基于GM(1, 1)模型对我国妇幼保健指标的预测

祝丽玲; 孟繁君; 杨迪

doi:10.16462/j.cnki.zhjbkz.2019.08.018

基于GM(1, 1)模型对我国妇幼保健指标的预测

doi: 10.16462/j.cnki.zhjbkz.2019.08.018

154007 佳木斯, 佳木斯大学公共卫生学院预防医学教研室

详细信息

中图分类号: R181
计量
- 文章访问数: 450
- HTML全文浏览量: 160
- PDF下载量: 62
- 被引次数: 0
出版历程
- 收稿日期: 2019-03-29
- 修回日期: 2019-06-12
- 刊出日期: 2019-08-10

Prediction of maternal and child health care indicators in China based on GM(1, 1) model

Department of Preventive Medicine, Academy of Public Health of Jiamusi University, Jiamusi 154007, China

摘要

摘要: 目的研究GM（1，1）模型在我国妇幼保健指标中的预测效果，并对未来妇幼保健指标进行短期预测，为我国妇幼保健服务水平的逐步完善提供科学依据。方法收集我国2008-2017年的孕产妇死亡率（maternal mortality rate，MMR）、新生儿死亡率（neonatal mortality rate，NMR）、婴儿死亡率（infant mortality rate，IMR）和5岁以下儿童死亡率（under-five mortality rate，U5MR），建立模型，应用MATLAB 2018b软件进行预测分析。结果我国MMR、NMR、IMR和U5MR的预测模型分别为：${\rm{\hat x}}\left( {{\rm{k}} + 1} \right) = - 476.08{{\rm{e}}^{ - 0.09{\rm{k}}}} + 510.28({C_1} = 0.165,{P_1} = 1.000)$，${\rm{\hat x}}\left( {{\rm{k}} + 1} \right) = - 108.43{{\rm{e}}^{{\rm{ - 0}}{\rm{.09k}}}} + 118.63({C_2} = 0.043,{P_2} = 1.000)$，${\rm{\hat x}}\left( {{\rm{k}} + 1} \right) = - 160.60{{\rm{e}}^{ - 0.09{\rm{k}}}} + 175.50({C_3} = 0.085,{P_3} = 1.000)$，${\rm{\hat x}}\left( {{\rm{k}} + 1} \right) = - 224.37{{\rm{e}}^{ - 0.08{\rm{k}}}} + 242.87({C_4} = 0.124,{P_4} = {\rm{ }}1.000)$，平均相对误差分别为：3.46%、0.67%、1.75%和2.36%。结论 GM（1，1）模型适用于对我国妇幼保健指标的预测，拟合精度均较高；预测未来三年各指标将继续逐年下降，相关部门应有针对性的加强管理工作。
- GM (1, 1)模型 /
- 妇幼保健 /
- 预测分析
Abstract: Objective To study the predictive effect of model[GM(1, 1)] in China's maternal and child health indicators, and to predict the future maternal and child health indicators in a short-term, and provide a scientific basis for the gradual improvement of maternal and child health care services in China. Methods The maternal mortality rate (MMR), neonatal mortality rate (NMR), infant mortality rate (IMR) and under-five mortality rate (U5MR) were collected from 2008 to 2017 in China. Models were established and MATLAB 2018b software was used for predictive analysis. Results The prediction models of maternal mortality rate, neonatal mortality rate, infant mortality rate and under-five mortality rate were as follows: ${\rm{\hat x}}\left( {{\rm{k}} + 1} \right) = - 476.08{{\rm{e}}^{ - 0.09{\rm{k}}}} + 510.28({C_1} = 0.165,{P_1} = 1.000)$, ${\rm{\hat x}}\left( {{\rm{k}} + 1} \right) = - 108.43{{\rm{e}}^{{\rm{ - 0}}{\rm{.09k}}}} + 118.63({C_2} = 0.043,{P_2} = 1.000)$, ${\rm{\hat x}}\left( {{\rm{k}} + 1} \right) = - 160.60{{\rm{e}}^{ - 0.09{\rm{k}}}} + 175.50({C_3} = 0.085,{P_3} = 1.000)$, ${\rm{\hat x}}\left( {{\rm{k}} + 1} \right) = - 224.37{{\rm{e}}^{ - 0.08{\rm{k}}}} + 242.87({C_4} = 0.124,{P_4} = {\rm{ }}1.000)$, the average relative errors were as follows: 3.46%, 0.67%, 1.75% and 2.36%。 Conclusions The GM (1, 1) is suitable for the prediction of maternal and child health indicators in China, and the fitting accuracy is high. It is predicted that the indicators will continue to decline year by year in the next three years, and relevant departments should strengthen the management work in a targeted manner.
- GM(1, 1) model /
- Maternal and child health care /
- Predictive analysis

HTML全文

表 1 GM(1, 1)模型精度判断标准^[7]

Table 1. GM (1, 1) model accuracy judgment criteria^[7]

P值	C值	拟合精度等级
＞0.95	＜0.35	好
＞0.80	＜0.50	合格
＞0.70	＜0.65	勉强合格
≤0.70	≥0.65	不合格

下载: 导出CSV

表 2 我国各年份妇幼保健指标的GM(1, 1)模型建立及检验^[8]

Table 2. Establishment and testing of GM(1, 1) model for maternal and child health indicators in various years in China^[8]

时间	MMR(1/10万)			NMR(‰)			IMR(‰)			U5MR(‰)
时间	实际值	预测值	相对误差(%)	实际值	预测值	相对误差(%)	实际值	预测值	相对误差(%)	实际值	预测值	相对误差(%)
2008	34.20	-	0.00	10.20	-	0.00	14.90	-	0.00	18.50	-	0.00
2009	31.90	31.08	2.57	9.00	9.06	0.67	13.80	14.03	1.67	17.20	17.46	1.51
2010	30.00	29.05	3.17	8.30	8.30	0.00	13.10	12.80	2.29	16.40	16.10	1.83
2011	26.10	27.15	4.02	7.80	7.61	2.44	12.10	11.68	3.47	15.60	14.85	4.81
2012	24.50	25.38	3.59	6.90	6.97	1.01	10.30	10.66	3.50	13.20	13.69	3.71
2013	23.20	23.73	2.28	6.30	6.39	0.00	9.50	9.73	2.42	12.00	12.63	5.25
2014	21.70	22.18	2.21	5.90	5.86	0.68	8.90	8.88	0.22	11.70	11.64	0.51
2015	20.10	20.73	3.13	5.40	5.37	0.56	8.10	8.11	0.12	10.70	10.74	0.37
2016	19.90	19.38	2.61	4.90	4.92	0.41	7.50	7.40	1.33	10.20	9.90	2.94
2017	19.60	18.12	7.55	4.50	4.51	0.22	6.80	6.75	0.74	9.10	9.13	0.33
C值		0.165			0.043			0.085			0.124
P值		1.000			1.000			1.000			1.000
平均误差(%)		3.46			0.67			1.75			2.36
注：符合率=预测值/实际值*100%，相对误差=1-符合率。

下载: 导出CSV

表 3 2018-2020年妇幼保健指标的预测情况

Table 3. Forecast of maternal and child health indicators from 2018 to 2020

时间	MMR预测值(1/10万)	NMR预测值(‰)	IMR预测值(‰)	U5MR预测值(‰)
2018	16.93	4.13	6.16	8.42
2019	15.83	3.79	5.62	7.77
2020	14.80	3.47	5.13	7.16

下载: 导出CSV

参考文献(17)

[1]	王四梅. 妇幼保健工作与母婴安全[J]. 世界最新医学信息文摘, 2018, 18(78): 206. DOI: 10.19613/j.cnki.1671-3141.2018.78.165. Wang SM. Maternal and child health work and maternal and child safety[J]. World Latest Medicine Information, 2018, 18(78): 206. DOI: 10.19613/j.cnki.1671-3141.2018.78.165.
[2]	卢月, 毕蕾, 张寒, 等. 我国妇幼保健资源配置现状及公平性分析[J]. 中国妇幼保健, 2018, 33(5): 965-969. DOI: 10.7620/zgfybj.j.issn.1001-4411.2018.05.02. Lu Y, Bi L, Zhang H, et al. Analysis on the status quo and fairness of maternal and child health resources allocation in China[J]. Maternal and Child Health Care of China, 2018, 33(5): 965-969. DOI: 10.7620/zgfybj.j.issn.1001-4411.2018.05.02.
[3]	张涛, 李书婷, 孙立奇, 等. 我国妇幼保健机构公共卫生服务效率测评及影响因素分析[J]. 中国卫生统计, 2018, 35(1): 89-91. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGWT201801025.htm Zhang T, Li ST, Sun LQ, et al. Evaluation of public health service efficiency and its influencing factors in maternal and child health care institutions in China[J]. Chin J Health Statistics, 2018, 35(1): 89-91. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGWT201801025.htm
[4]	刘元元. 2010年我国孕产妇、婴儿及5岁以下儿童死亡率的统计预测研究[D]. 成都: 四川大学, 2005. Liu YY. Statistical prediction of maternal, infant and under-five child mortality in China in 2010[D]. Chengdu: Sichuan University, 2005.
[5]	朱春梅. 灰色GM(1, 1)模型在孕产妇和新生儿死亡率预测研究中的应用[J]. 中国妇幼保健, 2016, 31(3): 455-457. DOI: 10.7620/zgfybj.j.issn.1001-4411.2016.03.03. Zhu CM. Application of grey GM(1, 1) model in predicting maternal and neonatal mortality[J]. Maternal and Child Health Care of China, 2016, 31(3): 455-457. DOI: 10.7620/zgfybj.j.issn.1001-4411.2016.03.03.
[6]	宋志慧, 王永斌, 冯福民. 残差修正GM(1, 1)模型在我国妇幼卫生健康指标预测中的应用[J]. 中国妇幼保健, 2015, 30(34): 5951-5954. DOI: 10.7620/zgfybj.j.issn.1001-4411.2015.34.05. Song ZH, Wang YB, Feng FM. Application of residual correction GM(1, 1) model in predicting maternal and child health indicators in China[J]. Maternal and Child Health Care of China, 2015, 30(34): 5951-5954. DOI: 10.7620/zgfybj.j.issn.1001-4411.2015.34.05.
[7]	庞艳蕾. 孕产妇、婴儿及5岁以下儿童死亡率分析与预测[D]. 潍坊: 潍坊医学院, 2016. Pang YL. Analysis and prediction of maternal, infant and under-five mortality rates[D]. Weifang: Weifang Medical College, 2016.
[8]	国家卫生健康委员会. 2018年《中国卫生健康统计年鉴》[C]. 北京: 中国协和医科大学出版社, 2018: 217-221. National Health and Wellness Committee. China health statistics yearbook in 2018[C]. Beijing: China Union Medical University Press, 2018: 217-221.
[9]	Wang Y, Wei F, Sun C, et al. The research of improved grey GM(1, 1) model to predict the postprandial glucose in type 2 diabetes[J]. Biomed Res Int, 2016: 1-6. DOI: 10.1155/2016/6837052.
[10]	Shen X, Ou L, Chen X, et al. The application of the grey disaster model to forecast epidemic peaks of typhoid and paratyphoid fever in China[J]. PLoS One, 2013, 8(4): e60601. DOI: 10.1371/journal.pone.0060601.
[11]	鲁红坤. 灰色GM(1, 1)模型的改进及应用[D]. 沈阳: 东北大学, 2013. Lu HK. Improvement and application of grey GM(1, 1) model[D]. Shenyang: Northeastern University, 2013.
[12]	刘洁, 何钦成. GM(1, 1)模型在5岁以下儿童死亡率预测研究中的应用[J]. 实用预防医学, 2012, 19(1): 27-29. DOI: 10.3969/j.issn.1006-3110.2012.01.008. Liu J, He QC. Application of GM(1, 1) model in prediction of mortality in children under 5 years old[J]. Pract Prevent Med, 2012, 19(1): 27-29. DOI: 10.3969/j.issn.1006-3110.2012.01.008.
[13]	郭玉秀, 宋国强, 周荣耀. GM(1, 1)模型在江苏省5岁以下儿童死亡率预测中的应用[J]. 南京医科大学学报(社会科学版), 2018, 18(4): 296-298. DOI: 10.7655/NYDXBSS20180410. Guo YX, Song GQ, Zhou RY. Application of GM(1, 1) model in predicting mortality rate of children under 5 years old in Jiangsu Province[J]. Acta Universitatis Medicinalis Nanjing (Social Sciences), 2018, 18(4): 296-298. DOI: 10.7655/NYDXBSS20180410.
[14]	王雅文, 沈忠周, 马帅, 等. GM(1, 1)模型在孕产妇死亡率预测中的应用[J]. 中华疾病控制杂志, 2018, 22(7): 755-757. DOI: 10.16462/j.cnki.zhjbkz.2018.07.023. Wang YW, Shen ZZ, Ma S, et al. Application of GM(1, 1) model in maternal mortality prediction[J]. Chin J Dis Control Prev, 2018, 22(7): 755-757. DOI: 10.16462/j.cnki.zhjbkz.2018.07.023.
[15]	祝丽玲, 孟繁君, 杜宁. 秩和比法综合评价2015年我国孕产妇保健状况[J]. 中华疾病控制杂志, 2018, 22(8): 859-861. DOI: 10.16462/j.cnki.zhjbkz.2018.08.025. Zhu LL, Meng FJ, Du N. Comprehensive evaluation of the maternal health status in China in 2015 by rank sum ratio method[J]. Chin J Dis Control Prev, 2018, 22(8): 859-861. DOI: 10.16462/j.cnki.zhjbkz.2018.08.025.
[16]	潘佳奇, 吕军, 励晓红, 等. 昆山市孕产妇保健服务利用现状分析[J]. 中国妇幼保健, 2018, 33(08): 1689-1692. DOI: 10.7620/zgfybj.j.issn.1001-4411.2018.08.03. Pan JQ, Lv J, Li XH, et al. Analysis of the utilization status of maternal health care services in Kunshan[J]. Maternal and Child Health Care of China, 2018, 33(8): 1689-1692. DOI: 10.7620/zgfybj.j.issn.1001-4411.2018.08.03.
[17]	张树栋, 严倩, 王文兴, 等. 烟台市妇幼卫生信息化促进孕产妇保健系统管理实践与成效[J]. 中国妇幼保健, 2018, 33(22): 5052-5053. DOI:CNKI:SUN:ZFYB.0.2018-22-003. Zhang SD, Yan Q, Wang WX, et al. Practice and effect of information technology of maternal and child health in promoting management of maternal health care system in Yantai[J]. Maternal and Child Health Care of China, 2018, 33(22): 5052-5053. DOI: 10.7620/zgfybj.j.issn.1001-4411.2018.22.03.