中国2008-2016年手足口病月发病率时间序列分析及预测模型

熊昱阳; 任静朝; 段广才

doi:10.16462/j.cnki.zhjbkz.2019.11.019

中国2008-2016年手足口病月发病率时间序列分析及预测模型

doi: 10.16462/j.cnki.zhjbkz.2019.11.019

1.
453003 新乡, 新乡医学院公共卫生学院
2.
450001 郑州, 郑州大学公共卫生学院

基金项目:

国家自然科学基金 81573205

详细信息

通讯作者:
段广才, E-mail: gcduan@zzu.edu.cn

中图分类号: R183;R181
计量
- 文章访问数: 464
- HTML全文浏览量: 164
- PDF下载量: 61
- 被引次数: 0
出版历程
- 收稿日期: 2019-07-02
- 修回日期: 2019-09-12
- 刊出日期: 2019-11-10

Application of the time series model in prediction of incidence of hand-foot-mouth disease from 2008 to 2016 in China

1.
School of Public Health of Xinxiang Medical University, Xinxiang 453003, China
2.
College of public health, Zhengzhou university, Zhengzhou 450001, China

Funds:

National Natural Science Foundation of China 81573205

More Information

Corresponding author: DUAN Guang-cai, E-mail: gcduan@zzu.edu.cn

摘要

摘要: 目的通过时间序列分析我国手足口病（hand-foot-mouth disease，HFMD）的发病趋势并构建时间序列预测模型，为制定防控策略提供科学依据。方法从公共卫生科学数据中心收集2008-2016年我国HFMD月发病数据，使用Excel 2007建立发病率数据库并进行图表绘制，通过SAS 9.1拟合自回归综合移动模型（autoregressive integrated moving average model，ARIMA model）。以2008年1月-2015年12月HFMD月发病率作为测试集构建时间序列模型，2016年发病率数据作为验证集检验预测效果，做出模型评价，利用该模型对2017年HFMD发病率做出预测。以P<0.05为差异有统计学意义。结果最终构建ARIMA（（12），2，0）疏系数模型，残差为白噪声序列，实际值均落在预测值95% CI内，模型回归系数有统计学意义，预测值与实际值总体吻合程度良好，均方误差平方根为3.6490，平均绝对误差=2.62，平均绝对百分比误差=28.24%。结论疏系数模型可较好的拟合我国HFMD发病率的时间序列趋势，对HFMD防控策略的制定有指导意义。
- 手足口病 /
- 疏系数模型 /
- 预测
Abstract: Objective To predict the monthly incidence of hand-foot-mouth disease (HFMD) in China by using autoregressive integrated moving average (ARIMA) model and provide evidence for prevention and control of HFMD. Methods The monthly incidence data of HFMD in China from 2008 to 2016 were collected from the Public Health Science data Center. The incidence database was established by Excel 2007 and graphed. SAS 9.1 was used to construct the ARIMA model, based on the data of the monthly reported incidence of HFMD in China from January 2008 to December 2015, and then the data in 2016 were used to verify the predicted results. The monthly incidence in 2017 was predicted in the same way.The difference was statistically significant when P<0.05. Result The model predicting monthly incidence of HFMD in China is ARIMA ((12), 2, 0) sparse coefficient and residuals is white noise. The parameters were as follows: moted mean squared error=3.6490, mean absolute error=2.62, mean absolute percentage error=28.24%. Conclusion The sparse coefficient model could well simulate the trend of HFMD case in time series, which has good reference of early warning and prevention of HFMD.
- HFMD /
- ARIMA /
- Prediction

HTML全文

图 1 测试集与差分后序列

Figure 1. Test set and differential sequence

下载: 全尺寸图片幻灯片

图 2 差分后序列自相关图与偏自相关图

Figure 2. Autocorrelation graphs and partial autocorrelation graphs of differential sequences

下载: 全尺寸图片幻灯片

图 3 2016年1月-2017年12月我国HFMD发病率趋势预测

Figure 3. Forecast of incidence trend of HFMD in China from January 2016 to December 2017

下载: 全尺寸图片幻灯片

表 1 差分后序列白噪声检验

Table 1. White noise test of difference sequence

Lag	χ²值	P值
6	15.30	0.018
12	61.24	<0.001
18	74.38	<0.001

下载: 导出CSV

表 2 ARIMA((1, 12), 2, 0)模型参数估计

Table 2. ARIMA (1, 12), 2, 0) model parameter estimation

参数	估计	标准误差	t值	近似Pr>\|t\|	滞后
AR1.1	-0.05 775	0.09 074	-0.64	0.526	1
AR1.2	-0.66 976	0.09 869	-6.78	<0.001	12

下载: 导出CSV

表 3 去掉AR1, 1项后参数估计

Table 3. Parameter estimation after removing AR1, 1 term

参数	估计	标准误差	t值	近似Pr>\|t\|	滞后
AR1.1	-0.65 523	0.09 602	-6.82	<0.001	12

下载: 导出CSV

表 4 模型残差序列白噪声检验

Table 4. Residual sequence white noise test

Lag	χ²值	P值
6	5.77	0.329
12	6.84	0.811
18	9.11	0.936

下载: 导出CSV

表 5 我国2016年发病率预测值与实际值比较(1/10万)

Table 5. Comparison between forecast and actual incidence of incidence in China in 2016 (1/10⁵)

月份	实际值	预测值	置信区间	绝对误差	相对误差(%)
1	5.56	4.83	-3.11~12.78	-0.73	13.04
2	2.25	3.72	-7.52~14.96	1.47	65.09
3	5.47	10.45	-3.32~24.21	4.98	91.12
4	17.33	26.21	10.30~42.10	8.88	51.21
5	32.30	34.31	16.53~52.08	2.01	6.22
6	32.73	32.62	13.14~52.08	-0.11	0.34
7	22.46	22.02	0.98~43.05	-0.44	1.96
8	11.93	12.20	-10.28~34.68	0.27	2.25
9	9.84	15.06	-8.78~38.91	5.23	53.11
10	13.44	14.90	-10.23~40.04	1.46	10.86
11	15.57	11.52	-14.84~37.88	-4.05	26.02
12	10.36	8.53	-19.01~36.06	-1.83	17.70

下载: 导出CSV

参考文献(22)

[1]	张静. 2008-2017年中国手足口病流行趋势和病原变化动态数列分析[J]. 中华流行病学杂志, 2019, 40(2): 147-154. DOI: 10.3760/cma.j.issn.0254-6450.2019.02.005. Zhang J. Trend of epidemics and variation of pathogens of hand, foot and mouth disease in china: a dynamic series analysis, 2008-2017[J]. Chin J Epidemiol, 2019, 40(2): 147-154. DOI: 10.3760/cma.j.issn.0254-6450.2019.02.005.
[2]	中华人民共和国卫生部. 手足口病预防控制指南(2008版)[J]. 首都公共卫生, 2008, 2(3): 210-213. DOI: 10.3969/j.issn.1673-7830.2008.04.002. Minister of Health of the People's Republic of China. Hand-foot-mouth disease control and prevention guide[J]. Capital Journal of Public Health, 2008, 2(3): 210-213. DOI: 10.3969/j.issn.1673-7830.2008.04.002.
[3]	Koh WM, Bogich T, Siegel K, et al. The epidemiology of hand, foot and mouth disease in Asia: a systematic review and analysis[J]. Pediatric Infectious Journal, 2016, 35(10): 1-5. DOI: 10.1097/inf.0000000000001242.
[4]	王琳琳. 2014年济南市市中区手足口病流行病学和病原学特征分析[J]. 疾病监测与控制, 2016, 10(4): 264-266. https://www.cnki.com.cn/Article/CJFDTOTAL-JBJK201604003.htm Wang LL. Analysis of epidemiology and etiology characteristics on hand, foot and mouth disease in Shizhong District, City of Jinan[J]. J Disease Monitor & Control, 2016, 10(4): 264-266. https://www.cnki.com.cn/Article/CJFDTOTAL-JBJK201604003.htm
[5]	王燕. 应用时间序列分析[M]. 第二版. 北京: 中国人民大学出版社, 2008: 1-239. Wang Y. Application of time series analysis[M]. 2nd ed. Beijing: China Renmin University Press, 2008: 1-239.
[6]	孙振球, 徐勇勇. 医学统计学[M]. 第四版. 北京: 人民卫生出版社, 2014: 384-401. Sun ZQ, Xu YY. Medical Statistics[M]. 4th. Beijing: People's Medical Publishing House, 2014: 384-401.
[7]	胡跃华, 肖革新, 郭莹, 等. 2008-2011年中国大陆手足口病流行特征分析[J]. 中华疾病控制杂志, 2014, 18(8): 693-697, 747. http://zhjbkz.ahmu.edu.cn/article/id/JBKZ201408001 Hu YH, Xiao GX, Guo Y, et al. The epidemic features of hand, foot and mouth disease during 2008-2011 in china[J]. Chin J Dis Control Prev, 2014, 18(8): 693-697, 747. http://zhjbkz.ahmu.edu.cn/article/id/JBKZ201408001
[8]	Deng T, Huang Y, Yu S, et al. Special-temporal clusters and risk factors of hand, foot and mouth disease at the districtlevel in Guangdong Province, China[J]. Plos One, 2013, 8(2): e56943. DOI: 10.1371/journal.pone.0056943.
[9]	吴恒超. 莱芜市手足口病的流行特征及其与气象因素的相关性研究[D]. 泰安: 泰山医学院, 2014. DOI: 10.7666/d.Y2748056. Wu HC. Research of the hand, foot and mouth disease in Laiwu about epidemic characteristics and correlation with meteorological factors[D]. Taian: Taishan Medical University, 2014. DOI: 10.7666/d.Y2748056.
[10]	吕世峰, 王慧君, 刘慧玲, 等. 山东日照市2009-2012年手足口病与气象因素的时间序列分析[J]. 华南预防医学, 2014, 8(4): 312-317. DOI: 10.13217/j.scjpm.2014.0312. Lv SF, Wang HJ, Liu HL, et al. Time series analysis of wether and hand, foot and mouth disease in Rizhao, Shandong, 2009-2012[J]. South China J Prev Med, 2014, 8(4): 312-317. DOI: 10.13217/j.scjpm.2014.0312.
[11]	梅梅, 姜允迪, 王遵娅, 等. 2016年中国气候主要特征及主要天气气候事件[J]. 气象, 2017, 43(4): 468-476. DOI: 10.7519/j.issn.1000-0526.2017.04.009. Mei M, Jiang YD, Wang ZY, et al. Climatic characteristics and major meteorological events over China in 2016[J]. Meteorological Monthly, 2017, 43(4): 468-476. DOI: 10.7519/j.issn.1000-0526.2017.04.009.
[12]	Chan JH, Law CK, Hamblion E, et al. Best practices to prevent transmission and control outbreaks of hand, foot and mouth disease in childcare facilities: A systematic review[J]. Hong Kong Med J, 2017, 23(2): 177-190. DOI: 10.12809/hkmj166098.
[13]	中国疾病预防控制中心. 肠道病毒71型灭活疫苗使用技术指南[J]. 中国病毒病杂志, 2016, 6(4): 241-247. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGJM201604020.htm Chinese Center for Disease Control and Prevention. Guidelines for use of inactivated enterovirus type 71 vaccine[J]. Chinese Journal of Viral Diseases, 2016, 6(4): 241-247. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGJM201604020.htm
[14]	白尧, 刘坤, 谷旭, 等. 陕西省2009-2013年手足口病疫情时空流行特征分析[J]. 中华流行病学杂志, 2018, 39(9): 1152-1158. DOI: 10.3760/cma.j.issn.0254-6450.2018.09.003. Bai Y, Liu K, Gu X, et al. Analysis on the spatial-temporal characteristics of hand-foot-mouth disease in Shanxi Province, 2009-2013[J]. Chin J Epidemiol, 2018, 39(9): 1152-1158. DOI: 10.3760/cma.j.issn.0254-6450.2018.09.003.
[15]	吴红霞, 王杰, 许晓楠, 等. 2008-2012年张家口地区手足口病流行病学调查[J]. 中国医药导报, 2017, 14(30): 59-62. https://www.cnki.com.cn/Article/CJFDTOTAL-YYCY201730015.htm Wu HX, Wang J, Xu XN, et al. Epidemiological survey of hand-foot-mouth disease in Zhangjiakou area from 2008-2012[J]. China Medical Herald, 2017, 14(30): 59-62. https://www.cnki.com.cn/Article/CJFDTOTAL-YYCY201730015.htm
[16]	卢小敏, 储海燕, 吴冬梅, 等. 盐城市空气污染物对手足口病发病风险的影响研究[J]. 中国妇幼健康研究, 2018, 29(3): 276-280. DOI: 10.3969/j.issn.1673-5293,2018.03.008. Lu XM, Chu HY, Wu DM, et al. Influence of air pollutants on occurrence risk of hand-foot-mouth disease in Yancheng city[J]. Chinese Journal of women and child health research, 2018, 29(3): 276-280. DOI: 10.3969/j.issn.1673-5293,2018.03.008.
[17]	王国华, 王春梅, 郭敏健, 等. 应用圆形分布法分析桐乡市手足口病的流行特征[J]. 预防医学, 2017, 29(10): 1027-1029. DOI: 10.19485/j.cnki.issn1007-0931.2017.10.014. Wang GH, Wang CM, Guo MJ, et al. Analysis of epidemic characteristics of hand, foot and mouth disease in Tongxiang City based on circular distribution method[J]. Prev Med, 2017, 29(10): 1027-1029. DOI: 10.19485/j.cnki.issn1007-0931.2017.10.014.
[18]	徐雪琴, 孙宁, 徐玉芳, 等. 基于BP神经网络的河南省甲乙类法定报告传染病预测研究[J]. 中华疾病控制杂志, 2014, 18(6): 561-563. http://zhjbkz.ahmu.edu.cn/article/id/JBKZ201406023 Xu XQ, Sun N, Xu YF, et al. Prediction of the notifible infectious disease in Henan Province based on BP neural network[J]. Chin J Dis Control Prev, 2014, 18(6): 561-563. http://zhjbkz.ahmu.edu.cn/article/id/JBKZ201406023
[19]	杨长庆, 姬冬静, 李峰, 等. ARIMA模型在盐城市肺结核发病率预测中的应用[J]. 南通大学学报(医学版), 2017, 37(40: 331-334. https://www.cnki.com.cn/Article/CJFDTOTAL-NTYX201704011.htm Yang CQ, Ji DJ, Li F, et al. Application of ARIMA model in predicting incidence of pulmonary tuberculosis incidence of Yancheng City[J]. Journal of Nantong University(Medical Sciences), 2017, 37(4): 331-334. https://www.cnki.com.cn/Article/CJFDTOTAL-NTYX201704011.htm
[20]	Peng Y, Yu B, Wang P, et al. Application of seasonal auto-regressive integrated moving average model in forecasting the incidence of hand-foot-mouth disease in Wuhan, China[J]. Journal of Huazhong University of Science and Technology (Medical Sciences), 2017, 37(6): 842-848. DOI: 10.1007/s11596-017-1815-8.
[21]	马晓梅, 刘颖, 杨梦利, 等. 手足口病月发病率ARIMA乘积季节模型预测探讨[J]. 现代预防医学, 2017, 44(9): 1541-1544, 1560. https://www.cnki.com.cn/Article/CJFDTOTAL-XDYF201709002.htm Ma XM, Liu Y, Yang ML, et al. Multiple seasonal ARIMA model in prediction of the monthly incidence of the hand-foot-mouth disease[J]. Modern Preventive Medicine, 2017, 44(9): 1541-1544, 1560. https://www.cnki.com.cn/Article/CJFDTOTAL-XDYF201709002.htm
[22]	郑代坤, 谭毅, 李佳, 等. 基于自回归求和移动平均模型预测我国手足口病月报告发病数[J]. 疾病监测, 2018, 33(1): 54-58. DOI: 10.3784/j.issn.1003-9961.2018.01.013. Zheng DK, Tan Y, Li J, et al. Prediction of monthly reported cases of hand, foot and mouth disease by ARIMA model in China[J]. Disease Surveillance, 2018, 33(1): 54-58. DOI: 10.3784/j.issn.1003-9961.2018.01.013.