| Background:Influenza causes an extremely high burden of disease around the world.It is of great significance to accurately estimate the intensity and characteristics of influenza death burden for formulating targeted influenza prevention and control policies.However,there are no studies that have explored the excess mortality rates for multiple underlying causes of death associated with influenza virus and its subtypes or lineages in different age groups and sex,taking into account the short-term harvesting effects of influenza.Objectives:The objectives of this study included describing influenza virus prevalence trends in Guangzhou from 2015 to 2018 based on appropriate proxy representing influenza activity,estimating the death burden of influenza in the whole population and in different age or sex subgroups,and analyzing the time course of the impact of influenza on mortality.Methods:Weekly influenza-like illness surveillance data and influenza virus etiology surveillance data from 2011 to 2018 were obtained from Guangzhou Center for Disease Control and Prevention,and daily mortality surveillance data were collected during the same period.The daily meteorological data was downloaded from China Meteorological Data Network.Excess pneumonia and influenza mortality were calculated based on the Serfling model to determine the best proxy for influenza activity.Quasi-Poisson regression model with Distributed lagged non-linear model(DLNM)was adopted.After controlling for long-term trends in mortality and effects of seasonality,temperature,absolute humidity,holiday and day of the week,the association of influenza activity proxies with population mortality and estimate expected mortality under actual influenza epidemic and no influenza epidemic(counterfactual)scenarios,which noting for E(Y|flu=fluobs)and E(Y|flu=0),was explored.The influenza-associated excess mortality was calculated as E(Y|flu=fluobs)-E(Y|flu=0).Results:It was found that influenza virus activity was best represented by multiplying the proportion of influenza-like illness with the positive rate of laboratory tests for different influenza subtypes/lineages.Based on this proxy,we found that influenza was prevalent throughout the year in Guangzhou,and the activity of different influenza subtypes or lineages was different in several years.The annual excess mortality rates for all causes,respiratory or cardiovascular disease,respiratory disease,pneumonia or influenza related to influenza,were 25.06(95%empirical confidence interval[eCI]19.85-30.16),20.36(95%eCI 16.75-23.74),9.41(95%eCI 7.78-10.90)and 3.88(95%eCI 2.80-4.84)per 100,000 persons,respectively.Influenza-related excess respiratory or cardiovascular deaths in people aged 60～79 years and>80 years accounted for 32.9%and 63.7%of excess death for corresponding reason associated with influenza in the whole population.The ratio of influenza-related excess mortality from respiratory diseases in men and excess mortality from the same cause in women was 1.34(95%CI 1.17-1.54)and that from cardiovascular diseases was 0.72(95%CI 0.63-0.82).The annual excess respiratory or cardiovascular diseases mortality rates associated with A(H3N2),B/Yamagata,B/Victoria and A(H1N1)were 8.47(95%eCI 6.60-10.30),5.81(95%eCI 3.35-8.25)and 3.68(95%eCI 0.81-6.49)and 2.83(95%eCI-1.26-6.71),respectively.Among these influenza subtypes or lineages,A(H3N2)had the highest excess mortality in people aged 60～79 years and>80 years,with 20.22(95%eCI 14.56-25.63)and 180.15(95%eCI 130.75-227.38),respectively.People aged<60 years were more affected by A(H1N1)than other influenza subtype or lineages,with an excess mortality rate of 1.29(95%eCI 0.07-2.32).The mortality displacement time of the A(H1N1),A(H3N2)and B/Yamagata was 2 to 5 days,while the mortality displacement time of the B/Victoria was 5 to 13 days.Conclusions:In Guangzhou,seasonal epidemics of influenza caused a certain number of deaths,mainly among the elderly.The burden of death caused by influenza B,especially B/Yamagata,should not be negligible.Reverse sex differences were found in excess mortality from influenza-related respiratory disease and cardiovascular disease,and the underlying mechanisms need to be explored in further studies.The findings of this study can help to understand the time-varying characteristics of the impact of influenza seasonal epidemic on human mortality,provide necessary scientific evidence for the formulation of influenza prevention and control measures,and provide methodological reference for the accurate estimation of influenza mortality burden. |
PDF Full Text Request