唐山地区结核病发病人数时间序列分析

Time series analysis of the number of tuberculosis cases in Tangshan area

目的 通过时间序列分析模型对不同时间唐山地区结核病发病人数进行预测预警,为该地区科学开展结核病疫情防控工作提供参考依据。方法 收集2005年1月至2021年12月唐山地区结核病发病人数,利用季节性自回归综合移动平均模型(seasonal autoregressive integrated moving average model, SARIMA)预测2022年结核病发病人数,同时利用该模型和秩检验,探索2020年新冠肺炎疫情期间该地区结核病发病预测人数与观察人数的差异性。结果 2005年1月至2021年12月,ARIMA(1,1,0)(1,1,2)s模型与实际结核病登记发病人数拟合效果较好(AR=-0.530,ARs=-0.967,MAs=0.861,P<0.05;Stationary R^(2)=0.558,R2=0.634,BIC=7.887;Ljung-Box Q=25.605,P<0.05),每年3、4和12月是结核病发病高峰,2022年发病人数预测值为1 800人。2005年至2019年,ARIMA(1,1,0)(1,1,2)s模型与实际发病人数拟合效果较好(AR=-0.544,ARs=-0.840,MAs=0.697,P<0.05;Stationary R^(2)=0.582,R^(2)=0.621,BIC=7.939;Ljung-Box Q=24.211,P<0.05),每年3、4和12月是发病高峰,2020年发病人数预测值为1 985人,2020年1月至5月结核病发病人数观察值与预测值差异有统计学意义(Z=-2.023,P<0.05)。结论 需提高唐山地区3、4和12月结核病的预警强度,防止该时期结核病疫情流行;同时加强新型冠状病毒肺炎等传染病疫情期间的结核病防治机构工作人员调度协调,完善应急体系,有效保障疫情期间结核病患者登记就诊。

Objective The time series analysis model was used to predict and warn the number of tuberculosis(TB) cases in Tangshan area in different time, which provided a reference for scientific prevention and control of TB epidemic in this area.Methods The number of monthly TB cases in Tangshan from January 2005 to December 2021 was collected, and the seasonal autoregressive integrated moving average(SARIMA) model was used to predict the number of TB cases in 2022.Meanwhile, the difference between the predicted number of TB cases and the actual observed number of TB cases in the area was explored during the period of COVID-19 in 2020 by this model and rank test.Results From January 2005 to December 2021,the ARIMA(1,1,0)(1,1,2)smodel was fitted well with the actual observed number of TB cases(AR=-0.530,ARs=-0.967,MAs=0.861,P<0.05;Stationary R^(2)=0.558,R^(2)=0.634,BIC=7.887;Ljung-Box Q=25.605,P<0.05),with peaks TB incidence in March, April, and December every year, and the predicted number of TB cases in 2020 was 1 800.From 2005 to 2019,ARIMA(1,1,0)(0,1,2)smodel was fitted well with the actual number of cases(AR=-0.544,ARs=-0.840,MAs=0.697,P<0.05;Stationary R^(2)=0.582,R^(2)=0.621,BIC=7.939;Ljung-Box Q=24.211,P<0.05),with peaks TB incidence in March, April, and December every year, and the predicted number of TB cases in 2020 was 1 985.The observed and predicted number of TB cases from January 2020 to May 2020 were statistically significant(Z=-2.023,P<0.05).Conclusion It is necessary to increase the intensity of early warning of TB in March, April, and December every year in Tangshan to prevent the epidemic of TB.At the same time, the coordination of the staff of TB prevention institutions and the emergency system should be strengthened during the epidemic situation of COVID-19,and effectively ensure the registration and medical treatment of TB patients during the epidemic situation.