Analysis Of The Tuberculosis Epidemic Characteristics And Prediction Of Its Incidence In Shanxi Province From 2010 To 2020

Objective:1.To analyze the spatial and temporal distribution characteristics of Tuberculosis(TB)in Shanxi Province from 2010 to 2020,including regional distribution,temporal distribution,age distribution,gender distribution,and occupational distribution.Also,to identify high-incidence areas,high-incidence times,and key populations of TB,thus,providing reference for the prevention and control of TB in Shanxi Province from both temporal and spatial perspectives.2.To construct an Autoregressive integrated moving average(ARIMA)multiplicative seasonal model to predict the incidence rate of TB from January to December 2021 in Shanxi Province,based on the monthly incidence rate data of TB in Shanxi Province from 2010 to 2020,and to analyze its epidemic trend,providing theoretical support for government decision-making,medical material reserves,and TB monitoring by the disease control department.Methods:1.First,we collected information on TB patients in Shanxi Province from 2010 to2020 from the Chinese Disease Prevention and Control Information System-Tuberculosis Management Information System,including the region of onset,time of onset,age,gender,and occupation.Second,we used Mann-Kendall trend test to analyze the dynamic trend of TB in Shanxi Province from 2010 to 2020.Afterwards,descriptive epidemiological methods were used to describe the spatial and temporal distribution,connect TB incidence data with Shanxi Province GIS spatial data,and Arc GIS 10.8software was used to draw annual and average incidence rate maps of TB in Shanxi Province from 2010 to 2020.Besides,Microsoft Excel 2019 was employed to organize the TB data from Shanxi Province from 2010 to 2020,and accordingly the high incidence months,high incidence age groups,male-to-female ratio,and high-risk occupations for TB in Shanxi Province were analyzed.2.We divided the TB incidence rate data in Shanxi Province from 2010 to 2020 into two parts.The monthly incidence rate data of TB from 2010 to 2019 was used to construct an ARIMA multiplicative seasonal model using R 4.1.0 software,and the monthly incidence rate data of TB in 2020 was used to test the model.Afterwards,the incidence rate of TB in Shanxi Province from January to December 2021 was predicted,and its epidemic trend was analyzed.Results:1.The dynamic trend of TB incidence rate in Shanxi Province from 2010 to 2020: a total of 183,408 cases of TB were reported in Shanxi Province from 2010 to 2020,with an average incidence rate of 47.29/100,000.The incidence rate decreased from68.25/100,000 in 2010 to 27.71/100,000 in 2020,a decrease of 59.40% over 11 years(Mann-Kendall trend test,Z=-3.893,P<0.001),with an increase in 2016.2.Triple distribution characteristics: In terms of regional distribution,the average incidence rate of TB in Shanxi Province from 2010 to 2020 was 47.29/100,000.The top three cities with the highest incidence rates were: Datong city(65.55/100,000),Xinzhou city(61.47/100,000),and Yuncheng city(56.80/100,000).In terms of time distribution,cases of TB were reported in Shanxi Province in every month from January to December from 2010 to 2020,with more cases reported in April,May,and June,and fewer cases in January,February,and October.There was a small peak in March.Regarding population distribution,the incidence rate increased with age,with the highest incidence rate in people over 65 years old,followed by those between 15-64 years old and the lowest incidence rate in those under 14 years old.The age group with the highest number of cases was 15-64 years old,especially high school and university students aged 15-24.The group with the second-highest number of cases was people over 65 years old,while the lowest number of cases was in the age group under 14 years old.In terms of gender,the incidence rate was higher in males than in females,with a male-to-female ratio of1.85:1.In terms of occupation,the main occupations with high incidence rates were farmers,students,and those who work in domestic services or are unemployed.3.Construction and prediction of ARIMA multiplicative seasonal model: Based on the monthly TB incidence rate data from January 2010 to December 2019,an ARIMA(0,1,1)(1,1,1)12 model was fitted.The Akaike information criterion,root mean square error,mean absolute error,mean absolute percentage error,and mean absolute scaled error of the model were 190.91,0.50,0.34,9.00,and 0.53,respectively.The model was validated using the monthly TB incidence rate data from January to December 2020,and the mean absolute percentage error was found to be 11.25%.The predicted TB incidence rates in Shanxi Province from January to December 2021 were1.39/100,000(95%CI: 0.12/100,000-2.67/100,000),0.97/100,000(95%CI:0-2.26/100,000),2.61/100,000(95%CI: 1.30/100,000-3.91/100,000),2.42/100,000(95%CI: 1.10/100,000-3.73/100,000),2.38/100,000(95%CI: 1.06/100,000-3.71/100,000),2.33/100,000(95%CI: 0.99/100,000-3.68/100,000),2.13/100,000(95%CI:0.78/100,000-3.49/100,000),2.07/100,000(95%CI: 0.70/100,000-3.44/100,000),2.06/100,000(95%CI: 0.68/100,000-3.44/100,000),1.47/100,000(95%CI:0.07/100,000-2.86/100,000),2.01/100,000(95%CI: 0.61/100,000-3.42/100,000),and2.18/100,000(95%CI: 0.76/100,000-3.60/100,000).Conclusion:1.From 2010 to 2020,the overall trend of TB incidence rate in Shanxi Province showed a decrease,with an increase in 2016.2.The cities with higher TB incidence rates in Shanxi Province from 2010 to 2020 were Datong,Xinzhou,and Yuncheng.The months with more cases were March,April,May,and June.The incidence rate increased with age,and the highest incidence rate was among people aged 65 and above.The age group with the most cases was 15-64 years old,especially high school and college students aged 15-24.The incidence rate was higher in males than in females,and most cases were found among farmers.3.The ARIMA(0,1,1)(1,1,1)12 model was used to predict the TB incidence rate in Shanxi Province from January to December 2021,and the predicted incidence rate ranged from 0.97/10,000 to 2.61/10,000,showing a decreasing trend.The average absolute percentage error was 9,indicating that the model’s prediction performance was good and had practical significance for the prevention and control of TB.