2017―2022年天津市结核病流行特征及发病预测

李文; 高颖; 张国钦; 沈玉祯; 王媛

doi:10.16462/j.cnki.zhjbkz.2025.05.006

2017―2022年天津市结核病流行特征及发病预测

doi: 10.16462/j.cnki.zhjbkz.2025.05.006

李文¹,
高颖^{2, 3},
张国钦⁴,
沈玉祯³,
王媛^1, ,

1.
天津医科大学公共卫生学院流行病与卫生统计学系，天津 300070
2.
天津医科大学公共卫生学院，天津 300070
3.
天津市滨海新区疾病预防控制中心，天津 300450
4.
天津市结核病控制中心，天津 300041

详细信息

通讯作者:
王媛，E-mail: wangyuan@tmu.edu.cn

中图分类号: R211;R521;R181
计量
- 文章访问数: 10
- HTML全文浏览量: 6
- PDF下载量: 1
- 被引次数: 0
出版历程
- 收稿日期: 2024-11-08
- 修回日期: 2025-03-04
- 刊出日期: 2025-05-10

Epidemiological characteristics and incidence prediction of tuberculosis in Tianjin from 2017 to 2022

LI Wen¹,
GAO Ying^{2, 3},
ZHANG Guoqin⁴,
SHEN Yuzhen³,
WANG Yuan^{1
, ,}

1.
Department of Epidemiology and Health Statistics, School of Public Health, Tianjin Medical University, Tianjin 300070, China
2.
School of Public Health, Tianjin Medical University, Tianjin 300070, China
3.
Binhai New District Center for Disease Control and Prevention, Tianjin 300450, China
4.
Tianjin Tuberculosis Control Center, Tianjin 300041, China

More Information

Corresponding author: WANG Yuan, E-mail: wangyuan@tmu.edu.cn

摘要

摘要: 目的分析2017―2022年天津市结核病的发病趋势及流行特征，为结核病防控提供科学依据。方法收集天津市2017―2023年结核病的月度报告数据，利用R 4.2.2软件构建差分自回归移动平均(autoregressive integrated moving average, ARIMA)模型，使用2023年的数据进行验证并预测2025年结核病发病情况。结果 2017―2022年，天津市共报告新发结核病病例22 342例，年发病率从2017年的26.20/10万下降至2022年的24.29/10万(P>0.05)。结核病发病率存在季节性波动：3月和11月略有上升。环城四区(29.97/10万)和≥80岁年龄组(64.72/10万)年均发病率最高，男性年均发病率(30.33/10万)高于女性(18.53/10万)(χ² =1 257.246, P＜0.001)。ARIMA(1, 1, 1)(0, 1, 1)₁₂模型是拟合天津市结核病发病率的最优模型(AIC=67.586)，预测结果与2023年的真实值基本一致，平均相对误差为13.73%。预测天津市2025年结核病月发病率为0.75/10万~1.59/10万，年发病率为16.78/10万，呈下降趋势。结论 2017―2022年天津市结核病发病率呈下降趋势，发病人群以男性、老年人为主，应关注冬春、秋冬季节交接时和环城四区的结核病防控及宣教工作。基于ARIMA(1, 1, 1)(0, 1, 1)₁₂模型的预测效果较好，具有重要的防控工作意义。
- 差分自回归移动平均模型 /
- 结核病 /
- 预测
Abstract: Objective To analyze the incidence trends of tuberculosis (TB) in Tianjin from 2017 to 2022, and to provide a basis for TB prevention and control. Methods Monthly TB data in Tianjin from 2017 to 2023 were collected. An autoregressive integrated moving average (ARIMA) model was built with R 4.2.2 software, validated with 2023 data, and used to predict 2025 incidence. Results From 2017 to 2022, Tianjin reported 22 342 new TB cases, with the annual incidence declining from 26.20 per 100 000 in 2017 to 24.29 per 100 000 in 2022 (P>0.05). The incidence showed seasonal fluctuations, slightly increase in March and November. The highest rates were in the four urban districts (29.97 per 100 000) and individuals aged 80 and older (64.72 per 100 000); males had a higher incidence (30.33 per 100 000) than females (18.53 per 100 000) (χ² =1 257.246, P < 0.001). The optimal ARIMA(1, 1, 1)(0, 1, 1)₁₂ model had an AIC of 67.586. Predictions matched the 2023 actual trends, with an average relative error of 13.73%. The model predicts a monthly TB incidence of 0.75 to 1.59 per 100 000 in 2025, with an annual rate of 16.78 per 100 000, indicating a downward trend. Conclusions TB incidence in Tianjin decreased from 2017 to 2022, predominantly affecting males and the elderly. Enhanced prevention and education efforts are needed in urban areas, especially in time of transition from winter to spring and from autumn to winter. The ARIMA(1, 1, 1)(0, 1, 1)₁₂ model effectively predicts TB incidence, aiding control efforts.
- Autoregressive integrated moving average model /
- Tuberculosis /
- Prediction

HTML全文

图 1 2017―2022年天津市结核病发病率月份趋势图

Figure 1. Monthly incidence rate of tuberculosis in Tianjin form 2017 to 2022

下载: 全尺寸图片幻灯片

图 2 天津市2017―2022年结核病月发病率原始时间序列分解图

A：原始序列; B：趋势项; C：季节项; D：残差项。

Figure 2. Decomposition of original time series of monthly incidence rate of tuberculosis in Tianjin from 2017 to 2022

A: original sequence; B: trend; C: seasonal; D: residual.

下载: 全尺寸图片幻灯片

图 3 差分后的自相关和偏自相关函数图

Figure 3. Autocorrelation function and partial autocorrelation function after difference

下载: 全尺寸图片幻灯片

图 4 ARIMA(1, 1, 1)(0, 1, 1)₁₂模型预测2025年天津市结核病发病率

Figure 4. ARIMA(1, 1, 1)(0, 1, 1)₁₂ model predicted tuberculosis incidence rate in Tianjin in 2025

下载: 全尺寸图片幻灯片

表 1 2017―2022年天津市结核病发病情况

Table 1. Tuberculosis incidence in Tianjin from 2017 to 2022

	2017年 year^①	2018年 year^①	2019年 year^①	2020年 year^①	2021年 year^①	2022年 year^①	合计 Total^①
发病情况Morbidity	4 093(26.20)	3 874(24.88)	4 008(25.69)	3 129(20.03)	3 903(28.15)	3 335(24.29)	22 342(24.82)
地区Region
市内六区The city six districts	1 583(29.93)	1 256(25.77)	1 297(26.39)	963(19.51)	1 125(27.73)	861(21.76)	7 094(25.28)
环城四区Four districts around the city	1 048(34.34)	1 097(32.64)	1 105(32.65)	890(26.23)	1 120(28.79)	1 029(26.35)	6 289(29.97)
远郊五区Exurbs five districts	971(21.90)	979(22.69)	1 021(23.66)	744(17.30)	964(25.03)	801(20.98)	5 480(21.89)
滨海新区Binhai New District	463(16.27)	474(15.88)	531(17.78)	470(15.73)	570(27.57)	559(27.26)	3 067(19.26)
性别Gender
男Male	2 687(31.60)	2 597(30.80)	2 665(31.54)	2 014(23.81)	2 500(34.99)	2 103(29.87)	14 566(30.33)
女Female	1 406(19.75)	1 277(17.89)	1 343(18.78)	1 115(15.58)	1 403(20.87)	1 232(18.42)	7 776(18.53)
发病率性别比(男: 女) Incidence gender ratio (male: female)	1.91:1.00	2.03:1.00	1.98:1.00	1.81:1.00	1.78:1.00	1.71:1.00	1.87:1.00
年龄组/岁Age group/years
0~＜10	7(0.61)	1(0.09)	25(2.24)	24(2.16)	20(1.58)	24(1.99)	101(1.45)
10~＜20	263(21.74)	235(19.32)	224(24.13)	168(17.84)	176(17.99)	129(11.13)	1 195(18.69)
20~＜30	861(25.05)	888(25.79)	772(27.35)	575(24.29)	646(36.60)	463(26.89)	4 205(27.66)
30~＜40	518(21.45)	557(22.95)	596(20.33)	540(17.33)	678(26.84)	552(21.71)	3 441(21.77)
40~＜50	422(16.61)	453(17.87)	411(17.45)	392(17.24)	447(24.75)	376(19.55)	2 501(18.91)
50~＜60	702(28.57)	597(24.62)	660(23.69)	493(17.05)	627(24.89)	576(26.72)	3 655(24.26)
60~＜70	676(50.88)	578(44.00)	671(38.43)	487(26.41)	702(38.77)	616(34.72)	3 730(38.87)
70~＜80	398(52.78)	349(47.29)	392(63.07)	291(39.64)	399(49.98)	394(44.79)	2 223(49.59)
≥80	246(75.81)	216(68.54)	257(88.24)	159(46.80)	208(52.82)	205(56.12)	1 291(64.72)
注：市内六区, 和平区、河东区、河西区、南开区、河北区、红桥区; 环城四区, 东丽区、西青区、津南区、北辰区; 远郊五区, 武清区、宝坻区、宁河区、静海区、蓟州区。 ①以病例数(发病率/10万^-1)表示。 Note: The city six districts: Heping District, Hedong District, Hexi District, Nankai District, Hebei District, Hongqiao District; Four districts around the city: Dongli District, Xiqing District, Jinnan District, Beichen District; Exurbs five districts: Wuqing District, Baodi District, Ninghe District, Jinghai District, Jizhou District. ① Number of cases (incidence/100 000^-1).

下载: 导出CSV

表 2 候选模型的评价

Table 2. Evaluation of potential models

序号 Number	ARIMA模型 ARIMA model	AIC值 value	BIC值 value	Ljung-Box检验 P值value
1	ARIMA(1, 0, 0)(2, 0, 0)₁₂	69.181	80.565	0.780
2	ARIMA(1, 1, 1)(0, 1, 0)₁₂	85.415	91.648	0.941
3	ARIMA(1, 1, 1)(0, 1, 1)₁₂	67.586	75.897	0.938
4	ARIMA(1, 1, 1)(0, 1, 2)₁₂	69.150	79.537	0.933
5	ARIMA(1, 1, 1)(1, 1, 0)₁₂	72.919	81.229	0.988
6	ARIMA(1, 1, 1)(1, 1, 1)₁₂	69.169	79.557	0.934
7	ARIMA(1, 1, 1)(1, 1, 2)₁₂	70.992	83.458	0.927
8	ARIMA(1, 1, 1)(2, 1, 0)₁₂	75.905	84.215	0.226
9	ARIMA(1, 1, 1)(2, 1, 1)₁₂	71.100	83.565	0.958
10	ARIMA(1, 1, 1)(2, 1, 2)₁₂	74.645	87.110	0.304
11	ARIMA(1, 1, 0)(0, 1, 0)₁₂	87.463	91.618	0.325
12	ARIMA(1, 1, 0)(0, 1, 1)₁₂	70.895	77.128	0.249
13	ARIMA(1, 1, 0)(1, 1, 0)₁₂	77.412	83.644	0.251
14	ARIMA(1, 1, 0)(1, 1, 1)₁₂	72.100	80.410	0.263
15	ARIMA(1, 1, 0)(2, 1, 0)₁₂	75.905	84.215	0.226
16	ARIMA(1, 1, 0)(2, 1, 1)₁₂	73.554	83.942	0.274
17	ARIMA(0, 1, 1)(0, 1, 0)₁₂	83.415	87.570	0.939
注：ARIMA，差分自回归移动平均; AIC，赤池信息准则; BIC，贝叶斯信息准则。 Note: ARIMA, autoregressive integrated moving average; AIC, Akaike information criterion; BIC, Bayesian information criterion.

下载: 导出CSV

表 3 ARIMA(1, 1, 1)(0, 1, 1)₁₂模型预测2023年天津市结核病发病率

Table 3. ARIMA(1, 1, 1)(0, 1, 1)₁₂ model predicted the incidence rate of tuberculosis in Tianjin in 2023

月份 Month	真实值 Actual value	预测值 Predicted value	绝对误差 Absolute error	相对误差 Relative error/%
1	1.61	1.51	0.11	6.53
2	2.25	0.99	1.26	56.08
3	2.36	1.71	0.64	27.35
4	1.97	1.81	0.16	8.10
5	1.81	1.64	0.17	9.52
6	1.99	1.83	0.16	8.07
7	1.59	1.79	0.20	12.67
8	1.80	1.74	0.06	3.24
9	1.63	1.82	0.19	11.61
10	1.58	1.41	0.16	10.32
11	1.65	1.70	0.04	2.69
12	1.54	1.67	0.13	8.54

下载: 导出CSV

参考文献(25)

[1]	World Health Organization. WHO consolidated guidelines on tuberculosis. Module 1: prevention-tuberculosis preventive treatment, second edition[R]. Geneva: World Health Organization, 2024: 3.
[2]	World Health Organization. Global tuberculosis report 2023[R]. Geneva: World Health Organization, 2023: 1.
[3]	Ledesma J R, Ma J, Zhang M, et al. Global, regional, and national age-specific progress towards the 2020 milestones of the WHO end TB strategy: a systematic analysis for the global burden of disease study 2021[J]. Lancet Infec Dis, 2024, 24 (7): 698-725. DOI: 10.1016/S1473-3099(24)00007-0.
[4]	王歆尧, 姜美丽, 庞元捷, 等. 中国结核病疾病负担现状[J]. 中华流行病学杂志, 2024, 45(6): 857-864. DOI: 10.3760/cma.j.cn112338-20240311-00111. Wang XY, Jiang ML, Pang YJ, et al. Current status of tuberculosis burden in China[J]. Chin J Epidemiol, 2024, 45(6): 857-864. DOI: 10.3760/cma.j.cn112338-20240311-00111.
[5]	张玉华. 2011-2020年天津市中心城区肺结核患者流行病学特征分析[J]. 结核与肺部疾病杂志, 2023, 4(4): 323-327. DOI: 10.19983/j.issn.2096-8493.20230069. Zhang YH. Epidemiological characteristics of pulmonary tuberculosis patients in central urban area of Tianjin City from 2011 to 2020[J]. J Tuberc Lung Dis, 2023, 4(4): 323-327. DOI: 10.19983/j.issn.2096-8493.20230069.
[6]	张文倩, 黄飞, 张国钦, 等. 2011-2022年天津市肺结核时空分布特征分析[J]. 中国防痨杂志, 2023, 45(12): 1170-1176. DOI: 10.19982/j.issn.1000-6621.20230287. Zhang WQ, Huang F, Zhang GQ, et al. Temporal-spatial distribution of pulmonary tuberculosis in Tianjin during 2011-2022[J]. Chin J Antituberc, 2023, 45(12): 1170-1176. DOI: 10.19982/j.issn.1000-6621.20230287.
[7]	Liu D, Huang F, Li Y, et al. Transmission characteristics in Tuberculosis by WGS: nationwide cross-sectional surveillance in China[J]. Emerg Microbes Infect, 2024, 13(1): 2348505. DOI: 10.1080/22221751.2024.2348505.
[8]	Lin Y, Chen M, Chen G, et al. Application of an autoregressive integrated moving average model for predicting injury mortality in Xiamen, China[J]. BMJ open, 2015, 5 (12): e008491. DOI: 10.1136/bmjopen-2015-008491.
[9]	Li Z Q, Pan H Q, Liu Q, et al. Comparing the performance of time series models with or without meteorological factors in predicting incident pulmonary tuberculosis in eastern China[J]. Infect Dis Poverty, 2020, 9: 1-11. DOI: 10.1186/s40249-020-00771-7.
[10]	Chen SS, Wang XQ, Zhao JW, et al. Application of the ARIMA model in forecasting the incidence of tuberculosis in Anhui during COVID-19 pandemic from 2021 to 2022[J]. Infect Drug Resist, 2022, 15: 3503-3512. DOI: 10.2147/IDR.S367528.
[11]	Aryee G, Kwarteng E, Essuman R, et al. Estimating the incidence of tuberculosis cases reported at a tertiary hospital in Ghana: a time series model approach[J]. BMC Public Health, 2018, 18: 1-8. DOI: 10.1186/s12889-018-6221-z.
[12]	Yang J, Zhou J, Luo TY, et al. Predicting pulmonary tuberculosis incidence in China using Baidu search index: an ARIMAX model approach[J]. Environ Health Prev Med, 2023, 28: 68. DOI: 10.1265/ehpm.23-00141.
[13]	Anwar M Y, Lewnard J A, Parikh S, et al. Time series analysis of malaria in Afghanistan: using ARIMA models to predict future trends in incidence[J]. Malaria J, 2016, 15: 1-10. DOI: 10.1186/s12936-016-1602-1.
[14]	Zhai MM, Li WH, Tie P, et al. Research on the predictive effect of a combined model of ARIMA and neural networks on human brucellosis in Shanxi Province, China: a time series predictive analysis[J]. BMC Infect Dis, 2021, 21(1): 280. DOI: 10.1186/s12879-021-05973-4.
[15]	Liu L, Luan RS, Yin F, et al. Predicting the incidence of hand, foot and mouth disease in Sichuan province, China using the ARIMA model[J]. Epidemiol Infec, 2016, 144(1): 144-151. DOI: 10.1017/S0950268815001144.
[16]	Wang M, Pan JH, Li XH, et al. ARIMA and ARIMA-ERNN models for prediction of pertussis incidence in mainland China from 2004 to 2021[J]. BMC Public Health, 2022, 22(1): 1447. DOI: 10.1186/s12889-022-13872-9.
[17]	Zhang Y, Wang XQ, Liu NQ, et al. Intervention effect of new tuberculosis control model on tuberculosis incidence in Xinjiang[J]. Infect Drug Resist, 2023, 16: 7485-7496. DOI: 10.2147/IDR.S441899.
[18]	刘伟, 刘远, 胡文穗, 等. 差分自回归移动平均乘积季节模型预测广州市肺结核发病趋势[J]. 中华疾病控制杂志, 2021, 25(2): 240-243, 248. DOI: 10.16462/j.cnki.zhjbkz.2021.02.023. Liu W, Liu Y, Hu WS, et al. Application of multiple seasonal ARIMA model for predicting the incidence trend of tuberculosis in Guangzhou City[J]. Chin J Dis Control Prev, 2021, 25(2): 240-243, 248. DOI: 10.16462/j.cnki.zhjbkz.2021.02.023.
[19]	Liu JW, Zeng W, Zhuo C, et al. Impact of the COVID-19 pandemic on the incidence of notifiable infectious diseases in China based on SARIMA models between 2013 and 2021[J]. J Epidemiol Glob Health, 2024, 14(3): 1191-1201. DOI: 10.1007/s44197-024-00273-x.
[20]	Yu SS, Zhan MR, Li KG, et al. Analysis of tuberculosis epidemiological distribution characteristics in Fujian Province, China, 2005-2021: spatial-temporal analysis study[J]. JMIR Public Health Surveill, 2024, 10: e49123. DOI: 10.2196/49123.
[21]	Xia CP, Wang JY, Wang ZX, et al. Correlation between notifiable infectious diseases and transportation passenger traffic from 2013 to 2019 in Mainland China[J]. BMC Public Health, 2024, 24(1): 3023. DOI: 10.1186/s12889-024-20479-9.
[22]	Hertz D, Schneider B. Sex differences in tuberculosis[J]. Semin Immunopathol, 2019, 41(2): 225-237. DOI: 10.1007/s00281-018-0725-6.
[23]	Allué-Guardia A, Torrelles JB, Sigal A. Tuberculosis and COVID-19 in the elderly: factors driving a higher burden of disease[J]. Front Immunol, 2023, 14: 1250198. DOI: 10.3389/fimmu.2023.1250198.
[24]	国务院联防联控机制. 关于进一步优化落实新冠肺炎疫情防控措施的通知[EB/OL]. (2022-12-07)[2025-01-04]. https://www.gov.cn/xinwen/2022-12/07/content_5730475.htm.
[25]	Zhao D, Zhang H, Cao Q, et al. The research of ARIMA, GM (1, 1), and LSTM models for prediction of TB cases in China[J]. Plos One, 2022, 17(2): e0262734. DOI: 10.1371/journal.pone.0262734.