| BackgroundThe global climate is undergoing significant changes characterized by global warming.Global climate change poses a series of serious threats to human health,including chronic non-communicable diseases,communicable diseases,mental diseases and adverse pregnancy outcomes.Meteorological factors have an important impact on the spread of infectious diseases.Previous studies have showed that the incidence of infectious diarrhea,influenza,hand foot and mouth disease,hemorrhagic fever renal syndrome,mumps are affected by meteorological factors.It is of great significance to explore the influence of meteorological factors on people’s health for improving people’s ability to cope with climate change in public health.Influenza(flu),an acute respiratory infectious disease caused by influenza viruses,is spread by droplets and direct contact.Influenza virus has strong infectivity and mutates easily.Therefore,influenza is likely to cause a widespread epidemic.Influenza is a seasonal infectious disease,and meteorological factors have an important impact on the incidence of influenza.Many studies on the correlation between meteorological factors and influenza have been carried out,suggesting that meteorological factors such as temperature,humidity,pressure,rainfall can affect the risk of influenza.Previous studies have shown that the effect of meteorological factors on the incidence of some infectious diseases is nonlinear effect instead of linear effect.And sometimes there is a single threshold or double threshold.In addition,there may be interactions between meteorological factors.However,these effects have not been thoroughly explored in influenza.Diurnal temperature range(DTR)is the difference between the highest and lowest temperatures in a day.It is used to represent the amplitude of the temperature daily variation.And it is an indicator reflecting the short-term change of temperature.Many researches on the health effects of DTR have been carried out,providing evidence for the effect of DTR on cardiovascular disease,chronic obstructive pulmonary disease,mental diseases,and so on.However,there is still a lack of research on the relationship between DTR and influenza.Exploring the influence of DTR on influenza can not only expand the research scope of the relationship between meteorological factors and influenza,but also provide more theoretical basis for the prevention and control of influenza.Objectives(1)Understanding the distributional characteristic of influenza in Beijing on the dimension of time,space and population;(2)Exploring the threshold and nonlinear effect of meteorological factors on influenza in Beijing;(3)Exploring the interaction between meteorological factors on influenza;(4)Quantifying the lag effects of DTR on influenza in Beijing,and screening vulnerable populations;(5)Exploring the interaction between DTR and temperature,DTR and relative humidity on influenzaMethodsBeijing city was chosen as research site.Influenzas data from 2014 to 2016 were collected from the National Notifiable Disease Surveillance System.Meteorological data of the same period were obtained from China Meteorological Data Sharing Service Network.The population data of Beijing were obtained from Statistical Yearbooks of Beijing.Descriptive analysis was conducted for the incidence of influenza and distributional characteristics of the influenza on the dimension of time,space and population.After the cleaning of influenza data and meteorological factors data,the data was transformed into the form of time series.The difference between daily maximum and minimum temperature was taken as the DTR.Spearman rank correlation was used to analyze the correlation among various meteorological factors.In order to avoid multicollinearity,two closely related meteorological factors are not included in one model.The distributed lag non-linear models were used to explore the threshold and the nonlinear effects of meteorological factors on influenza.The daily rainfall was put into the model as binary variable due to the its particularity.The rain day was assigned as 1,while the non-rain day was assigned as 0.Other meteorological factors were put into the models as continuous variables.A generalized additive model based on thin plate splines was used to fit the interaction between meteorological factors,which was aimed to plot the graphs of"factor-factor-outcomes".A generalized linear model was used to quantify the interaction between meteorological factors,and the Relative Excess Risk due to Interaction(RERI),Attributable ratio(AP),Synergy Index(S)was used to evaluate the values of interaction.To explore the influence of DTR on the incidence of influenza,a generalized additive model was used to determine whether the association between DTR and influenza is a linear one.And then a distributed lag nonlinear model was used to quantify the lagged effect of DTR on influenza.The whole population was divided into subgroups according to gender and age to screen vulnerable populations.The effects of DTR on the incidence of influenza in each subgroup were discussed respectively.The interaction between DTR and temperature,DTR and relative humidity was also analyzed.ResultsA total of 34 804 cases of influenza were reported in Beijing from 2014 to 2016,with a male-female ratio of 0.95:1.The variation of influenza incidence is obviously seasonal,which is high in winter and spring and low in summer and autumn.Children under 14 years old and people over the age of 70 accounted for the highest proportion.For occupational distribution,residentially-scattered children(6919 cases),childcare(6629 cases)and students(5540 cases)accounted for the highest proportion.The incidence of influenza in Chaoyang district,Shunyi district,Huairou county were at a relatively high level,while the incidence in Daxing district,Yanqing county was at a relatively low level.During the study period,the mean temperature was 13.93℃,the mean relative humidity was 52.75%,and the mean DTR was 10.15℃.There was a strong correlation(|rs|>0.6)between the average air pressure and the average air temperature(rs=-0.878),sunshine duration and DTR(rs=0.657).With the median of each meteorological factors as the control,temperature lower than-1.3℃,low humidity,wind speed above 5.7 m/s,air pressure above 1031.4 hPa increase the risk of flu in Beijing.With non-rain day as the control,the rain day showed a protective effect on influenza.The 3D diagram of the interaction showed that the number of influenza cases are higher at low temperature-low humidity,high pressure-low humidity,low humidity-high wind speed,high pressure-high wind speed.High pressure and low humidity have a synergistic effect on influenza with RERI of 0.158(95%CI:0.025-0.290)and AP of 0.136(95%CI:0.015-0.257).The generalized additive model showed that there is a linear relationship between the DTR and the number of influenza cases.At lag 0 day and 1 day,the RR values of DTR on the incidence of influenza were 1.012(95%CI:1.003～1.021)and 1.011(95%Cl:1.003～1.019),respectively.The cumulative lag effect of DTR on the number of influenza cases reached the maximum at lag 0-3 days,and the RR value was 1.029(95%CI:1.012～1.045).The RR values of male,female,people aged 0-4 years,people aged 15-64 years and people aged 65 years and older were 1.014(lagO,95%CI:1.004～1.025),1.011(lagl,95%CI:1.003～1.020),1.012(lagl,95%CI:1.001～1.024),1.016(lag0,95%CI:1.006～1.027),1.022(lag0,95%CI:1.006～1.039),respectively.But no significant result was observed in people aged 5-14 years.The cumulative RR value of lag 0-3 days changed little when changed maximum lag time from 5 days to 11 days and degree of freedom for time from 4 to 10.The risk of flu in high DTR-low temperature was higher than in low DTR-high temperature with RR value of 1.543(95%CI:1.245～1.913).The risk of flu in high DTR-low relative humidity was higher than in low DTR-relative humidity with RR value of 1.092(95%CI:1.032～1.156).But no interaction between low DTR and high temperature,low DTR and high relative humidity was observed.Conclusion(1)The flu season in Beijing is from November to April.And the incidence of children and elderly people are relatively high.(2)With the median of each meteorological factors as the control,temperature lower than-1.3℃,low humidity,wind speed above 5.7 m/s,air pressure above 1031.4 hPa increase the risk of flu in Beijing.With non-rain day as the control,the rain day showed a protective effect on influenza.(3)There is an interaction between air pressure and humidity,which manifests as a synergistic effect between high pressure and low humidity.(4)The increase of DTR was associated with the increased risk of influenza in Beijing.And the cumulative lag effect reached maximum at lag 0-3 days. |
PDF Full Text Request