If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Emergency Medicine, Second Affiliated Hospital, Department of Public Health, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
Department of Emergency Medicine, Second Affiliated Hospital, Department of Public Health, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
Department of Emergency Medicine, Second Affiliated Hospital, Department of Public Health, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
Department of Emergency Medicine, Second Affiliated Hospital, Department of Public Health, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
Department of Emergency Medicine, Second Affiliated Hospital, Department of Public Health, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
Department of Emergency Medicine, Second Affiliated Hospital, Department of Public Health, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China
Global influenza virus circulation was characterized by seven influenza transmission zones (ITZs).
•
Influenza surveillance should be enhanced in Africa and Southern-America ITZs.
•
More attention should be given to Eastern & Southern-Asia and Africa ITZs.
Abstract
Objectives
This study aimed to investigate region-specific epidemiologic characteristics of influenza and influenza transmission zones (ITZs).
Methods
Weekly influenza surveillance data of 156 countries from 1996 to 2021 were obtained using FluNet. Joinpoint regression was used to describe global influenza virus trends, and clustering analyses were used to classify the ITZs.
Results
The global median average positive rate for total influenza virus was 16.19% (interquartile range: 11.62-25.70%). Overall, three major subtypes (influenza H1, H3, and B viruses) showed alternating epidemics. Notably, the proportion of influenza B viruses increased significantly from July 2020 to June 2021, reaching 62.66%. The primary peaks of influenza virus circulation in the north were earlier than those in the south. Global influenza virus circulation was significantly characterized by seven ITZs, including "Northern America" (primary peak: week 10), "Eastern & Southern-Asia" (primary peak: week 10), "Europe" (primary peak: week 11), "Asia-Europe" (primary peak: week 12), "Southern-America" (primary peak: week 30), "Oceania-Melanesia-Polynesia" (primary peak: week 39), and "Africa" (primary peak: week 46).
Conclusion
Global influenza virus circulation was significantly characterized by seven ITZs that could be applied to influenza surveillance and warning.
], and its seasonal epidemics infect approximately 5-10% of adults and 20-30% of children, as well as cause 290,000-650,000 deaths worldwide every year [
]. According to the countries, areas, or territories with similar influenza transmission patterns, the World Health Organization (WHO) established 18 influenza transmission zones (ITZs) globally in 2009 [
], which are useful for the surveillance of influenza virus circulation and the selection of vaccine candidate strains. The classification of ITZs, impeded by the limitations of fighting against influenza only at the national level, can help to develop global strategies for the prevention and control of influenza [
]. However, based on the 18 ITZs, there are still many countries or regions from geographically adjacent ITZs with the same epidemiological characteristics as influenza, such as influenza seasonality, peaks, and spatiotemporal patterns [
]. The study by Caini et al. in 2017 further improved the ITZ system, with the reclassification of two ITZs in the WHO European Region, which were divided into five ITZs by the WHO [
]. Globally, there is no universal partitioning of ITZs that can reflect common influenza characteristics to better pursue and implement global policies for fighting against influenza.
In this study, we aimed to investigate the region-specific epidemiologic characteristics of influenza and explore the universal ITZs which reflect common influenza characteristics and contribute to the implementation of global policies.
Methods
Data sources
Global influenza virus surveillance data
Global influenza surveillance data were obtained from FluNet (https://www.who.int/tools/flunet). In this study, weekly influenza surveillance data from 156 countries for the period between week 27 in 1996 and week 26 in 2021 were available on 12 November 2021. The following data were extracted: positive numbers of A/H1, A/H3, A/H5, and unsubtyped influenza A viruses; positive number of influenza B viruses; positive number of total influenza viruses; and number of samples processed.
The definitions of indexes used in this study
The influenza season was defined as the period from 1 July (week 27) of 1 year to 30 June (week 26) of the next year [
]. The positive rates of A/H1, A/H3, B, and total influenza viruses in each country and for each influenza season were calculated by dividing the positive numbers of influenza viruses (A/H1, A/H3, B, and total influenza virus) by the total number of samples processed. If the total number of samples processed was not accessed or fewer than 100 samples were collected in one influenza season in one country, the data of this country in this influenza season were excluded.
] to examine the global trends in positive rates for A/H1, A/H3, B, and total influenza viruses from July 1996 (week 27) to June 2021 (week 26) influenza seasons. Annual percentage changes (APCs) with 95% confidence intervals (CIs) were calculated. If the APC was not significant (P≥0.05), we regarded trends as stable; otherwise, they increased or decreased [
Epidemiological features of and changes in incidence of infectious diseases in China in the first decade after the SARS outbreak: an observational trend study.
The classification of influenza transmission zones
To obtain complete data from more countries and exclude the impact of the 2009 influenza pandemic, data from July 2010 to June 2019 were included in the ITZ analysis. For each included country, we calculated the timing of the primary peaks of influenza viruses (total influenza virus, and influenza H1, H3, and B viruses). For each country, we detrended the time series of influenza data with a quadratic polynomial, and the periodic annual function (PAF) was generated by summing the annual, semi-annual, and quarterly harmonics obtained by Fourier decomposition [
]. The timing of the primary peaks of the PAF of each included country was extracted, which indicated the period when the maximum intensity of disease burden usually occurs [
]. The week of the start and peak of the influenza season was further calculated for total influenza virus and influenza H1, H3, and B viruses in each country using the weekly average of the positive number of influenza viruses. The start week of the influenza season was defined retrospectively as the 1st week in which the cumulative positive number of influenza viruses reached at least 10% of the total positive number in the whole influenza season, while the peak week of the influenza season was defined retrospectively as the week with the highest positive number of influenza viruses among the whole influenza season [
]. Countries with missing data on the weekly average of the positive number of influenza viruses were excluded. The “timing” parameters (primary peak, start and peak week, for total influenza virus, influenza H1, H3, and B viruses), “virus” parameters (positive rate for total influenza virus, influenza H1, H3, and B viruses), and location information (longitude and latitude of capital) for 109 countries were included to perform cluster analysis. We applied K-means clustering based on the Hartigan–Wong algorithm [
] to identify the regional clusters. The “elbow method”, which computed the sum of squared errors within clusters (SSEWC), was used to select the appropriate number of clusters [
]. To evaluate the stability of K-means clustering, three hierarchical clustering approaches (Ward's minimum variance method, complete linkage, and average linkage) relying on Euclidean distance [
] were also performed for 109 countries. The consistency of the results of the three hierarchical clustering and K-means clustering methods was calculated (consistency represents the number of countries falling into the same cluster based on the four methods divided by the total number of countries).
We used R (version 3.6.1), Joinpoint (version 4.8.0.1), and EPIPOI [
Geographical distribution and primary peaks of influenza virus circulation
In our study, global influenza surveillance data, including 39,637,339 samples from week 27 of 1996 to week 26 of 2021. The median average positive rate of total influenza virus in the 156 countries was 16.19% (interquartile range:11.62-25.70%) (Supplementary Material 1). The highest average positive rates of total influenza virus were observed in Central America and the Caribbean (22.38%), Northern Africa (20.10%), South-West Europe (19.35%), and South-East Asia (18.50%), whereas the lowest average positive rate of total influenza virus was found in tropical South America (4.35%) (Figure 1a).
Figure 1Geographical distribution and primary peaks of influenza virus. (a) The average positive rate of total influenza virus; (b) The primary peaks of influenza virus.
The time of the primary peaks in each country was significantly negatively correlated with the latitude gradient. The primary peaks of the countries in the north were earlier than those in the south (Figure 1b). The correlation was strongest for the influenza B virus (R = -0.53; P < 0.001), followed by the total influenza virus (R = -0.44; P < 0.001), A/H1 (R = -0.37; P < 0.001), and A/H3 (R = -0.40; P < 0.001) viruses (Supplementary Materials 2-4). A non-significant correlation between the primary peaks and the longitudinal gradient was found (Supplementary Material 5).
Global trends of influenza virus from July 1996 to June 2021 influenza seasons
From July 1996 to June 2016, the positive rate of the total influenza virus was stable (APC = 1.46%, 95% CI: -0.61-3.56%; P = 0.17 > 0.05). From 2016/07 to 2019/06, it slight but not significant increase (APC = 18.85%, 95% CI: -40.65 to 137.97%; P = 0.61 > 0.05). Between July 2019 and June 2021, the positive rate of total influenza virus significantly decreased (APC = -87.06%, 95% CI: -93.54% to -74.08%; P <0.01) (Figure 2a). The average positive rate of total influenza virus significantly decreased from 16.66% (6.23-49.40%) during July 1996 to June 2020 influenza season to 0.23% (0-16.58%) during July 2020 to June 2021 influenza season (P <0.01). The positive rate of A/H1 from July 1996 to June 2010 showed a significant increase (APC = 42.86%, 95% CI:18.50-72.23%; P <0.01). In the periods from July 2010 to June 2019 and July 2019 to June 2021, the positive rate of A/H1 decreased but was not statistically significant, with APC = -8.02% (95% CI: -37.67 to 35.74%; P = 0.66 > 0.05) and APC = -89.62% (95% CI: -99.71 to 267.50%; P = 0.20 > 0.05), respectively (Figure 2b). The positive rate of A/H3 significantly increased from July 2001 to June 2019 (APC = 7.04%, 95% CI:1.78-12.57%; P = 0.01) and significantly decreased after July 2019 (APC = -88.37%, 95% CI: -97.24 to -50.95%; P <0.01) (Figure 2c). Overall, the positive rate of influenza B virus was relatively stable, although two turning points were found in June 2000 and June 2018 (all P-values [0.59, 0.07, and 0.17] > 0.05) (Figure 2d).
Figure 2Global trends of positive rate of influenza virus from July 1996 to June 2021 influenza seasons. (a) Global trends of positive rate of total influenza virus; (b) Global trends of positive rate of influenza H1 virus; (c) Global trends of positive rate of influenza H3 virus; (d) Global trends of positive rate of influenza B virus. The orange solid dot represents the observed positive rate for each influenza seasons. The green empty dot circle represents the turning point in join-point regression.
Spatiotemporal characteristics and transmission zones of influenza
We conducted clustering analyses that included 3,546,236 influenza cases in 109 countries from July 2010 to June 2019. The elbow method suggests that 109 countries can be divided into at least four ITZs where the SSEWC was significantly reduced (Supplementary Material 6). After combining the results of the K-mean and the three hierarchical clustering algorithms (Supplementary Materials 7-13), we divided the 109 countries into seven ITZs according to the characteristics of influenza virus circulation and geographic location of the country (Figure 3). The average positive rate of total influenza virus was highest in the Europe ITZ (16.46%) and lowest in the Southern-America ITZ (5.58%). The primary peak was earliest in the Northern-America ITZ (week 10) and Eastern & Southern-Asia (week 10), followed by Europe (week 11), Asia-Europe (week 12), Southern-America (week 30), Oceania-Melanesia-Polynesia (week 39), and Africa ITZs (week 46) (Figure 3). However, the seasonality of the winter epidemics became less pronounced among some countries closer to the equator, some showing even two peaks or were sustained with low intensity year-round in both the Eastern & Southern-Asia and Africa ITZs (Figure 4).
Figure 3Geographical distribution of seven global ITZs. The geographical distribution, primary peaks, and average positive rate of seven ITZs.
Figure 4The heatmap of seven ITZs from July 2010 to June 2019. Color bar represents the intensity of influenza incidence, from high (orange) to low (white). Weekly incidence counts were standardized for each influenza season and shown as the proportion of the maximum number of cases in a week for that country and period (hence, weeks with the maximum number of cases for a given influenza season were assigned the value 1).
Dynamic pathogen spectrum of influenza virus from the July 1996 to June 2021 influenza season
Figure 5 shows that the three major subtypes (influenza H1, H3, and B viruses) showed an alternating epidemic during the study period. Deducted from influenza A virus (unsubtyped) due to limited data, the overall proportion of A/H1 accounted for 25.50% of the subtype spectrum of influenza virus, followed by influenza B virus (24.72%) and A/H3 (19.65%). Notably, the proportion of A/H1 increased significantly from July 2009 to June 2010, reaching 79.53%, and the proportion of influenza B viruses increased significantly from July 2020 to June 2021, reaching 62.66%.
Figure 5Dynamic pathogen spectrum of influenza virus from July 1996 to June 2021 influenza seasons. (a) Global dynamic pathogen spectrum of influenza virus; (b) Dynamic pathogen spectrum of influenza virus in Northern-America ITZ; (c) Dynamic pathogen spectrum of influenza virus in Eastern & Southern-Asia ITZ; (d) Dynamic pathogen spectrum of influenza virus in Europe ITZ; (e) Dynamic pathogen spectrum of influenza virus in Asia-Europe ITZ; (f) Dynamic pathogen spectrum of influenza virus in Southern-America ITZ; (g) Dynamic pathogen spectrum of influenza virus in Oceania-Melanesia-Polynesia ITZ; (h) Dynamic pathogen spectrum of influenza virus in Africa ITZ. The proportion of A/H5 was extremely small, therefore we did not show it.
In this study, we first performed a comprehensive analysis of global influenza surveillance data including 39,637,339 samples from 1996 to 2021 in 156 countries and further investigated the global trends of influenza virus circulation and explored universal ITZs. The median average positive rate of total influenza virus in 156 countries was 16.19%. We found that the average positive rates of influenza virus were highest in the regions of Central America and the Caribbean, Northern Africa, South-West Europe, and South-East Asia. We found a substantial increase in the proportion of influenza A/H1 during the A/H1N1(pdm09) and influenza B viruses during the COVID-19 pandemics , respectively. It could have been viral interference with the complexity of these viruses, and when a new virus emerges it may affect other viruses. Simultaneously, the implementation of various nonpharmacological interventions is usually strengthened during the pandemic, which might also affect the prevalence of some other specific strains. In future studies, we need to further explore the causes of strain changes in specific periods.
The global patterns of influenza virus circulation, including the positive rates, primary peaks, and dominant subtypes of the virus, vary significantly in different regions around the world. To elucidate the global epidemiologic characteristics of the influenza virus, the WHO established 18 ITZs globally in 2009 (Supplementary Material 14) [
]. Based on global influenza surveillance data, we further reclassified these ITZs. Our findings indicated that countries among the same ITZs had similar characteristics of influenza seasonality, peaks, and spatiotemporal patterns (Figure 3 and Supplementary Materials 15-16). The patterns of influenza virus circulation in each ITZ indicated in this study would be helpful for implementing a preparedness plan against an influenza epidemic within the ITZ, such as surveillance for circulating strains, assessments of severity and healthcare resources, and vaccine distribution and deployment in the ITZ [
Among the seven ITZs, the primary peaks of the four ITZs in the Northern Hemisphere occurred earlier than those of the three ITZs in the Southern Hemisphere. At the same time, we found that the primary peak of the Northern-America ITZ occurred earlier than that of the Europe ITZ. In future research, more dimensional data, such as epidemiological data and pathogenic genome sequences, should be obtained to verify this phenomenon and clarify its underlying reasons. Epidemic influenza usually lasts for a mean of four months [
], after which influenza virus circulation weakens and even vanishes during the rest of the season. These various patterns have been speculated to be linked to climatological diversity [
]. Another factor for the unsustainable activities of the influenza virus throughout the year is speculated to be linked with sufficient herd immunity after the epidemic [
]. Therefore, we hypothesized that influenza viruses may travel from one region to another annually in climate-dependent and herd-susceptibility-driven patterns.
However, the seasonality of winter epidemics became less pronounced in some regions closer to the equator with complex climatic features [
]. This result is consistent with our findings that the influenza virus circulation had two peaks and was sustained with low intensity year-round in both the Eastern & Southern-Asia and Africa ITZs (Figure 4), although relative primary peaks of weeks 10 and 46 were identified (Figure 3). Sustained year-round influenza virus circulation among these ITZs will provide potential opportunities for supplying the annually circulating influenza virus strain to the globe, including regions with a typical seasonal pattern. Thus, we supposed that the Eastern & Southern-Asia ITZ is not necessarily the only source region for the annually circulating influenza virus strain, as the Africa ITZ might be a tropical exchange zone of influenza viruses between hemispheres. A previous hypothesis on the global circulation of the influenza virus suggested that annual epidemic waves may propagate from Asia, where influenza activity frequently occurs year-round [
], but due to the high frequency of influenza virus mutations, the protective effect of influenza vaccination is only 37% (32-42%) in the Northern Hemisphere and 54% (48-59%) in the Southern Hemisphere [
Variable seasonal influenza vaccine effectiveness across geographical regions, age groups and levels of vaccine antigenic similarity with circulating virus strains: a systematic review and meta-analysis of the evidence from test-negative design studies after the 2009/10 influenza pandemic.
]. The WHO Global Influenza Surveillance Network (GISRS) was established in 1952 to provide early warning of changes in influenza virus circulating in the global population and evaluate the efficacy of seasonal influenza vaccines [
]. They are mainly committed to providing scientific and timely information on the antigenic characteristics of influenza viruses and the selection of suitable vaccine viruses [
]. Given the different characteristics of influenza virus circulation in the seven ITZs indicated in our study, we recommend that influenza surveillance be enhanced in Africa and Southern-America ITZs.
Limitation
Our study had several limitations. First, in some large countries, the spatiotemporal patterns of influenza epidemics may be more complex which could “be an ITZ on its own” or even “encompass more than one ITZ”. However, the influenza surveillance data currently entered into the FluNet database are from individual countries, and there are no regional data within one country. As global influenza virus surveillance ability improves and many regional data are more available, updating the definition of ITZs may be required. Second, the test volume, sample source, and testing abilities varied among the countries, which may have an impact on the positive rate of influenza virus.
Conclusion
First, global influenza virus circulation was significantly characterized by seven ITZs, which could be applied to influenza surveillance and warning. We recommend that influenza surveillance be enhanced in the Africa and Southern-America ITZs. Influenza virus circulation is sustained year-round in the Eastern & Southern-Asia and Africa ITZs, which warrants more attention and health prevention services.
Declarations of competing interest
The authors have no competing interests to declare.
Funding
This study was supported by grants from the National Natural Science Foundation of China (grant number: 82173577, 81672005, U1611264, 81001271), the Mega-Project of National Science and Technology for the 12th and 13th Five-Year Plan of China (grant number: 2018ZX10715-014-002 and 2014ZX10004008).
Ethics approval and consent to participate
Not applicable.
Author contributions
SY designed the study. CC, DJ, DY, XZ, YD, XL, MY and YZ collected data. SY, CC and LP analyzed data. CC, CD and LL checked the data and results. SY and CC interpreted data and wrote the report. SY and CC revised the report from preliminary draft to submission. All authors have read and approved the manuscript.
Epidemiological features of and changes in incidence of infectious diseases in China in the first decade after the SARS outbreak: an observational trend study.
Variable seasonal influenza vaccine effectiveness across geographical regions, age groups and levels of vaccine antigenic similarity with circulating virus strains: a systematic review and meta-analysis of the evidence from test-negative design studies after the 2009/10 influenza pandemic.
00052-8&id=gr1.jpg){kind=link}
00052-8&id=gr2.jpg){kind=link}
00052-8&id=gr3.jpg){kind=link}
00052-8&id=gr4.jpg){kind=link}
00052-8&id=gr5.jpg){kind=link}