2004-2016年珠海市道路交通伤害时间序列分析
|
摘要
目的建立珠海市道路交通伤害(road traffic injuries,RTIs)发生的时间序列模型,了解RTIs的时间变化规律。方法对2004-2016年珠海市3家哨点监测医院RTIs病例发生时间进行描述性分析。以2004-2015年RTIs按发生年、月份建立自回归移动平均混合模型(ARIMA),以2016年资料进行验证;同时按RTIs发生星期、时点(24 h)建立ARIMA模型进行时间序列分析。结果 2004-2016年共监测到70 813例RTIs。1-2月份RTIs较少,7月达高峰;星期一、六和日RTIs发生数量较多;18-21点RTIs呈现最高峰,7-9点、0-2点分别呈现次高峰。构建得到RTIs发生人数按年、月份的ARIMA(0,1,1)模型(Ljung-Box检验Q=16.586,P=0.413),预测2016年RTIs人数,预测值与实际观测结果较相符;随着预测时间延长,CI范围扩大。对RTIs发生星期及时点序列分析建模为ARIMA(1,0,0)(Ljung-Box检验Q=13.283,P=0.652),观测值与拟合值基本相符。结论 2004-2016年珠海市RTIs流行状态具有一定时间变化规律,ARIMA模型适合进行RTIs发生时间趋势拟合并进行短期预测分析。
Objective To establish an autoregressive integrated moving average(ARIMA) model of time series analysis in predicting road traffic injuries(RTIs) based on the surveillance data in Zhuhai City, and to understand the changing rule of occurrence time of RTIs. Methods Descriptive epidemiological analysis was conducted on the occurrence time of RTIs collected from 3 sentinel monitoring hospitals in Zhuhai City from 2004 to 2016. An ARIMA model was established according to years and months of occurrence of RTIs during 2004-2015, and tested by the data about RTIs in 2016. And according to weeks and time points of occurrence of RTIs, another ARIMA model was simultaneously established to perform a time series analysis. Results We had totally monitored 70,813 cases of RTIs in 3 sentinel monitoring hospitals in Zhuhai City during 2004-2016. RTIs were fewer in January to February, but peaked in July. Most of the RTIs occurred on Monday, Saturday and Sunday. RTIs peaked at eighteen and twenty-one o'clock, and the second spike happened between seven and nine o'clock as well as between zero and two o'clock. According to months and years of occurrence of RTIs, the fitting model was ARIMA(0,1,1), and the Ljung-Box test for the model was not statistically significant(Q=16.586, P=0.413). The predictive analysis results of RTIs in 2016 showed that the predictive values were similar to the actual observed values, and the scope of the confidence interval expanded as the prediction time extended. According to weeks and time points of occurrence of RTIs, the fitting model was ARIMA(1,0,0), and the Ljung-Box test for the model was not statistically significant(Q=13.283, P=0.652). The observed values were basically similar to the fitted values. Conclusions There were certain changing rules of time of RTIs in Zhuhai City during 2004-2016. ARIMA model is suitable for fitting the trend of occurrence time of RTIs and performing short-term forecast analysis.
Objective To establish an autoregressive integrated moving average(ARIMA) model of time series analysis in predicting road traffic injuries(RTIs) based on the surveillance data in Zhuhai City, and to understand the changing rule of occurrence time of RTIs. Methods Descriptive epidemiological analysis was conducted on the occurrence time of RTIs collected from 3 sentinel monitoring hospitals in Zhuhai City from 2004 to 2016. An ARIMA model was established according to years and months of occurrence of RTIs during 2004-2015, and tested by the data about RTIs in 2016. And according to weeks and time points of occurrence of RTIs, another ARIMA model was simultaneously established to perform a time series analysis. Results We had totally monitored 70,813 cases of RTIs in 3 sentinel monitoring hospitals in Zhuhai City during 2004-2016. RTIs were fewer in January to February, but peaked in July. Most of the RTIs occurred on Monday, Saturday and Sunday. RTIs peaked at eighteen and twenty-one o'clock, and the second spike happened between seven and nine o'clock as well as between zero and two o'clock. According to months and years of occurrence of RTIs, the fitting model was ARIMA(0,1,1), and the Ljung-Box test for the model was not statistically significant(Q=16.586, P=0.413). The predictive analysis results of RTIs in 2016 showed that the predictive values were similar to the actual observed values, and the scope of the confidence interval expanded as the prediction time extended. According to weeks and time points of occurrence of RTIs, the fitting model was ARIMA(1,0,0), and the Ljung-Box test for the model was not statistically significant(Q=13.283, P=0.652). The observed values were basically similar to the fitted values. Conclusions There were certain changing rules of time of RTIs in Zhuhai City during 2004-2016. ARIMA model is suitable for fitting the trend of occurrence time of RTIs and performing short-term forecast analysis.
引文
[1] WHO.Health in 2015:from MDGs,millennium development goals to SDGs,sustainable development goals [R].Geneva:World Health Organization,2015:174.
[2] WHO.Global status report on road safety 2015 [R].Geneva:World Health Organization,2015:vi-xii,2-4.
[3] 邓晓,杨超,段蕾蕾.2015年全国伤害监测道路交通伤害病例分布特征分析[J].道路交通科学技术,2017,3(1):35-39.
[4] 公安部交管局.2016年全国机动车和驾驶人保持快速增长[J].道路交通管理,2017,2(1):7.
[5] 龚思红,李德云,梁小冬,等.2008-2010年广东省珠海市哨点监测医院交通伤害住院病例分析[J].中国健康教育,2012,28(4):335-337.
[6] WHO.World report on road traffic injury prevention[R].Geneva:World Health Organization,2004:200.
[7] 邓晓,吴春眉,蒋炜,等.2006-2008年全国伤害监测道路交通伤害病例分布特征分析[J].中华流行病学杂志,2010,31(9):1005-1008.
[8] 梁文杰,韦波,黄开勇,等.广西三城市2000-2009年道路交通伤害死亡的经济负担研究[J].中华疾病控制杂志,2015,19(4):361-363.
[9] 周林,武献锋,刘守钦,等.济南市道路交通伤害死亡特征与疾病负担分析[J].现代预防医学,2016,43(23):4394-4396.
[10] 顾月,柯维夏,崔梦晶,等.中美道路交通伤害的比较分析[J].伤害医学(电子版),2012,1(2):33-36.
[11] 王伶,姚文清.利用时间序列模型分析预测辽宁手足口病疫情趋势[J].中国卫生统计,2016,33(5):847-849.
[12] 周美兰,周志华,罗美玲,等.湖南省哨点医院流感样病例SARIMA模型预测[J].实用预防医学,2018,25(3):370-373.
[13] Mehmandar M,Soori H,Mehrabi Y.Predicting and analyzing the trend of traffic accidents deaths in Iran in 2014 and 2015[J].Int J Crit Illn Inj Sci,2016,6(2):74-78.
[14] Chandran A,Pérez-Nunez R,n Bachani AM,et al.Early impact of a national multi-faceted road safety intervention program in Mexico:results of a time-series analysis[J].PLoS One,2014,9(1):e87482.
[15] 庞媛媛,张徐军,涂志斌,等.自回归移动平均混合模型在中国道路交通伤害预测中的应用[J].中华流行病学杂志,2013,34(7):736-739.
[16] 高景宏,朱瑶,熊黎黎,等.汕头市某三甲医院2002-2012年交通伤害病例的时间序列分析[J].中华疾病控制杂志,2014,18(10):917-921.
[17] 叶鹏鹏,邓晓,高欣,等.2006-2013年全国伤害监测系统中儿童道路交通伤害病例变化趋势及现况特征分析[J].中华流行病学杂志,2015,36(1):7-11.
[18] 黄开勇,杨莉,王晓敏.集中度和圆形分布法分析道路交通伤害发生时间规律研究[J].应用预防医学,2014,20(1):11-14.
[19] McWade CM,McWade MA,Quistberg DA,et al.Epidemiology and mapping of serious and fatal road traffic injuries in Guyana:results from a cross-sectional study[J].Inj Prev,2017,23(5):303-308.
[20] 戴璟,杨云娟.中国2006-2010年机动车道路交通伤害中驾驶员违法行为研究[J].中华流行病学杂志,2015,36(6):603-606.
[21] 陈芳,罗乐,杨傲,等.机动车道路交通伤害危险因素的Meta分析[J].中国循证医学杂志,2014,14(12):1434-1441.
[2] WHO.Global status report on road safety 2015 [R].Geneva:World Health Organization,2015:vi-xii,2-4.
[3] 邓晓,杨超,段蕾蕾.2015年全国伤害监测道路交通伤害病例分布特征分析[J].道路交通科学技术,2017,3(1):35-39.
[4] 公安部交管局.2016年全国机动车和驾驶人保持快速增长[J].道路交通管理,2017,2(1):7.
[5] 龚思红,李德云,梁小冬,等.2008-2010年广东省珠海市哨点监测医院交通伤害住院病例分析[J].中国健康教育,2012,28(4):335-337.
[6] WHO.World report on road traffic injury prevention[R].Geneva:World Health Organization,2004:200.
[7] 邓晓,吴春眉,蒋炜,等.2006-2008年全国伤害监测道路交通伤害病例分布特征分析[J].中华流行病学杂志,2010,31(9):1005-1008.
[8] 梁文杰,韦波,黄开勇,等.广西三城市2000-2009年道路交通伤害死亡的经济负担研究[J].中华疾病控制杂志,2015,19(4):361-363.
[9] 周林,武献锋,刘守钦,等.济南市道路交通伤害死亡特征与疾病负担分析[J].现代预防医学,2016,43(23):4394-4396.
[10] 顾月,柯维夏,崔梦晶,等.中美道路交通伤害的比较分析[J].伤害医学(电子版),2012,1(2):33-36.
[11] 王伶,姚文清.利用时间序列模型分析预测辽宁手足口病疫情趋势[J].中国卫生统计,2016,33(5):847-849.
[12] 周美兰,周志华,罗美玲,等.湖南省哨点医院流感样病例SARIMA模型预测[J].实用预防医学,2018,25(3):370-373.
[13] Mehmandar M,Soori H,Mehrabi Y.Predicting and analyzing the trend of traffic accidents deaths in Iran in 2014 and 2015[J].Int J Crit Illn Inj Sci,2016,6(2):74-78.
[14] Chandran A,Pérez-Nunez R,n Bachani AM,et al.Early impact of a national multi-faceted road safety intervention program in Mexico:results of a time-series analysis[J].PLoS One,2014,9(1):e87482.
[15] 庞媛媛,张徐军,涂志斌,等.自回归移动平均混合模型在中国道路交通伤害预测中的应用[J].中华流行病学杂志,2013,34(7):736-739.
[16] 高景宏,朱瑶,熊黎黎,等.汕头市某三甲医院2002-2012年交通伤害病例的时间序列分析[J].中华疾病控制杂志,2014,18(10):917-921.
[17] 叶鹏鹏,邓晓,高欣,等.2006-2013年全国伤害监测系统中儿童道路交通伤害病例变化趋势及现况特征分析[J].中华流行病学杂志,2015,36(1):7-11.
[18] 黄开勇,杨莉,王晓敏.集中度和圆形分布法分析道路交通伤害发生时间规律研究[J].应用预防医学,2014,20(1):11-14.
[19] McWade CM,McWade MA,Quistberg DA,et al.Epidemiology and mapping of serious and fatal road traffic injuries in Guyana:results from a cross-sectional study[J].Inj Prev,2017,23(5):303-308.
[20] 戴璟,杨云娟.中国2006-2010年机动车道路交通伤害中驾驶员违法行为研究[J].中华流行病学杂志,2015,36(6):603-606.
[21] 陈芳,罗乐,杨傲,等.机动车道路交通伤害危险因素的Meta分析[J].中国循证医学杂志,2014,14(12):1434-1441.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |