Objective: To analyze the epidemicological characteristics of influenza in
Yuxi city from 2010 to 2019, and to establish an autoregressive integrated moving average(ARIMA) model to predict the trend of influenza incidence. Methods: The data of influenza was analyzed
by descriptive epidemiological methods, and the ARIMA model was established by SPSS 25.0 software. Results: A total of 9 925 influenza-like
cases were detected in Yuxi influenza network laboratory, and 2 849 nucleic acid positive
cases were detected, with a positive rate of 28.71%. The nucleic acid positive cases
showed an upward trend year by year in general, with an average increment speed of 2.01%. Among influenza positive
cases, the ratio of
male to female was 1.25∶1, and children
aged 5-14 accounted for the highest proportion (40.86%). Influenza
had a strong seasonality(concentration was 0.60). The dominant influenza virus strains changed year by year. In 2010
and 2018, A/H1N1 pdm
was dominant; in 2011 and
2012, B/Victoria
was dominant; in 2013 and
2016, B/Yamagata
was dominant; and in 2015
and 2017, A/H3N2 was
dominant; in 2014, B/Yamagata and A/H1N1 pdm
dominated, and in 2019, B/Victoria and A/H3N2 dominated.
The best prediction model was ARIMA(1,0,0)(0,1,1)12, whose
standardized BIC value was 5.873, Ljung-Box Q was 21.258, P was 0.169. The actual values of the prediction results were
within the 95% confidence interval(95%CI) of the
predicted value, and the
average relative error was 10.14%. The model was used to predict a decreasing
trend of influenza incidence from January to June in 2020. Conclusion: The influenza in Xuyi city has
obvious seasonality and the nucleic acid positive cases show an upward trend
year by year in general, and the dominant influenza strains change year by year. ARIMA(1,0,0)(0,1,1)12 model can effectively predict
the trend of short-term influenza in Yuxi city.