臺灣2006~07年不同經濟狀況民眾事故傷害住院分析

Characteristics of Hospitalizations Caused by Injury by Different Levels of Family Income in Taiwan, 2006-2007

目的：比較低收入戶及非低收入戶傷患住院原因、傷病特質、醫療利用與住院預後。方法：使用2006-07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)」、「醫事機構基本資料檔(HOSB)」與「承保抽樣檔(ID)」，選取ICD-9-CM中N Code、E Code為事故傷害的個案，並以統計軟體SPSS 18.0進行分析。結果：台灣2006-07年低收入戶傷患住院人數為13,860人，住院率為每10萬人3,189.1，男性住院率（每10萬人4,172.7）高於女性（每10萬人2,217.2），65歲以上老年人住院率遠高於其他年齡層，住院率前三名的地區為花蓮縣、宜蘭縣及台中縣，主要住院原因為交通事故(E800-E848)、跌倒墜落(E880-E888)，就醫層級以區域醫院為主。與非低收入戶比較，低收入戶住院傷患的合併罹患病症較多且嚴重、使用精神科的比例較高、住院天數較長、醫療花費較多、住院預後較差。以多變項邏輯斯回歸分析，結果顯示低收入戶傷患住院死亡風險是非收入戶的1.674倍，整體模式的解釋力為37.5%。結論：低收入傷患住院的死亡風險較非低收入傷患高，未來除了醫療體系之因應外，仍需有關單位擬定改善經濟之合適政策，例如就業輔導、提供工作機會來增加收入等方式，或能局部消弭低收入戶健康不平等的現象。

Objective: The aim of this study was to determine the causes of injury, patient characteristics, medical utilization and prognosis for inpatients hospitalized as a result of injury by different levels of family income. Methods: This research gathered data about “inpatient expenditures by admissions”, “registry for contracted medical facilities” and “registry for beneficiaries” from the National Health Insurance Research database for 2006-2007 and analyzed it with SPSS 18.0 software. We defined injury as ICD-9-CM N Code “800-999” and E Code “E800-E999”. Results: There were 13,860 low-income patients hospitalized as a result of injury in Taiwan in 2006-2007. The total hospitalization rate was 3,189.1/105. Among the various low-income groups, males had a higher hospitalization rate than females (males: 4,172.7/105, females: 2,217.2/105), the population over 65 had the highest rate, and the top three counties were Hualien County, Ilan County, and Taichung County. The major causes of injury were traffic accidents (E800-E848) and falls (E880-E888), and most of the cases were hospitalized in local hospitals. Compared with non-low-income families, inpatients with low-income had more comorbid conditions, higher CCI (Charlson Comorbidity Index) scores, more consultations from the psychiatry department, longer lengths of stay more medical costs, and worse prognoses. The risk of mortality for low-income inpatients was 1.674 times that of non-low-income ones (OR=1.674, p<0.001) by multivariate logistic regression analysis, and the overall model explained 37.5% of the variance. Conclusions: The risk of mortality for low-income inpatients was higher than that for non-lowincome inpatients. In order to eliminate health inequality in low-income households, we should provide medical services, employment assistance and job opportunities in order for them to increase their income.