Objective   To explore the impact of meteorological factors on influenza incidence in Xiamen city of Fujian province and to provide evidences for prediction and early warning of influenza epidemic.

  Methods   The data on daily meteorological factors such as average atmospheric pressure, atmospheric pressure difference, relative humidity, temperature, temperature difference and sunshine hours and the data on daily number of influenza incidence reported in Xiamen city during January 2013 to February 2019 were collected and analyzed using correlation and distributed lag nonlinear model.

  Results   Totally 10 573 influenza cases were reported in the city during the period; the male to female ratio of the cases was 1.29 ∶ 1 and 52.44% and 32.19% of the cases were 4 – 12 and 0 – 3 years old children. The daily atmospheric pressure and pressure difference were positively correlated with the incidence of influenza (r > 0), but the daily average temperature and sunshine hours were reversely correlated with the incidence (r < 0). Low atmospheric pressure (< 980 hPa) was a lag 0 – 5 day risk factor for influenza incidents, with the highest lag 4-day risk correlated with an atmospheric pressure of 999 hPa (relative risk RR = 1.05, 95% confidence interval 95% CI : 1.01 – 1.11). A ‘U’ shaped correlation between atmospheric pressure difference and influenza incidence was observed and both the atmospheric pressure difference of > 18 hPa and < 3 hPa were risk factors for influenza incidence. There was also a ‘U’ shaped correlation between average daily temperature and influenza incidence and both a daily temperatures of < 9 °C and > 23 °C were risk factors for influenza incidents, with the highest lag 4-day risk (RR = 2.40, 95% CI: 1.01 – 5.71) for the average daily temperature of 1 °C. The sunshine time of 1 – 5 hours and 11 – 13 hours showed lag 1 – 15-day risks for influenza incidents and the highest risk was for lag 15-day sunshine time of 13 hours (RR = 8.79, 95% CI: 1.22 – 63.30). The cumulative effect of low atmospheric pressure of 975 hPa showed a significant lag 0 – 15-day cumulative risk of influenza incidents among children aged 4 – 12 year (RR = 7.82, 95% CI: 6.40 – 9.55). The difference in daily atmospheric pressure of 5 hPa showed the highest lag 0 – 15-day cumulative risk of influenza incident among the people aged 60 years and above (RR = 3.69, 95% CI: 1.56 – 8.68). Significant lag 0 – 15-day cumulative risk of influenza incidents related to low temperature of 1 °C was observed among the people aged 13 – 59 years (RR = 40.82, 95% CI: 3.53 – 554.19) and that related to high temperature of 31 °C was observed among those aged 4 – 12 years. The most significant lag 0 – 15-day cumulative risk of influenza incident (RR = 20.41, 95% CI : 4.12 – 99.34) related to daily sunshine time of 13 hours was observed among 4 – 12 years old children.

  Conclusion   Meteorological factors including daily average atmospheric pressure, daily atmospheric pressure difference, daily average temperature and sunshine hours have lag influence on the incidence of influenza in Xiamen city and the results need to be concerned in early warning and prevention of influenza epidemic.