two authors (CEGS and JP) but did not have any additional role in the data collection, analysis, We describe the epidemiological characteristics, pattern of circulation, and geographical decision to publish, or preparation of the manuscript. The specific roles of these authors are distribution of influenza B viruses and its lineages using data from the Global Influenza B articulated in the “author contribution section”. Study. We included over 1.8 million influenza cases occurred in thirty-one countries during None of the surveillance schemes in the 2000–2018. We calculated the proportion of cases caused by influenza B and its lineages; participating countries was supported by this grant. determined the timing of influenza A and B epidemics; compared the age distribution of B/ Competing interests: Clotilde El-Guerche Se ´blain is Victoria and B/Yamagata cases; and evaluated the frequency of lineage-level mismatch for an employee of Sanofi Pasteur. She was the the trivalent vaccine. The median proportion of influenza cases caused by influenza B virus coordinator of the research project at Sanofi Pasteur, she helped define the study objectives, was 23.4%, with a tendency (borderline statistical significance, p = 0.060) to be higher in and critically revised the manuscript. When tropical vs. temperate countries. Influenza B was the dominant virus type in about one every reviewing the manuscript, the revisions concerned seven seasons. In temperate countries, influenza B epidemics occurred on average three the epidemiological findings of the study and not weeks later than influenza A epidemics; no consistent pattern emerged in the tropics. The the public health findings or conclusions. This does not alter our adherence to PLOS ONE policies on two B lineages caused a comparable proportion of influenza B cases globally, however the sharing data and materials. Cheryl Cohen has B/Yamagata was more frequent in temperate countries, and the B/Victoria in the tropics (p = received grant support from Sanofi Pasteur, 0.048). B/Yamagata patients were significantly older than B/Victoria patients in almost all Advanced Vaccine Initiative, US Centers for Disease Control and Prevention, and payment of countries. A lineage-level vaccine mismatch was observed in over 40% of seasons in tem- travel costs from Parexel. All of the other authors perate countries and in 30% of seasons in the tropics. The type B virus caused a substantial declare that they have no conflict of interests to st proportion of influenza infections globally in the 21 century, and its two virus lineages dif- disclose. fered in terms of age and geographical distribution of patients. These findings will help inform health policy decisions aiming to reduce disease burden associated with seasonal influenza. Introduction Influenza causes a major burden of disease on populations globally . The impact of seasonal and pandemic influenza on population health in high-income countries has been described extensively . In Europe, influenza ranks first among infectious diseases in terms of burden, accounting for 30% of the total disability-adjusted life years (DALYs) lost due to infectious dis- eases annually . More recently, evidence has been accumulating that the burden of disease associated with influenza is high in low- and middle-income countries as well . Type A viruses cause most influenza cases and are responsible for pandemics, but influenza B is an important cause of morbidity and mortality during interpandemic periods, and its prevention represents an important public health priority globally [5–6]. Influenza B viruses split into two antigenically distinct phylogenetic lineages (B/Victoria/2/87 representative, abbreviated B/Vic- toria, and B/Yamagata/16/88 representative, abbreviated B/Yamagata) in the early 1980s . Whilst the circulation of the Victoria lineage was geographically limited to eastern Asia for st most of the 1990s, the two lineages have co-circulated globally in the 21 century [8–9]. The mainstay of influenza prevention is vaccination, aimed primarily at individuals who are at greater risk of developing complications when infected, like the elderly, pregnant PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 2 / 17 Epidemiology of influenza B virus and its lineages women, and people with underlying medical conditions, and more recently at children to pro- vide both direct protection to the children and indirect protection to the wider population . Both the composition and period of administration of the vaccine are crucial to ensure adequate immunity against influenza; thus, investigating the timing of influenza epidemics and the patterns of circulation of the different influenza virus types, subtypes and lineages is of great importance from a public health perspective . Until 2012, trivalent influenza vaccines (TIV) were available that contained only one type B influenza virus, belonging to either the B/ Victoria or B/Yamagata lineage. Each year, the WHO issued recommendations for the B virus strain to be included in the TIV, but the frequency of lineage-level vaccine mismatch has been high (around 50%) . This often resulted in unsatisfactory protection, considering the limited efficacy against the mismatched lineage , and substantial health impact . Quadrivalent influenza vaccine (QIV), containing one B/Victoria and one B/Yamagata virus strain, were first approved by the US Food and Drug Administration (US-FDA) in 2012. This prompted the establishment of several studies that aimed to describe the epidemiology, clinical character- istics and burden of disease of influenza B in more depth, and to compare the cost-effective- ness of QIV vs. TIV in different settings and populations. Despite significant advances in our understanding of influenza B , important knowl- edge gaps persist, in particular concerning the pattern of circulation and geographical distribu- tion of B/Victoria and B/Yamagata viruses, and the age distribution of patients who are most often infected with either lineage. These aspects are of great importance, however, for instance in order to evaluate the cost-effectiveness of alternative influenza prevention strategies and thus enable more appropriate public health choices (e.g. around the choice of tri- and quadri- valent vaccines) . In fact, while some reports have addressed this topic in single countries [16–22], a global overview is still lacking. Here, we conducted a global analysis of the epidemi- ology of influenza B virus and its lineages using the database of the Global Influenza B Study (GIBS). Methods The Global Influenza B Study The rationale and methodology of the Global Influenza B Study has been described in detail elsewhere [14,23–24]. The GIBS database encompasses epidemiological and virological influ- enza surveillance data from thirty-one countries around the world (Fig 1). Participating coun- tries were classified as located in the Northern hemisphere (n = 11), inter-tropical belt (n = 15), or Southern hemisphere (n = 5) based on the latitude of the country’s population cen- tre . Participating countries were required to make available influenza surveillance data for as many consecutive years as possible (from 2000 onwards). Data include the weekly number of laboratory-confirmed influenza cases broken down by virus type (A, B), type A subtype (H1N1, 2009 pandemic H1N1, H3N2, A not subtyped), and type B lineage (Victoria, Yama- gata, B not characterized) and weekly influenza-like illness/acute respiratory infection rates (per 100,000 population or 100 consultations, depending on country). Information on age (exact age or age groups) was also available. In most participating countries, the influenza sur- veillance system covers the whole national territory, and a sample of influenza virus positive specimens or virus isolates is routinely sent to a WHO collaborating centre for further investi- gations. However, countries differ in other important aspects of their surveillance system, e.g. the clinical case definition in use, the distribution of outpatients and inpatients being sampled, and the availability of population denominators [14,24]. The GIBS database was initially assembled in 2013–14 by merging influenza surveillance data up to as late as December 2013 from thirty countries , but was updated for the purposes of this paper with data until 2015 PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 3 / 17 Epidemiology of influenza B virus and its lineages Fig 1. Countries participating in the Global Influenza B Study (GIBS), 2000–2018. https://doi.org/10.1371/journal.pone.0222381.g001 to 2018 (depending on the country) for twenty-five countries. The Netherlands joined the project in 2018, bringing the total number of participating countries to thirty-one. Data for China were provided separately for the Northern and Southern parts of the coun- try, which were therefore entered separately in the analyses (for brevity, however, we will us the term “country” to refer to a whole country or a part of it henceforth). Brazil also provided data stratified by sub-national regions (i.e. five administrative regions: north, north-east, cen- tral-west, south, and south-east); however, geographically-stratified data were too sparse in some years, thus we opted to treat Brazil as a single country in the analysis. Statistical methods Similar to previous GIBS publications [14,23–24], the unit of analysis was the “season”: this corresponded to the calendar year in tropical countries and countries located in the Southern th th hemisphere, and was defined as the period between the 27 week of a year and the 26 week of the following year for countries located in the Northern hemisphere. As previously explained , the purpose of this methodological approach is to give each “season” an equal weighting in the analysis, thus limiting the impact on results arising from any differences in reporting between countries (e.g. high- vs. low-resources countries) and over time within the same country (e.g. before vs. after the 2009 pandemic). For each country, we included in the analyses the seasons with at least 20 weeks of reported data and at least 100 laboratory-confirmed influenza cases. In each season, we calculated the percentage of influenza cases that were due to influenza B. We then determined the percentage of cases that were caused by viruses belonging to the B/Victoria or B/Yamagata lineage: in this analysis, to increase the stability of the results, we only included (for each country) the subset of seasons with 50 or more characterized influenza B viruses. The non-parametric Kruskal- PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 4 / 17 Epidemiology of influenza B virus and its lineages Wallis test was applied to compare the median proportion of influenza A and B cases (over all influenza cases) and B/Victoria and B/Yamagata cases (over all cases with characterized influ- enza B viruses) between countries based on their latitude (Northern hemisphere, inter-tropical belt, or Southern hemisphere) and population age structure (median age below 30 years, between 30 and 35 years, or above 35 years) . We compared the timing of the primary peak of influenza A and B epidemics in each coun- try using the EPIPOI software . For this analysis, we excluded data from the 2009 season (2009–2010 in Northern hemisphere countries) because of the markedly atypical timing of the influenza A(H1N1)pdm09 pandemics. Time series were first standardized by dividing the weekly number of influenza cases by the maximum weekly number (per country by defined season). Next, a periodic annual function (PAF) was generated by summing up the annual, semi-annual and quarterly harmonics (obtained by Fourier decomposition). The timing of the primary peak (defined as the month where the PAF reaches its maximum value) was then compared for A and B epidemics in each country. We aimed to evaluated the frequency of influenza B lineage-level vaccine mismatch, which was defined as a mismatch between the lineage that caused the majority (>50%) of influenza B cases in a given season and country, and the lineage included in the trivalent vaccine. Informa- tion on influenza vaccine formulation was retrieved from the WHO website . The above definition is straightforward for temperate countries in the Northern and Southern hemi- sphere, but much less so for countries of the inter-tropical belt. While the WHO issues recom- mendations on the composition of influenza vaccines to be used in tropical countries , these countries often adopt either formulation based on local considerations or make no spe- cific recommendations . To cope with this and consider all possible scenarios (and also for consistency with previous GIBS publications ), the frequency of B lineage-level vaccine mismatch in tropical countries was calculated by assuming that: (i) all countries situated north of the equator use the northern hemisphere vaccine, and all countries south of the equator use the southern hemisphere vaccine formulation; (ii) all countries use the northern hemisphere vaccine formulation; or (iii) all countries use the southern hemisphere vaccine formulation. The recent update of the GIBS database and the resulting increase in the number of cases with characterized influenza B viruses made it possible to compare the age distribution of influenza B/Victoria and B/Yamagata cases in a similar way to what had been done previously for influenza B vs. A (and its subtypes) [14,24]. This analysis was limited to countries for which information on the exact age was available for at least 50 B/Victoria and at least 50 B/ Yamagata influenza cases over all seasons available. The virus-lineage-specific age distribution of influenza B cases was visualized, separately in each country, using histograms with 5-age- year-wide bars. Since the distributions were skewed to the right, we used the non-parametric Wilcoxon rank sum test to compare the median age of influenza B cases by virus lineage. Statistical analyses were conducted using Stata software version 14 (StataCorp LP, College Station, TX, USA). All statistical tests were two-sided, and p-values were considered statisti- cally significant when below 0.05. Results The database of the Global Influenza B Study encompassed a total of 1.820.301 influenza cases between 2000 and 2018, of which 419,167 (23.0%) had type B influenza. Cases were unevenly distributed between countries, with USA and Australia contributing 54.1% and 25.3% of all influenza cases to the database, respectively. Information on age was not available for the USA; for the remaining countries, information on age was available for most (95.0%) influenza cases. After applying our exclusion criteria, 299 seasons with 100 or more reported influenza PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 5 / 17 Epidemiology of influenza B virus and its lineages cases overall were included in the analysis, of which 110 were from Northern hemisphere countries, 131 from countries located in the inter-tropical belt, and 58 from Southern hemi- sphere countries. The median number of seasons per country was nine, ranging from three (for Turkey) to eighteen (for New Zealand). The number of overall and influenza B cases in each country and season included in the analysis, the proportion of influenza B cases of which the virus was characterized, and their breakdown into B/Victoria and B/Yamagata lineages, are provided in S1 Table. Frequency of influenza B Influenza B virus caused a median 23.4% (interquartile range [IQR] 9.3–38.9%) of reported influenza cases in a season. More specifically, the proportion of influenza cases caused by the B virus type was <20% in 118 seasons (42.3%), between 20% and 50% in 125 seasons (41.8%), and above 50% in 45 seasons (15.1%). The proportion of influenza B cases over all influenza cases reported in the season varied geographically (Fig 2): its median value tended to be higher (with borderline statistical significance, p = 0.060) in countries of the inter-tropical belt (27.4%, IQR 12.2–41.7%) compared to temperate countries of the Northern (21.0%, IQR 7.3– 37.4%) and Southern (22.2%, IQR 9.1–34.5%) hemispheres. The proportion of seasons in which influenza B cases constituted between 20 and 50% or more than 50% of all reported influenza cases did not significantly differ (p = 0.275) between tropical countries (47.3% and 16.0%, respectively) and temperate countries of the two hemispheres (39.9% and 14.3%, respectively). Timing of influenza A and B epidemics In countries of the Southern hemisphere, influenza A epidemics typically peaked in July-Sep- tember, on average 1.1 month earlier than influenza B (August-September) (Table 1). Influ- enza epidemics also peaked in the winter months of Northern hemisphere countries, earlier for influenza A (January-February, except Ukraine, in March) than for influenza B (February- March, except England, in January), with an average difference of 0.6 months. The only excep- tion was Bhutan (i.e. the southernmost countries among those situated in the Northern hemi- sphere), where both influenza A and B epidemics typically peaked in August. There was more variability in countries of the inter-tropical belt: the timing of the primary peak varied widely between countries and could take place in practically any month of the year, with a small dif- ference in timing (0.3 months earlier on average) for the primary peak of influenza A and B epidemics. Frequency of B/Victoria and B/Yamagata lineages viruses There were 84 seasons where the virus lineage was determined for 50 or more influenza B cases (out of a total 299 seasons, 28.1%) and this data covered 18 countries (out of 31). In these seasons, the B/Victoria and B/Yamagata lineage viruses caused a similar range of proportions of influenza B cases. More precisely, the median proportion of influenza B cases reported in a season that were caused by B/Victoria lineage viruses was 46.0% (IQR 5.3–89.6%). The range of the proportion of B/Victoria over all influenza B cases was 0%-100%, meaning that there were seasons in which all influenza B cases were caused by either virus lineage. Co-circulation of the two lineages was frequent: in 27 (32.1%) of 84 seasons, both lineages accounted for at least 20% of influenza B cases, while the B/Victoria and B/Yamagata lineages caused over 80% of all influenza B cases in 27 (32.1%) and 30 (35.8%) seasons, respectively. The proportion of either lineage over all influenza B cases varied according to the age struc- ture of countries, and consequently, according to the countries’ latitude as well (Fig 3). The PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 6 / 17 Epidemiology of influenza B virus and its lineages Fig 2. Distribution of influenza seasons by proportion of influenza B cases and geographical area (Inter-tropical th th th belt, Northern hemisphere, Southern hemisphere). Red bars indicate 25 , 50 (median) and 75 percentiles. The Global Influenza B Study, 2000–2018. https://doi.org/10.1371/journal.pone.0222381.g002 proportion of influenza B cases caused by the B/Victoria lineage in a given season varied widely (ranging from 0% to nearly 100%) also in the analyses stratified by the country’s age structure and latitude. However, the B/Victoria lineage was more frequent (p-value 0.049) in countries with a median age below 30 years (median proportion over all influenza B cases = 67.0%) compared to countries with a median age was 30–35 years (42.0%) or above 35 years (36.2%). Likewise, the median proportion of B/Victoria over all characterized influenza B cases in a season was higher in countries of the inter-tropical belt (57.5%) compared to tem- perate countries of the Northern (27.9%) and the Southern hemisphere (43.8%) (p- value = 0.048). Age distribution of B/Victoria and B/Yamagata influenza cases In Table 2, we compared the median age of influenza B cases according to virus lineage. In Australia, B/Victoria cases were non-significantly older than B/Yamagata cases: the median PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 7 / 17 Epidemiology of influenza B virus and its lineages Table 1. Typical timing of the peak of influenza A and B epidemics according to countries’ latitude. The Global Influenza B Study, 2000–2018. Country Latitude Typical timing of influenza peak A B Southern hemisphere st nd New Zealand 41.8 S Aug (1 half) Aug (2 half) nd nd Chile 35.8 S Jul (2 half) Sep (2 half) st nd Argentina (Santa Fe) 31.4 S Aug (1 half) Sep (2 half) st nd South Africa 29.0 S Jul (1 half) Sep (2 half) st st Australia 25.8 S Sep (1 half) Sep (1 half) Inter-tropical belt st st Madagascar 19.4 S Feb (1 half) Mar (1 half) nd st Brazil 10.8 S Jun (2 half) Oct (1 half) st st Ecuador 2.0 S May (1 half) Jul (1 half) st nd Indonesia 1.7 S Feb (1 half) Apr (2 half) nd nd Kenya 0.4 S Jul (2 half) Mar (2 half) st nd Singapore 1.2 N Jun (1 half) May (2 half) st st Cameroon 5.7 N Nov (1 half) Nov (1 half) nd st Ivory Coast 7.6 N Nov (2 half) Oct (1 half) nd nd Panama 8.6 N Jun (2 half) Jul (2 half) nd st Costa Rica 10.0 N Dec (2 half) Oct (1 half) st nd Nicaragua 12.9 N Nov (1 half) Aug (2 half) nd st El Salvador 13.8 N Jun (2 half) Jul (1 half) st st Honduras 14.8 N Nov (1 half) Jul (1 half) nd st Guatemala 15.7 N Mar (2 half) Sep (1 half) st st Viet Nam 16.7 N Aug (1 half) Dec (1 half) Northern hemisphere nd nd Bhutan 27.4 N Aug (2 half) Aug (2 half) nd nd China South 31.1 N Feb (2 half) Mar (2 half) nd st Morocco 32.0 N Jan (2 half) Feb (1 half) st nd Turkey 39.0 N Feb (1 half) Feb (2 half) nd st Portugal 39.3 N Jan (2 half) Feb (1 half) nd st China North 39.5 N Jan (2 half) Mar (1 half) nd nd Italy 42.9 N Feb (2 half) Feb (2 half) st nd USA 45.6 N Feb (1 half) Mar (2 half) nd nd Kazakhstan 48.0 N Feb (2 half) Feb (2 half) st st Ukraine 49.1 N Mar (1 half) Mar (1 half) st st Netherlands 52.3 N Feb (1 half) Mar (1 half) nd nd England 52.3 N Jan (2 half) Jan (2 half) https://doi.org/10.1371/journal.pone.0222381.t001 age was 11 vs. 8 years, respectively. In all other countries, B/Yamagata cases were older than B/ Victoria cases, and the difference between the median age was always significant except in Madagascar (p-value 0.248) and in Ukraine (p-value 0.082). The visual inspection of the graphs revealed a fairly consistent pattern in the age distribu- tion of B/Victoria and B/Yamagata cases (S1 File). In most countries, influenza B/Victoria cases tended to be distributed according to their age along a unimodal curve, with a peak below 10 years of age, while the age distribution of B/Yamagata cases frequently followed a bimodal curve, with an earlier, larger peak below 10 years of age, and a smaller, yet discernible peak at older age, mostly between 25 and 50 years of age. The adult age peak of B/Yamagata PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 8 / 17 Epidemiology of influenza B virus and its lineages Fig 3. Proportion of influenza B cases in a season that were caused by the B/Victoria lineage viruses, according to countries median age (top) and geographical area (bottom). The Global Influenza B Study, 2000–2018. https://doi.org/10.1371/journal.pone.0222381.g003 cases was barely noticeable in countries where the majority of inhabitants are younger than 25 years (e.g. Madagascar and Ivory Coast), and became increasingly evident as the median age of the country population increased. A partial exception was Portugal, where the peak in the age PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 9 / 17 Epidemiology of influenza B virus and its lineages Table 2. Age distribution of influenza B/Victoria and B/Yamagata patients by country (sorted by median age of the general population). The Global Influenza B Study, 2000–2018. (a) Country Median age general population No. influenza B cases Median age (IQR) (b) Victoria Yamagata Victoria Yamagata p-value Madagascar 19.7 787 405 7 (4–16) 8 (3–17) 0.248 Ivory Coast 20.9 240 153 3 (1–8) 5 (2–32) 0.002 South Africa 27.1 61 102 3 (1–29) 29 (1–54) 0.003 Indonesia 30.2 422 235 10 (6–20) 25 (7–36) <0.001 Turkey 30.9 98 339 12 (6–22) 34 (11–48) <0.001 Chile 34.4 1351 1950 8 (3–18) 14 (6–21) <0.001 Singapore 34.6 413 921 15 (8–32) 35 (12–51) <0.001 New Zealand 37.9 232 1036 15 (7–33) 35 (11–52) <0.001 Australia 38.7 648 276 11 (5–23) 8 (3–36) 0.468 England 40.5 478 877 16 (8–29) 40 (17–53) <0.001 Ukraine 40.6 109 186 13 (8–19) 15 (7–32) 0.082 Portugal 42.2 57 183 19 (13–36) 46 (34–57) <0.001 Netherlands 42.6 354 773 19 (7–36) 46 (26–58) <0.001 Italy 45.5 68 510 10 (6–19) 12 (6–47) 0.009 IQR: inter-quartile range (a) Countries were included if information on exact age was available for�50 B Victoria and�50 B Yamagata influenza cases. (b) Wilcoxon rank-sum test for comparison of medians https://doi.org/10.1371/journal.pone.0222381.t002 distribution of B/Victoria cases was between 10 and 20 years of age, and B/Yamagata distrib- uted along a unimodal curve peaking at above 50 years of age. Frequency of lineage-level influenza B trivalent vaccine mismatches The proportion of influenza B lineage-level vaccine mismatch was 54.2% in countries in the Southern hemisphere, and 42.9% in countries of the Northern hemisphere. For countries in the inter-tropical belt, the proportion of seasons with a lineage-level vaccine mismatch was 29.6% in the scenario in which these countries use the vaccine composition recommended for the hemisphere they are situated in; 40.7% assuming all these countries used the recommended vaccine composition for the Northern hemisphere; and 22.2% if all countries used the recom- mended vaccine composition for the Southern hemisphere. Discussion Our analysis found that influenza B virus was responsible for nearly one fourth of all influenza cases in an average season between 2000 and 2018. More specifically, influenza B virus accounted for over 20% of all influenza cases in more than half of the seasons, and was the dominant virus type in about one out of every seven seasons. This proportion varied geograph- ically, as the proportion of influenza B over all influenza cases tended to be higher on average in tropical countries compared to countries of the Northern and Southern hemispheres. These findings were broadly in line with reports from countries around the world [5,9,20–21,30–31] and with GIBS data published in 2015 , with moderate discrepancies (less than 5%) being most likely linked to the variability of the time span covered by different studies. Influenza B epidemics tended to peak on average three weeks later than influenza A epidemics during the winter period in temperate countries of both hemispheres, which is consistent with previous country reports [21,30,32–34]. The timing of influenza A and B epidemics was different in PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 10 / 17 Epidemiology of influenza B virus and its lineages tropical countries where they were highly heterogeneous and showed no consistent pattern in the timing of the different epidemics. Influenza viruses exhibit differential ability to induce temporary immunity against re-infection with other viruses in experimental settings . This “viral hierarchy” may help explain the differences in timing of epidemics caused by different virus types and subtypes where the weather conditions that are most favourable to the occur- rence of influenza epidemics (e.g. cold-dry) last for only a limited number of weeks each year. The study of patterns of circulation of B/Victoria and B/Yamagata lineages revealed two main findings. Firstly, the proportion of B cases caused by either lineage in a given season ran- ged between 0% and nearly 100% in each geographical area (i.e. anywhere around the world seasons occurred where all influenza B cases were caused by either the B/Victoria or the B/ Yamagata lineage), which emphasizes the challenges faced when trying to predict which line- age will cause most influenza B cases next season. This helps explain the very high frequency of lineage-level vaccine mismatch for the TIV which was around 50% (i.e. close to the scenario in which the choice of the vaccine lineage is made at random) in temperate countries of the two hemispheres, which has been reported in the past [30,36–37]. The second important finding that emerges from the B/Victoria-B/Yamagata lineage analy- sis is that whilst the two lineages caused a comparable proportion of influenza B cases globally during the study period, their distribution varied geographically, as B/Victoria was relatively more frequent in tropical countries, while B/Yamagata was more frequent in temperate climate countries of the Southern and Northern hemispheres. The reason for this pattern of circulation of the two lineages is not clear. One possibility is that the survival and transmissibility of B/Vic- toria and B/Yamagata viruses is differentially affected by “cold-dry” and “humid-rainy” envi- ronmental conditions that drive the occurrence of influenza epidemics in temperate and tropical countries, respectively . This hypothesis does not seem, however, to be supported by the available evidence, which suggests instead that influenza viruses of different types and subtypes (e.g. A(H3N2), A(H1N1)pdm09, and B) are similarly affected by weather conditions . Based on our findings on the differential age distribution of B/Victoria and B/Yamagata infected cases, a more likely explanation for the unequal geographical distribution of the two B lineages lies is in the diverse demographic structure of world countries, with those located around the tropics having a lower median age, on average, than those in temperate climates. In this regard, our study showed consistently at a global level (i.e. across countries that dif- fer greatly under many aspects, including the type of influenza surveillance system) that B/ Yamagata cases are on average older than B/Victoria cases, and in particular, that the former virus lineage infects adult individuals (�25 years) more frequently than the latter. The only exception (i.e. the only country in which B/Victoria cases were older than B/Yamagata cases, albeit not significantly) was Australia. The reason for this divergent pattern is not clear; how- ever, since the majority of B/Yamagata cases in the database from Australia occurred in 2008 (see S1 Table), this finding might be driven by a single season with an unusual age distribution of B/Yamagata patients. Importantly, our findings support earlier work showing that each influenza A subtype (H3N2, pre-2009-pandemic H1N1, and H1N1pdm09) tends to affect a different age group . A recent study has shown that the B/Victoria viruses antigenic drift parallels that of A(H3N2), with limited cross-reactivity between phylogenetic clusters com- pared to B/Yamagata (which, in contrast, resembled A(H1N1) under this regard, with multiple variants circulating and greater levels of antigenic cross-reactivity) . According to these observations, one would expect re-infection among adult individuals to occur more frequently for B/Victoria rather than B/Yamagata, but the observed pattern runs counter to this expecta- tion. Vijaykrishna and colleagues suggested that the observed differential age distribution may be explained by either the higher effective reproductive number of B/Victoria; a broader response against B/Victoria viruses in older people; and/or differences in the prevalence of PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 11 / 17 Epidemiology of influenza B virus and its lineages receptor binding molecules that help B/Victoria and B/Yamagata viruses infect respiratory tract cells in young children and adults . Another hypothesis is that the bimodal age distri- bution of B/Yamagata cases may be the result of a shorter duration of acquired immunity developed after infection with B/Yamagata vs. B/Victoria viruses . Finally, antigenic origi- nal sin (i.e. preferential response to influenza B viruses encountered in the past ) may also play a role. If this were true, the mean age of patients infected with either B lineage would be expected to drift over time: this has occurred in Australia between 2008–2011 , but we were unable to replicate these results in the GIBS database (results not shown). These findings have important implications for assessing the impact of influenza B viruses in a population each season. Knowing early in the season (through surveillance activities) which influenza B lineage virus is circulating will allow one to predict which age groups will be more affected, with B/Yamagata seasons being generally characterized by more adults and elderly people being infected and, therefore, by a greater burden of disease compared to B/Vic- toria seasons. This appears to have been the case during the 2017/18 season in Europe, when B/Yamagata was dominant (and there was a lineage-level vaccine mismatch) and a high bur- den of influenza B was observed in many European countries . We analyzed influenza surveillance data from around the world to determine the epidemi- st ology of influenza B virus and its lineages in the 21 century. We have enriched the analysis of an earlier GIBS paper  by including more countries (31 vs. 26) and additional years (data to up to 2018): the overall number of reports analysed is now twice as large (>1.8 million vs. 935,000 cases). This allowed us to conduct a comprehensive comparative analysis of the epide- miological characteristics of B/Victoria and B/Yamagata virus lineages, thus achieving a better understanding of the epidemiology of influenza B globally. The most important limitation of our study is that it relies on data that were collected within national surveillance systems rather than for research purposes. Countries participating in GIBS inevitably differ in terms of a number of important characteristics, including percentage of out- and in-patients included, number of years with data and average number of influenza cases reported per year, propor- tion of influenza B viruses that were characterized availability of information on age of these cases and sampling strategies across age groups. In each individual country, changes in the aforementioned characteristics may have occurred over time, for instance before and after the 2009 pandemic. Another limitation of our study is the uneven distribution of cases among countries and seasons, which is partly due to differences among countries in terms of popula- tion, the intensity of surveillance activities, and the coverage of each country’s influenza sur- veillance system. Throughout this paper, we have used statistical methods that were intended to minimize the impact of this variability on the results. These include the use of the ‘season’ as a unit of analysis for the comparison of the frequency of B influenza across countries, the stan- dardization offered by the EPIPOI software in the analysis of temporal patterns of influenza epidemics, and the use of medians to compare the age distribution of influenza cases between countries. However, we acknowledge that these sources of variability may have affected our findings to an extent that is difficult to quantify precisely, and a further improvement in the harmonization of influenza surveillance data remains desirable as it would allow researchers to conduct more detailed and in-depth analyses and obtain more reliable results. In conclusion, our study provides important new information that describes the epidemiol- st ogy of influenza B viruses and its lineages during the start of the 21 century. We showed that influenza B virus caused nearly one fourth (on average) of all influenza cases occurring annu- ally in the world during 2000–2018, and that both lineages accounted each for a comparable proportion of all influenza B cases globally. We also observed that cases infected with the B/ Victoria or B/Yamagata lineages differed consistently (i.e. in a similar fashion across countries) in terms of their age distribution, and suggested that this may help explain the uneven PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 12 / 17 Epidemiology of influenza B virus and its lineages geographical distribution of the two lineages globally. Finally, we confirmed that the TIV suf- fers from a high frequency of lineage-level mismatch, especially in temperate countries of the two hemispheres. While expanding our knowledge of the epidemiology and pattern of circula- tion of influenza B virus, these findings will help inform health policy decisions aiming to reduce disease burden associated with seasonal influenza locally, regionally and globally. Supporting information S1 Table. Laboratory-confirmed influenza cases reported in each country and season, and proportion of those caused by the influenza B virus type (overall and by lineage). The Global Influenza B Study, 2000–2018. (a) A season was defined as the period between the 27th week of a given year and the 26th week of the following year for countries in the Northern hemisphere, and as the period between the first and last week of a given year for other countries. (b) Seasons with fewer than 100 influenza cases were not included. (c) Number of influenza B cases in each season, and proportion of influenza B over all influenza cases in the same season. (d) Number of influenza B cases for which the virus lineage was characterized, and proportion over all influenza B cases in the same season. (e) Number of influenza B cases caused by viruses belonging to the Victoria and Yamagata line- ages, and proportion over all characterized influenza B viruses in the same season. Not reported in seasons with fewer than 50 characterized influenza B cases. (DOCX) S1 File. Age distribution of laboratory-confirmed influenza patients infected with type B viruses belonging to the Victoria or Yamagata lineage. Countries which provided informa- tion on exact age for�50 B Victoria and�50 B Yamagata influenza for the whole study period cases were included. Countries were ordered according to the median age of the general popu- lation. The Global Influenza B Study, 2000–2018. (PDF) S2 File. Contact information of data owners. (DOC) Acknowledgments The “Global Influenza B Study team” (group authorship) includes the following scientists: Juan Manuel Rudi (email@example.com), National Institute of Respiratory Diseases "Emilio Coni", Santa Fe, Argentina; Dorji Wangchuk (firstname.lastname@example.org) and Sangay Zangmo (email@example.com), Royal Centre for Disease Control, Department of Public Health, Ministry of Health, Thimphu, Bhutan; Daiana Araujo da Silva (daiana.silva@saude. gov.br), Ministry of Health, Department of Surveillance of Transmissible Diseases, Brası ´lia/ DF, Brazil; Winston Andrade (firstname.lastname@example.org), Sub-Department of Viral Diseases, Insti- tuto de Salud Publica de Chile, Santiago, Chile; Jiandong Zheng (email@example.com) and Ying Qin (firstname.lastname@example.org), Division of Infectious Diseases, Chinese Center for Disease Control and Prevention, Beijing, P.R. China; Joanna Ellis (email@example.com), Public Health England, London, United Kingdom; Simona Puzelli (firstname.lastname@example.org), National Influenza Center, Department of Infectious Diseases, National Institute of Health, Rome, Italy; Caterina Rizzo (email@example.com), Bambino Gesu ` Children’s Hospital, Rome, Italy; Linus Ndegwa (firstname.lastname@example.org), Influenza Program, Centers for Disease Control and Preven- tion, Nairobi, Kenya; Marit MA de Lange (email@example.com) and Anne C. Teirlinck PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 13 / 17 Epidemiology of influenza B virus and its lineages (firstname.lastname@example.org), National Institute for Public Health and the Environment, Centre for Infectious Diseases, Epidemiology and Surveillance, Bilthoven, The Netherlands; Jeffery Cutter (email@example.com) and Raymond Tzer Pin Lin (firstname.lastname@example.org), Public Health Group, Ministry of Health, Singapore, Singapore; Than T. Le (lmot82@yahoo. com), National Institute of Hygiene and Epidemiology, Hanoi, Vietnam; Peter Kinuthia (email@example.com), IHRC Inc., Atlanta, USA. We thank the Australia Influenza Surveillance Network for making their data publically available to researchers. In the Netherlands, we would like to thank the general practitioner network coordinated by Nivel and their patients for participation in the surveillance of influ- enza-like illness and other acute respiratory infections; the technicians who did all the labora- tory analyses of the surveillance specimens; and epidemiologists with which this surveillance collaboratively has been performed (Marit de Lange, MSc; Dr. Anne Teirlinck; Dr. Wim van der Hoek). Influenza surveillance data Turkey were provided by Meral A. Ciblak (Sanofi Pas- teur, Lyon, France) while working at the Istanbul University, Istanbul, Turkey. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. Author Contributions Conceptualization: Saverio Caini, Clotilde El Guerche-Se ´blain, John Paget. Data curation: Saverio Caini, Gabriela Kusznierz, Verònica Vera Garate, Sonam Wangchuk, Binay Thapa, Francisco Jose ´ de Paula Ju ´ nior, Walquiria Aparecida Ferreira de Almeida, Richard Njouom, Rodrigo A. Fasce, Patricia Bustos, Luzhao Feng, Zhibin Peng, Jenny Lara ´ ´ ´ Araya, Alfredo Bruno, Domenica de Mora, Monica Jeannette Barahona de Gamez, Richard Pebody, Maria Zambon, Rocio Higueros, Rudevelinda Rivera, Herman Kosasih, Maria Rita Castrucci, Antonino Bella, Herve ´ A. Kadjo, Coulibaly Daouda, Ainash Makusheva, Olga Bessonova, Sandra S. Chaves, Gideon O. Emukule, Jean-Michel Heraud, Norosoa H. Raza- najatovo, Amal Barakat, Fatima El Falaki, Adam Meijer, Ge ´ A. Donker, Q. Sue Huang, Tim Wood, Angel Balmaseda, Rakhee Palekar, Brechla Moreno Are ´valo, Ana Paula Rodrigues, Raquel Guiomar, Vernon Jian Ming Lee, Li Wei Ang, Cheryl Cohen, Florette Treurnicht, Alla Mironenko, Olha Holubka, Joseph Bresee, Lynnette Brammer, Mai T. Q. Le, Phuong V. M. Hoang, John Paget. Formal analysis: Saverio Caini. Funding acquisition: Clotilde El Guerche-Se ´blain, John Paget. Investigation: Saverio Caini, John Paget. Methodology: Saverio Caini, Clotilde El Guerche-Se ´blain, John Paget. Project administration: Clotilde El Guerche-Seblain, John Paget. Resources: Gabriela Kusznierz, Verònica Vera Garate, Sonam Wangchuk, Binay Thapa, Fran- ´ ´ cisco Jose de Paula Junior, Walquiria Aparecida Ferreira de Almeida, Richard Njouom, Rodrigo A. Fasce, Patricia Bustos, Luzhao Feng, Zhibin Peng, Jenny Lara Araya, Alfredo Bruno, Dome ´nica de Mora, Mo ´nica Jeannette Barahona de Ga ´mez, Richard Pebody, Maria Zambon, Rocio Higueros, Rudevelinda Rivera, Herman Kosasih, Maria Rita Castrucci, Antonino Bella, Herve A. Kadjo, Coulibaly Daouda, Ainash Makusheva, Olga Bessonova, Sandra S. Chaves, Gideon O. Emukule, Jean-Michel Heraud, Norosoa H. Razanajatovo, Amal Barakat, Fatima El Falaki, Adam Meijer, Ge ´ A. Donker, Q. Sue Huang, Tim Wood, Angel Balmaseda, Rakhee Palekar, Brechla Moreno Arevalo, Ana Paula Rodrigues, Raquel Guiomar, Vernon Jian Ming Lee, Li Wei Ang, Cheryl Cohen, Florette Treurnicht, Alla PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 14 / 17 Epidemiology of influenza B virus and its lineages Mironenko, Olha Holubka, Joseph Bresee, Lynnette Brammer, Mai T. Q. Le, Phuong V. M. Hoang, Clotilde El Guerche-Se ´blain. Supervision: John Paget. Visualization: Saverio Caini. Writing – original draft: Saverio Caini, John Paget. Writing – review & editing: Gabriela Kusznierz, Verònica Vera Garate, Sonam Wangchuk, Binay Thapa, Francisco Jose ´ de Paula Ju ´ nior, Walquiria Aparecida Ferreira de Almeida, Richard Njouom, Rodrigo A. Fasce, Patricia Bustos, Luzhao Feng, Zhibin Peng, Jenny Lara Araya, Alfredo Bruno, Dome ´nica de Mora, Mo ´nica Jeannette Barahona de Ga ´mez, Richard Pebody, Maria Zambon, Rocio Higueros, Rudevelinda Rivera, Herman Kosasih, Maria Rita Castrucci, Antonino Bella, Herve ´ A. Kadjo, Coulibaly Daouda, Ainash Makusheva, Olga Bessonova, Sandra S. Chaves, Gideon O. Emukule, Jean-Michel Heraud, Norosoa H. Raza- najatovo, Amal Barakat, Fatima El Falaki, Adam Meijer, Ge ´ A. Donker, Q. Sue Huang, Tim Wood, Angel Balmaseda, Rakhee Palekar, Brechla Moreno Are ´valo, Ana Paula Rodrigues, Raquel Guiomar, Vernon Jian Ming Lee, Li Wei Ang, Cheryl Cohen, Florette Treurnicht, Alla Mironenko, Olha Holubka, Joseph Bresee, Lynnette Brammer, Mai T. Q. Le, Phuong V. M. Hoang, Clotilde El Guerche-Se ´blain, John Paget. References 1. Iuliano AD, Roguski KM, Chang HH, Muscatello DJ, Palekar R, Tempia S, et al. Estimates of global sea- sonal influenza-associated respiratory mortality: a modelling study. Lancet. 2018; 391(10127):1285– 1300. https://doi.org/10.1016/S0140-6736(17)33293-2 PMID: 29248255 2. Rolfes MA, Foppa IM, Garg S, Flannery B, Brammer L, Singleton JAet al. Annual estimates of the bur- den of seasonal influenza in the United States: A tool for strengthening influenza surveillance and pre- paredness. Influenza Other Respir Viruses. 2018; 12(1):132–137. https://doi.org/10.1111/irv.12486 PMID: 29446233 3. Cassini A, Colzani E, Pini A, Mangen MJ, Plass D, McDonald SA, et al. Impact of infectious diseases on population health using incidence-based disability-adjusted life years (DALYs): results from the Burden of Communicable Diseases in Europe study, European Union and European Economic Area countries, 2009 to 2013. Euro Surveill. 2018; 23(16). 4. Lee VJ, Ho ZJM, Goh EH, Campbell H, Cohen C, Cozza V, et al. Advances in measuring influenza bur- den of disease. Influenza Other Respir Viruses. 2018; 12(1):3–9. https://doi.org/10.1111/irv.12533 PMID: 29460425 5. Paul Glezen W, Schmier JK, Kuehn CM, Ryan KJ, Oxford J. The burden of influenza B: a structured lit- erature review. Am J Public Health. 2013; 103(3):e43–51. https://doi.org/10.2105/AJPH.2012.301137 PMID: 23327249 6. Tafalla M, Buijssen M, Geets R, Vonk Noordegraaf-Schouten M. A comprehensive review of the epide- miology and disease burden of Influenza B in 9 European countries. Hum Vaccin Immunother. 2016; 12 (4):993–1002. https://doi.org/10.1080/21645515.2015.1111494 PMID: 26890005 7. Rota PA, Wallis TR, Harmon MW, Rota JS, Kendal AP, Nerome K. Cocirculation of two distinct evolu- tionary lineages of influenza type B virus since 1983. Virology. 1990; 175(1):59–68. https://doi.org/10. 1016/0042-6822(90)90186-u PMID: 2309452 8. Paiva TM, Benega MA, Silva DB, Santos KC, Cruz AS, Hortenci MF, et al. Evolutionary pattern of reemerging influenza B/Victoria lineage viruses in São Paulo, Brazil, 1996–2012: Implications for vac- cine composition strategy. J Med Virol. 2013; 85(11):1983–9. https://doi.org/10.1002/jmv.23684 PMID: 9. Jennings L, Huang QS, Barr I, Lee PI, Kim WJ, Buchy P, et al. Literature review of the epidemiology of influenza B disease in 15 countries in the Asia-Pacific region. Influenza Other Respir Viruses. 2018; 12 (3):383–411. https://doi.org/10.1111/irv.12522 PMID: 29127742 10. Mertz D, Kim TH, Johnstone J, Lam PP, Science M, Kuster SP, et al. Populations at risk for severe or complicated influenza illness: systematic review and meta-analysis. BMJ. 2013; 347:f5061. https://doi. org/10.1136/bmj.f5061 PMID: 23974637 PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 15 / 17 Epidemiology of influenza B virus and its lineages 11. Caini S, Schellevis F, El-Guerche Se ´ blain C, Paget J. Important changes in the timing of influenza epi- demics in the WHO European Region over the past 20 years: virological surveillance 1996 to 2016. Euro Surveill. 2018; 23(1). 12. Tricco AC, Chit A, Soobiah C, Hallett D, Meier G, Chen MH, et al. Comparing influenza vaccine efficacy against mismatched and matched strains: a systematic review and meta-analysis. BMC Med. 2013; 11:153. https://doi.org/10.1186/1741-7015-11-153 PMID: 23800265 13. Heikkinen T, Ikonen N, Ziegler T. Impact of influenza B lineage-level mismatch between trivalent sea- sonal influenza vaccines and circulating viruses, 1999–2012. Clin Infect Dis. 2014; 59(11):1519–24. https://doi.org/10.1093/cid/ciu664 PMID: 25139969 14. Caini S, Huang QS, Ciblak MA, Kusznierz G, Owen R, Wangchuk S, et al. Epidemiological and virologi- cal characteristics of influenza B: results of the Global Influenza B Study. Influenza Other Respir Viruses. 2015; 9 Suppl 1:3–12. 15. de Boer PT, van Maanen BM, Damm O, Ultsch B, Dolk FCK, Cre ´ pey P, et al. A systematic review of the health economic consequences of quadrivalent influenza vaccination. Expert Rev Pharmacoecon Out- comes Res. 2017; 17(3):249–265. https://doi.org/10.1080/14737167.2017.1343145 PMID: 28613092 16. Sočan M, Prosenc K, Učakar V, Berginc N. A comparison of the demographic and clinical characteris- tics of laboratory-confirmed influenza B Yamagata and Victoria lineage infection. J Clin Virol. 2014; 61 (1):156–60. https://doi.org/10.1016/j.jcv.2014.06.018 PMID: 25034374 17. Le TT, Pham TH, Pham TH, Nguyen le KH, Nguyen CT, Hoang VM, et al. Circulation of influenza B line- ages in northern Viet Nam, 2007–2014. Western Pac Surveill Response J. 2015; 6(4):17–23. https:// doi.org/10.5365/WPSAR.2015.6.1.022 PMID: 26798557 18. Xu C, Chan KH, Tsang TK, Fang VJ, Fung RO, Ip DK, et al. Comparative epidemiology of influenza B Yamagata- and Victoria-lineage viruses in households. Am J Epidemiol. 2015; 182(8):705–13. https:// doi.org/10.1093/aje/kwv110 PMID: 26400854 19. Barr IG, Vijaykrishna D, Sullivan SG. Differential age susceptibility to influenza B/Victoria lineage viruses in the 2015 Australian influenza season. Euro Surveill. 2016; 21(4). 20. Seleka M, Treurnicht FK, Tempia S, Hellferscee O, Mtshali S, Cohen AL, et al. Epidemiology of influ- enza B/Yamagata and B/Victoria lineages in South Africa, 2005–2014. PLoS One. 2017; 12(5): e0177655. https://doi.org/10.1371/journal.pone.0177655 PMID: 28542324 21. Moa AM, Muscatello DJ, Turner RM, MacIntyre CR. Epidemiology of influenza B in Australia: 2001– 2014 influenza seasons. Influenza Other Respir Viruses. 2017; 11(2):102–109. https://doi.org/10.1111/ irv.12432 PMID: 27650482 22. Lapinscki B, Pereira LA, Nogueira MB, Vidal LR, Riediger I, Debur MC, et al. Molecular epidemiology of influenza B virus and implications in immunization strategy, Southern Brazil. Vaccine. 2018; 36(1):107– 113. https://doi.org/10.1016/j.vaccine.2017.11.033 PMID: 29174679 23. Caini S, Andrade W, Badur S, Balmaseda A, Barakat A, Bella A, et al. Temporal patterns of influenza A and B in tropical and temperate countries: what are the lessons for influenza vaccination? PLoS One. 2016; 11(3):e0152310. https://doi.org/10.1371/journal.pone.0152310 PMID: 27031105 24. Caini S, Spreeuwenberg P, Kusznierz GF, Rudi JM, Owen R, Pennington K, et al. Distribution of influ- enza virus types by age using case-based global surveillance data from twenty-nine countries, 1999– 2014. BMC Infect Dis. 2018; 18(1):269. https://doi.org/10.1186/s12879-018-3181-y PMID: 29884140 25. Central Intelligence Agency. The World Factbook. Available from https://www.cia.gov/library/ publications/the-world-factbook/ Cited 20 June 2018 26. Alonso WJ, McCormick BJ. EPIPOI: a user-friendly analytical tool for the extraction and visualization of temporal parameters from epidemiological time series. BMC Public Health. 2012; 12(1):982. 27. World Health Organization. Recommendations on the composition of influenza virus vaccines. http:// www.who.int/influenza/vaccines/virus/recommendations/en/ Cited 18 June 2018 28. World Health Organization. Which vaccine formulation to use–Northern or Southern hemisphere? http://www.who.int/influenza/vaccines/tropics/vaccination_formulation/en/ Cited 18 June 2018 29. Hirve S, Lambach P, Paget J, Vandemaele K, Fitzner J, Zhang W. Seasonal influenza vaccine policy, use and effectiveness in the tropics and subtropics—a systematic literature review. Influenza Other Respir Viruses. 2016; 10(4):254–67. https://doi.org/10.1111/irv.12374 PMID: 26842617 30. Mosnier A, Caini S, Daviaud I, Bensoussan JL, Stoll-Keller F, Bui TT, et al. Ten influenza seasons in France: distribution and timing of influenza A and B circulation, 2003–2013. BMC Infect Dis. 2015; 15:357. https://doi.org/10.1186/s12879-015-1056-z PMID: 26289794 31. Barros EN, Cintra O, Rossetto E, Freitas L, Colindres R. Patterns of influenza B circulation in Brazil and its relevance to seasonal vaccine composition. Braz J Infect Dis. 2016; 20(1):81–90. https://doi.org/10. 1016/j.bjid.2015.09.009 PMID: 26626166 PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 16 / 17 Epidemiology of influenza B virus and its lineages 32. Pan Y, Zhang Y, Yang P, Qian H, Shi W, Wu S, et al. Epidemiological and phylogenetic characteristics of influenza B infection in severe acute respiratory infection cases in Beijing, 2014 to 2015. Medicine (Baltimore). 2015; 94(52):e2399. 33. van Asten L, Bijkerk P, Fanoy E, van Ginkel A, Suijkerbuijk A, van der Hoek W, et al. Early occurrence of influenza A epidemics coincided with changes in occurrence of other respiratory virus infections. Influenza Other Respir Viruses. 2016; 10(1):14–26. https://doi.org/10.1111/irv.12348 PMID: 26369646 34. Caini S, Alonso WJ, Se ´ blain CE, Schellevis F, Paget J. The spatiotemporal characteristics of influenza A and B in the WHO European Region: can one define influenza transmission zones in Europe? Euro Surveill. 2017; 22(35). 35. Laurie KL, Guarnaccia TA, Carolan LA, Yan AW, Aban M, Petrie S, et al. Interval between infections and viral hierarchy are determinants of viral interference following influenza virus infection in a ferret model. J Infect Dis. 2015; 212(11):1701–10. https://doi.org/10.1093/infdis/jiv260 PMID: 25943206 36. Belshe RB, Coelingh K, Ambrose CS, Woo JC, Wu X. Efficacy of live attenuated influenza vaccine in children against influenza B viruses by lineage and antigenic similarity. Vaccine. 2010; 28(9):2149–56. https://doi.org/10.1016/j.vaccine.2009.11.068 PMID: 20003926 37. de Boer PT, Kelso JK, Halder N, Nguyen TP, Moyes J, Cohen C, et al. The cost-effectiveness of triva- lent and quadrivalent influenza vaccination in communities in South Africa, Vietnam and Australia. Vac- cine. 2018; 36(7):997–1007. https://doi.org/10.1016/j.vaccine.2017.12.073 PMID: 29373192 38. Tamerius JD, Shaman J, Alonso WJ, Bloom-Feshbach K, Uejio CK, Comrie A, et al. Environmental pre- dictors of seasonal influenza epidemics across temperate and tropical climates. PLoS Pathog. 2013; 9 (3):e1003194. https://doi.org/10.1371/journal.ppat.1003194 PMID: 23505366 39. Lowen AC, Steel J. Roles of humidity and temperature in shaping influenza seasonality. J Virol. 2014; 88(14):7692–5. https://doi.org/10.1128/JVI.03544-13 PMID: 24789791 40. Vijaykrishna D, Holmes EC, Joseph U, Fourment M, Su YC, Halpin R, et al. The contrasting phylody- namics of human influenza B viruses. Elife. 2015; 4:e05055. https://doi.org/10.7554/eLife.05055 PMID: 41. Nyirenda M, Omori R, Tessmer HL, Arimura H, Ito K. Estimating the lineage dynamics of human influ- enza B viruses. PLoS One. 2016; 11(11):e0166107. https://doi.org/10.1371/journal.pone.0166107 PMID: 27829058 42. Skowronski DM, Hottes TS, De Serres G, Ward BJ, Janjua NZ, Sabaiduc S, et al. Influenza B/Victoria antigen induces strong recall of B/Yamagata but lower B/Victoria response in children primed with two doses of B/Yamagata. Pediatr Infect Dis J. 2011; 30(10):833–9. https://doi.org/10.1097/INF. 0b013e31822db4dc PMID: 21857263 43. Adlhoch C, Snacken R, Melidou A, Ionescu S, Penttinen; The European Influenza Surveillance Net- work. Dominant influenza A(H3N2) and B/Yamagata virus circulation in EU/EEA, 2016/17 and 2017/18 seasons, respectively. Euro Surveill. 2018; 23(13). PLOS ONE | https://doi.org/10.1371/journal.pone.0222381 September 12, 2019 17 / 17
PLoS ONE – Public Library of Science (PLoS) Journal
Published: Sep 12, 2019
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera