 for motor vehicle injury in comparison with aboriginal patient. Taking 50㎎/dL as a cutting point for blood alcohol level , the injury patients were more likely to have a blood alcohol over 50㎎/dL than non-injury patients（OR= 4.9737）,and the patients who used to drink occasionally or daily were more likely to have a blood alcohol over 50㎎/dL than patients who never drank（OR= 2.0262 and OR=17.9675）