台灣地區機動車事故死亡率之年齡、年代及世代效應分析
Age, Period and Cohort Effect Analyses of Motor Vehicle Mortality in Taiwan
中文摘要
台灣地區直到近年才逐漸重視事故傷害的預防與控制，許多事故傷害防制的策略 也在近年來逐步執行，包括安全帽法案、傷害防制研究增加等。此研究的目的在 藉由年齡、年代及世代效應之模式分析，比較台灣地區主要傷害死亡率是否減緩。
本研究以 1972 年到 2001 年衛生署的死亡資料檔及內政部的人口統計數字作分 析。先以圖形描述的方法，分析台灣地區三十年來死亡率的趨勢變化，另外，使 用卜瓦松線性迴歸分析法加上 Osmond 氏所修正過的年齡-年代-世代迴歸模式，
分析年齡、年代及世代效應的個別影響。以及利用年代與地區的分層分析，比較 在不同年代與地區的年齡、年代及世代效應是否不同。最後綜合以上分析探討造 成趨勢變化的因素。
本研究結果發現，青少年（15-19）與老年人（70 歲以上）為最危險族群，且年 輕族群之危險族群有年齡下滑的現象，男性年輕族群由 25-29 歲，女性由 20-24 歲下降到 15-19 歲年齡層。且藉由年代分層分析，發現女性年輕族群（15-29 歲）
危險性的上升較男性更高。在控制年齡及世代效應之後，年代效應確實影響機動 車事故死亡率。在 1997 到 2001 之間，機動車事故死亡率的危險性下降到三十年 前的水準。而地區分層之後的比較中發現，都會地區的年代效應趨勢下滑，而鄉 村地區卻呈現上升趨勢。然而，在 1997 年以後的趨勢卻一致下滑。而不同地區 的世代效應在老年族群出現反向差異，但年輕世代的世代效應趨勢卻趨於一致。
本研究建議，應加強 15-19 歲青少年的機動車傷害防制，以及城鄉之間醫療資源 的平衡，並應積極透過教育及播傳的途徑，建立民眾傷害防制的觀念。此外，女 性危險性增加大於男性的原因應進一步研究。其他事故傷害的年齡、年代、世代 效應的分析，也必須進一步研究。
英文摘要
Recently in Taiwan, the importance of injury prevention and control has been placed on great emphasis. Different strategies of injury prevention and controls have been implemented including the helmet law and increasing injury prevention research.
The purpose of this study is to analyze the trend of motor vehicle (MV) mortality by the method of Age-Period-Cohort model analysis.
The national death record was obtained from the Department of Health, and the national census data was acquired from the Ministry of Interior (1972-2001). The trend of the MV mortality was described via means of the graphical method. The age, period, and cohort effect was analyzed respectively using the Log-Linear model
modified by Osmond. Stratified analyses were also performed utilizing the period and area stratified. Finally, the results of these analyses used for the interpretation of the relative factor of the MV mortality trend were combined.
The results showed that both the adolescents (15-19 years old) and elders (beyond 70 years old) were the highest group at risk for MV mortality. Among the younger age group, the riskiest age group decreased from 25-29 years old and 20-24 years old to 15-19 years old in men and women respectively. The increasing MV mortality risk in female was higher than in male utilizing the stratified period. The motor vehicle mortality was obviously influenced by the period effect and is evident through the significant decrease in the risk of MV mortality in Taiwan during 1997 through 2001 to a level equal to that of 30 years ago. By the stratified area, the MV mortality trend in urban area decreased continuously, but the trend in rural area increased. However after 1997, the trends both decreased. The trends of cohort effect in rural area increased continuously and urban area decreased continuously at elder cohorts. However the trends of cohort effect at young cohort in both rural and urban areas were similar.
Our findings suggested that the prevention and control of MV mortality in
adolescents must improve, and the medical resource between urban and rural area must progress towards some sort of equilibrium. Education and mass
communication is also an important way for people to obtain and understand the concept of injury prevention. There are also many other serious injuries in Taiwan for example suicide and falls. The age, period, and cohort effect in these injuries are still unclear therefore more research about them is needed.
