Interruption time series analysis using autoregressive integrated moving average model: evaluating the impact of COVID-19 on the epidemic trend of gonorrhea in China
Abstract Background Interrupted time series (ITS) analysis is a growing method for assessing intervention impacts on diseases. However, it remains unstudied how the COVID-19 outbreak impacts gonorrhea. This study aimed to evaluate the effect of COVID-19 on gonorrhea and predict gonorrhea epidemics u...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2023-10-01
|
| Series: | BMC Public Health |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12889-023-16953-5 |
| _version_ | 1827707876521541632 |
|---|---|
| author | Yanyan Li Xingyan Liu Xinxiao Li Chenlu Xue Bingjie Zhang Yongbin Wang |
| author_facet | Yanyan Li Xingyan Liu Xinxiao Li Chenlu Xue Bingjie Zhang Yongbin Wang |
| author_sort | Yanyan Li |
| collection | DOAJ |
| description | Abstract Background Interrupted time series (ITS) analysis is a growing method for assessing intervention impacts on diseases. However, it remains unstudied how the COVID-19 outbreak impacts gonorrhea. This study aimed to evaluate the effect of COVID-19 on gonorrhea and predict gonorrhea epidemics using the ITS-autoregressive integrated moving average (ARIMA) model. Methods The number of gonorrhea cases reported in China from January 2005 to September 2022 was collected. Statistical descriptions were applied to indicate the overall epidemiological characteristics of the data, and then the ITS-ARIMA was established. Additionally, we compared the forecasting abilities of ITS-ARIMA with Bayesian structural time series (BSTS), and discussed the model selection process, transfer function, check model fitting, and interpretation of results. Result During 2005–2022, the total cases of gonorrhea were 2,165,048, with an annual average incidence rate of 8.99 per 100,000 people. The highest incidence rate was 14.2 per 100,000 people in 2005 and the lowest was 6.9 per 100,000 people in 2012. The optimal model was ARIMA (0,1, (1,3)) (0,1,1)12 (Akaike’s information criterion = 3293.93). When predicting the gonorrhea incidence, the mean absolute percentage error under the ARIMA (16.45%) was smaller than that under the BSTS (22.48%). The study found a 62.4% reduction in gonorrhea during the first-level response, a 46.47% reduction during the second-level response, and an increase of 3.6% during the third-level response. The final model estimated a step change of − 2171 (95% confidence interval [CI] − 3698 to − 644) cases and an impulse change of − 1359 (95% CI − 2381 to − 338) cases. Using the ITS-ARIMA to evaluate the effect of COVID-19 on gonorrhea, the gonorrhea incidence showed a temporary decline before rebounding to pre-COVID-19 levels in China. Conclusion ITS analysis is a valuable tool for gauging intervention effectiveness, providing flexibility in modelling various impacts. The ITS-ARIMA model can adeptly explain potential trends, autocorrelation, and seasonality. Gonorrhea, marked by periodicity and seasonality, exhibited a downward trend under the influence of COVID-19 intervention. The ITS-ARIMA outperformed the BSTS, offering superior predictive capabilities for the gonorrhea incidence trend in China. |
| first_indexed | 2024-03-10T16:54:52Z |
| format | Article |
| id | doaj.art-54adf82623e44d799e85656f7421af15 |
| institution | Directory Open Access Journal |
| issn | 1471-2458 |
| language | English |
| last_indexed | 2024-03-10T16:54:52Z |
| publishDate | 2023-10-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Public Health |
| spelling | doaj.art-54adf82623e44d799e85656f7421af152023-11-20T11:10:39ZengBMCBMC Public Health1471-24582023-10-0123111110.1186/s12889-023-16953-5Interruption time series analysis using autoregressive integrated moving average model: evaluating the impact of COVID-19 on the epidemic trend of gonorrhea in ChinaYanyan Li0Xingyan Liu1Xinxiao Li2Chenlu Xue3Bingjie Zhang4Yongbin Wang5Department of Epidemiology and Health Statistics, School of Public Health, The First Affiliated Hospital, Xinxiang Medical UniversityDepartment of Epidemiology and Health Statistics, School of Public Health, The First Affiliated Hospital, Xinxiang Medical UniversityDepartment of Epidemiology and Health Statistics, School of Public Health, The First Affiliated Hospital, Xinxiang Medical UniversityDepartment of Epidemiology and Health Statistics, School of Public Health, The First Affiliated Hospital, Xinxiang Medical UniversityDepartment of Epidemiology and Health Statistics, School of Public Health, The First Affiliated Hospital, Xinxiang Medical UniversityDepartment of Epidemiology and Health Statistics, School of Public Health, The First Affiliated Hospital, Xinxiang Medical UniversityAbstract Background Interrupted time series (ITS) analysis is a growing method for assessing intervention impacts on diseases. However, it remains unstudied how the COVID-19 outbreak impacts gonorrhea. This study aimed to evaluate the effect of COVID-19 on gonorrhea and predict gonorrhea epidemics using the ITS-autoregressive integrated moving average (ARIMA) model. Methods The number of gonorrhea cases reported in China from January 2005 to September 2022 was collected. Statistical descriptions were applied to indicate the overall epidemiological characteristics of the data, and then the ITS-ARIMA was established. Additionally, we compared the forecasting abilities of ITS-ARIMA with Bayesian structural time series (BSTS), and discussed the model selection process, transfer function, check model fitting, and interpretation of results. Result During 2005–2022, the total cases of gonorrhea were 2,165,048, with an annual average incidence rate of 8.99 per 100,000 people. The highest incidence rate was 14.2 per 100,000 people in 2005 and the lowest was 6.9 per 100,000 people in 2012. The optimal model was ARIMA (0,1, (1,3)) (0,1,1)12 (Akaike’s information criterion = 3293.93). When predicting the gonorrhea incidence, the mean absolute percentage error under the ARIMA (16.45%) was smaller than that under the BSTS (22.48%). The study found a 62.4% reduction in gonorrhea during the first-level response, a 46.47% reduction during the second-level response, and an increase of 3.6% during the third-level response. The final model estimated a step change of − 2171 (95% confidence interval [CI] − 3698 to − 644) cases and an impulse change of − 1359 (95% CI − 2381 to − 338) cases. Using the ITS-ARIMA to evaluate the effect of COVID-19 on gonorrhea, the gonorrhea incidence showed a temporary decline before rebounding to pre-COVID-19 levels in China. Conclusion ITS analysis is a valuable tool for gauging intervention effectiveness, providing flexibility in modelling various impacts. The ITS-ARIMA model can adeptly explain potential trends, autocorrelation, and seasonality. Gonorrhea, marked by periodicity and seasonality, exhibited a downward trend under the influence of COVID-19 intervention. The ITS-ARIMA outperformed the BSTS, offering superior predictive capabilities for the gonorrhea incidence trend in China.https://doi.org/10.1186/s12889-023-16953-5Autoregressive integrated moving average modelsInterrupted time series analysisIntervention analysisCOVID-19Gonorrhea |
| spellingShingle | Yanyan Li Xingyan Liu Xinxiao Li Chenlu Xue Bingjie Zhang Yongbin Wang Interruption time series analysis using autoregressive integrated moving average model: evaluating the impact of COVID-19 on the epidemic trend of gonorrhea in China BMC Public Health Autoregressive integrated moving average models Interrupted time series analysis Intervention analysis COVID-19 Gonorrhea |
| title | Interruption time series analysis using autoregressive integrated moving average model: evaluating the impact of COVID-19 on the epidemic trend of gonorrhea in China |
| title_full | Interruption time series analysis using autoregressive integrated moving average model: evaluating the impact of COVID-19 on the epidemic trend of gonorrhea in China |
| title_fullStr | Interruption time series analysis using autoregressive integrated moving average model: evaluating the impact of COVID-19 on the epidemic trend of gonorrhea in China |
| title_full_unstemmed | Interruption time series analysis using autoregressive integrated moving average model: evaluating the impact of COVID-19 on the epidemic trend of gonorrhea in China |
| title_short | Interruption time series analysis using autoregressive integrated moving average model: evaluating the impact of COVID-19 on the epidemic trend of gonorrhea in China |
| title_sort | interruption time series analysis using autoregressive integrated moving average model evaluating the impact of covid 19 on the epidemic trend of gonorrhea in china |
| topic | Autoregressive integrated moving average models Interrupted time series analysis Intervention analysis COVID-19 Gonorrhea |
| url | https://doi.org/10.1186/s12889-023-16953-5 |
| work_keys_str_mv | AT yanyanli interruptiontimeseriesanalysisusingautoregressiveintegratedmovingaveragemodelevaluatingtheimpactofcovid19ontheepidemictrendofgonorrheainchina AT xingyanliu interruptiontimeseriesanalysisusingautoregressiveintegratedmovingaveragemodelevaluatingtheimpactofcovid19ontheepidemictrendofgonorrheainchina AT xinxiaoli interruptiontimeseriesanalysisusingautoregressiveintegratedmovingaveragemodelevaluatingtheimpactofcovid19ontheepidemictrendofgonorrheainchina AT chenluxue interruptiontimeseriesanalysisusingautoregressiveintegratedmovingaveragemodelevaluatingtheimpactofcovid19ontheepidemictrendofgonorrheainchina AT bingjiezhang interruptiontimeseriesanalysisusingautoregressiveintegratedmovingaveragemodelevaluatingtheimpactofcovid19ontheepidemictrendofgonorrheainchina AT yongbinwang interruptiontimeseriesanalysisusingautoregressiveintegratedmovingaveragemodelevaluatingtheimpactofcovid19ontheepidemictrendofgonorrheainchina |