Co-benefits effect of non-pharmacological intervention against COVID-19 on tuberculosis in Ningxia Hui Autonomous Region: a SARIMA analysis
-
摘要:
目的 分析2020年新型冠状病毒肺炎疫情期间的非药物干预策略(NPIs)对宁夏回族自治区(简称宁夏)肺结核发病的影响。 方法 构建季节性差分自回归移动平均(SARIMA)模型对2020年宁夏肺结核发病进行预测,并根据2020年突发公共卫生事件不同的应急响应级别将肺结核实际发病数分别与2019年同期数据以及模型得出的预期发病数计算相对变化率进行对比分析。 结果 2004 — 2020年宁夏肺结核发病存在明显的季节性趋势;ARIMA(1,1,2)(1,1,2)12模型(RMSE = 1280.500,MAPE = 13.15 %)为筛选出的最优模型。2020年突发公共卫生事件一至三级响应时发病的预期降幅分别为63.58 %、29.61 %和8.59 %。 结论 新冠肺炎的非药物干预措施对宁夏肺结核病具有降低发病的伴随效应,且该效应与应急响应级别呈正比,突发公共卫生事件应急响应级别越高,该抑制作用越明显。 Abstract:Objective To analyze the influence of non-pharmacological interventions (NPIs) against coronavirus disease 2019 (COVID-19) on tuberculosis (TB) incidence in Ningxia Hui Autonomous Region (Ningxia) in 2020. Methods Using TB incidents reported in Ningxia during 2004 – 2018 as a training set and those during 2019 – 2020 as a prediction set, a seasonal autoregressive moving average (SARIMA) model was established to predict TB incidents during the three COVID-19 epidemic periods with different response level in 2020 in Ningxia. The relative rate of change in the number of TB incidence for the three response periods were calculated by comparing the TB incidence number in the COVID-19 response periods with the number of same periods in 2019 or the number predicted based on the established SARIMA model. Results The established ARIMA(1,1,2) (1,1,2)12 model was fitted to the data well, with a root mean square error of 1280.50 and a mean absolute percentage error of 13.15%. The NPIs against COVID-19 showed an inhibitive co-benefits effect on the incidence of TB in Ningxia, resulting the expected incidence reductions of 63.58%, 29.61% and 8.59% for the response periods of level I, level II and level III, respectively. Conclusion Non-pharmacological interventions against COVID-19 could exert an incidence reduction co-benefits effect on TB in Ningxia, and the effect may be positively related to the grade of emergency response. -
表 1 不同模型的AIC值和log likelihood值汇总表
模型 AIC值 log likelihood值 ARIMA(1,1,2)(0,1,0)12 1774.05 – 883.02 ARIMA(1,1,2)(0,1,1)12 1736.24 – 863.12 ARIMA(1,1,2)(0,1,2)12 1738.13 – 863.07 ARIMA(1,1,2)(1,1,0)12 1737.91 – 863.95 ARIMA(1,1,2)(1,1,1)12 1739.76 – 863.88 ARIMA(1,1,2)(1,1,2)12 1734.09 – 860.05 ARIMA(1,1,2)(2,1,0)12 1739.43 – 863.71 ARIMA(1,1,2)(2,1,1)12 1739.84 – 862.92 ARIMA(1,1,2)(2,1,2)12 1737.76 – 860.88 表 2 新型冠状病毒肺炎的非药物干预策略对宁夏回族自治区肺结核病的伴随效应
应急响
应级别月份 2020年与2019年同期的实际变化情况 2020年肺结核发病的预期变化情况 N2020real N2019real 相对
变化率(%)实际相对变化率(%) N2020real N2020pre 相对
变化率(%)预期相对变化率(%) 响应前 1 152 186 – 18.28 – 18.28 152 233 – 34.76 (– 20.72~ – 48.80) – 34.76 (– 20.72~ – 48.80) 一级响应 2 63 157 – 59.87 – 59.87 63 173 – 63.58 (– 49.54~ – 77.62) – 63.58 (– 49.54~ – 77.62) 二级响应 3 132 212 – 37.74 – 31.24 132 253 – 47.83 (– 33.79~ – 61.87) – 29.61 (– 15.57~ – 43.65) 4 196 265 – 26.04 196 213 – 7.98 (6.06~ – 22.02) 三级响应 5 151 213 – 29.11 – 17.64 151 222 – 31.98 (– 17.94~ – 46.02) – 8.59 (5.45~ – 22.63) 6 186 176 5.68 186 176 5.68 (19.72~ – 8.36) 7 198 266 – 25.56 198 187 5.88 (19.92~ – 8.16) 8 184 186 – 1.08 184 183 0.55 (14.59~ – 13.49) 9 177 206 – 14.08 177 187 – 5.35 (8.69~ – 19.39) 10 137 169 – 18.93 137 173 – 20.81 (– 6.77~ – 34.85) 11 151 201 – 24.88 151 173 – 12.72 (1.32~ – 26.76) 12 156 210 – 25.71 156 165 – 5.45 (8.59~ – 19.49) -
[1] World Health Organization. Global tuberculosis report 2021. World Health Organization[EB/OL]. (2022 – 01 – 19)[2022 – 01 – 19]. https://apps.who.int/iris/handle/10665/346387. [2] 疾病预防控制局. 2020年全国法定传染病疫情概况[EB/OL]. (2021 – 03 – 12)[2022 – 01 – 19]. http://www.nhc.gov.cn/jkj/s3578/202103/f1a448b7df7d4760976fea6d55834966.shtml. [3] Rana MS, Usman M, Alam MM, et al. Impact of COVID - 19 preventive measures on other infectious and non - infectious respiratory diseases in Pakistan[J]. Journal of Infection, 2021, 82(5): e31 – e32. doi: 10.1016/j.jinf.2021.01.018 [4] Liu XS, Xu X, Li GG, et al. Differential impact of non - pharmaceutical public health interventions on COVID - 19 epidemics in the United States[J]. BMC Public Health, 2021, 21(1): 965. doi: 10.1186/s12889-021-10950-2 [5] 邹旋, 吴永胜, 刘晓剑, 等. 深圳市新型冠状病毒肺炎应急响应策略和措施效果评价[J]. 中华流行病学杂志, 2020, 41(8): 1225 – 1230. doi: 10.3760/cma.j.cn112338-20200316-00360 [6] 李艳君. 气象因素对呼吸性传染病的发病分布影响及非药物干预措施对法定报告传染病监测影响的时间序列分析[D]. 沈阳: 中国医科大学, 2021,doi: 10.27652/d.cnki.gzyku.2021.000290. [7] Xiao JP, Dai JY, Hu JX, et al. Co - benefits of nonpharmaceutical intervention against COVID - 19 on infectious diseases in China: a large population - based observational study[J]. The Lancet Regional Health Western Pacific, 2021, 17: 100282. doi: 10.1016/j.lanwpc.2021.100282 [8] 宁夏回族自治区人民政府. 宁夏启动重大突发公共卫生事件一级响应[EB/OL]. (2021 – 01 – 25)[2022 – 01 – 19]. http://www.nx.gov.cn/zwxx_11337/wztt/202001/t20200125_1932950.html. [9] 宁夏回族自治区人民政府. 宁夏重大突发公共卫生事件一级响应调整为二级响应[EB/OL]. (2021 – 02 – 28)[2022 – 01 – 19]. http://www.nx.gov.cn/zwxx_11337/wztt/202002/t20200228_1974311.html. [10] 宁夏回族自治区人民政府. 今天17时起宁夏应急响应级别由二级调整为三级[EB/OL]. (2021 – 05 – 06)[2022 – 01 – 19]. http://www.nx.gov.cn/zwxx_11337/wztt/202005/t20200506_2056196.html. [11] 姜丹丹, 张晓蕾. 应用季节指数法和季节性指数平滑法预测某院妇科肿瘤门诊人次[J]. 中国病案, 2021, 22(1): 65 – 68. doi: 10.3969/j.issn.1672-2566.2021.01.025 [12] 运玲, 王福才, 张秋芬, 等. ARIMA模型在德国小蠊密度季节消长预测中应用及抗药性[J]. 中国公共卫生, 2020, 36(3): 410 – 415. doi: 10.11847/zgggws1124086 [13] 刘超, 孟园园, 张庆雯. 手足口病发病预测4种时间序列预测模型比较[J]. 中国公共卫生, 2022, 38(2): 218 – 223. doi: 10.11847/zgggws1133517 [14] 茅蓉, 王远航, 葛锐. 基于自回归移动平均模型的浙江省肺结核发病趋势预测[J/OL]. 疾病监测: 1-8(2022 – 01 – 18)[2022 – 01 – 19]. http://www.jbjc.org/article/id/bd9d4dab-859b-452f-a561-9eb3b45cfa68. [15] 高雅, 王伶, 吴伟, 等. 辽宁省手足口病疫情季节性ARIMA模型预测效果评价[J]. 中国公共卫生, 2017, 33(10): 1482 – 1484. doi: 10.11847/zgggws2017-33-10-16 [16] 周扬, 梁士杰. ARIMA乘积季节模型在郑州市肺结核月发病趋势预测中的应用[J]. 中国卫生统计, 2021, 38(4): 554 – 555. [17] 王若男, 杨正贵, 马宁, 等. 2004 — 2019年宁夏地区肺结核病的时空分布研究[J/OL]. 疾病监测: 1-8(2022 – 01 – 05)[2022 – 01 – 19]. http://www.jbjc.org/article/id/287feace-ceeb-4388-a345-b2f80814efde. [18] 杨振, 聂艳武, 孙亚红, 等. 基于Prophet等时间序列季节模型的肺结核发病预测及对比分析[J]. 现代预防医学, 2021, 48(21): 3841 – 3846, 3883. [19] 梁达, 商越, 王兆芬, 等. 青海省肺结核发病与气象因素的时间序列分析[J]. 中华疾病控制杂志, 2021, 25(10): 1186 – 1193. doi: 10.16462/j.cnki.zhjbkz.2021.10.013 [20] 丁哲渊, 吴昊澄, 吴晨, 等. 浙江省新型冠状病毒肺炎疫情应急响应期间其他法定传染病监测分析[J]. 疾病监测, 2020, 35(8): 746 – 752. doi: 10.3784/j.issn.1003-9961.2020.08.015 [21] 姜海波, 史宏博, 冯宏伟, 等. 新型冠状病毒肺炎疫情对浙江省宁波市艾滋病防制工作的影响评估[J]. 疾病监测, 2020, 35(9): 803 – 808. doi: 10.3784/j.issn.1003-9961.2020.09.008 [22] 汤吉军. 为什么过去统计数据无法预测未来 —— 兼论经济计量学的局限性[J]. 经济学家, 2016(5): 29 – 36. doi: 10.16158/j.cnki.51-1312/f.2016.05.005 [23] 李方友, 孟凡强, 禹长兰, 等. 基于自回归滑动平均混合模型的安丘市肾综合征出血热发病率预测研究[J]. 预防医学论坛, 2020, 26(2): 109 – 112. doi: 10.16406/j.pmt.issn.1672-9153.2020.02.008 -