{
  "id": "influenza",
  "title": "Influenza",
  "lang": "en",
  "source_lang": "en",
  "summary": "A contagious respiratory illness caused by influenza viruses, producing seasonal epidemics, occasional pandemics and zoonotic spillover.",
  "category": "health",
  "subcategory": "diseases",
  "type": "concept",
  "aliases": [
    "flu",
    "grippe"
  ],
  "generated_at": "2026-10-10",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/influenza/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/influenza/",
      "json": "https://aiwiki.page/zh/influenza.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Influenza, commonly called the flu, is an acute infection of the respiratory tract caused by influenza viruses, enveloped RNA viruses in the virus family Orthomyxoviridae. Its importance comes from three overlapping patterns: annual seasonal epidemics that affect large shares of the population each year and cause substantial illness and death among the elderly, the very young and people with chronic conditions; sporadic but potentially catastrophic pandemics, such as the 1918–1919 event; and zoonotic outbreaks in which viruses circulating in birds and other animals infect humans. The disease is a central subject of epidemiology, immunology and public health, and it has shaped the development of vaccines and antiviral drugs.\n\nThe entry first defines the virus and its classification into types A, B, C and D, with particular attention to the hemagglutinin and neuraminidase subtypes that distinguish influenza A. It then explains how the virus is structured and how its segmented genome, antigenic drift and reassortment generate new variants, followed by transmission, pathogenesis, clinical course, complications and diagnosis. Immune responses, vaccination and antiviral treatment are treated as separate topics, as is the surveillance system that selects vaccine strains. The later sections cover the history of epidemics and pandemics, the ecology of avian and mammalian reservoirs, the economic and social burden, and the open scientific questions that remain, including universal vaccines and pandemic forecasting."
    },
    {
      "heading": "Definition and scope",
      "text": "Influenza, commonly called flu, is an acute, contagious respiratory infectious disease caused by influenza viruses. It affects the nose, throat and sometimes the lungs, with illness ranging from mild infection to severe disease. Typical illness begins abruptly with fever, cough, muscle aches and marked fatigue, although fever is not present in every case. Most people recover from fever and other symptoms within about a week without medical attention, but influenza can cause pneumonia, hospitalization and death. The term identifies an infection with a particular group of viruses, rather than any illness accompanied by respiratory symptoms or fever. (who.int)\n\nInfluenza differs from the common cold in both cause and usual clinical pattern. Colds can result from several groups of viruses, including rhinoviruses and seasonal coronaviruses; influenza is caused only by influenza viruses. Cold symptoms generally develop more gradually, are milder and feature nasal congestion or a runny nose more prominently, whereas influenza more often produces abrupt, intense systemic symptoms. COVID-19 is a separate infection caused by SARS-CoV-2, a coronavirus first identified in 2019. Both COVID-19 and influenza range from asymptomatic infection to life-threatening disease, and COVID-19 can cause more severe illness in some people. Their overlapping symptoms mean that symptoms alone cannot reliably establish which infection is present; laboratory testing can distinguish them. (cdc.gov)\n\nThere are four types of influenza virus: A, B, C and D. Types A, B and C infect humans, but A and B are responsible for seasonal epidemics. Influenza A includes viruses circulating in humans and viruses associated with animal hosts; A(H1N1) and A(H3N2) are examples of human seasonal subtypes. Influenza B also causes seasonal human disease. Type C is detected less frequently and generally causes mild illness rather than the large epidemics associated with A and B. Type D primarily affects cattle, can infect other animals and is not known to cause illness in people. These letters denote distinct virus types, not successive stages of disease or grades of clinical severity. (cdc.gov)\n\nSeasonal influenza refers to recurrent epidemics caused by influenza A and B viruses circulating among people. In temperate regions, epidemics occur mainly in winter; in tropical regions, influenza may occur throughout the year with less regular outbreaks. Seasonal influenza is distinct from pandemic influenza, a global epidemic associated with the emergence of a new influenza A virus capable of sustained human-to-human spread in a population with little or no immunity to it. In everyday usage, “flu” is sometimes applied more broadly, but “stomach flu” is not influenza: it usually denotes viral gastroenteritis, an intestinal infection caused by other viruses. (who.int)\n\nAvian influenza, or bird flu, is a collective term for infection with influenza A viruses associated with birds, not the name of one virus or a synonym for seasonal influenza. A(H5N1) is one example. Avian influenza viruses can occasionally infect mammals, including humans, but an animal-to-human infection does not by itself constitute a human pandemic. The labels “highly pathogenic” and “low pathogenic” describe characteristics of these viruses and their effects in chickens; they do not directly specify the severity of disease in an infected person. (cdc.gov)"
    },
    {
      "heading": "Virus structure and genome",
      "text": "Influenza virions are enveloped particles whose lipid membrane is acquired from the host cell membrane during budding. They are pleomorphic: roughly spherical particles are commonly about 80–120 nm in diameter, whereas filamentous forms can extend several micrometres. Their genome consists of separate segments of negative-sense, single-stranded RNA. Influenza A and B viruses have eight segments; influenza C and D viruses have seven. “Negative-sense” means that genomic RNA is complementary to the coding sequence of messenger RNA and cannot itself be translated into viral proteins. Consequently, the virion carries the polymerase needed to initiate transcription after infection. (ictv.global)\n\nIn influenza A and B, two major glycoproteins project from the envelope: trimeric hemagglutinin (HA) and tetrameric neuraminidase (NA). HA binds sialic-acid-containing molecules on the cell surface, allowing uptake into an endosome. Host proteases must first cleave its precursor, HA0, into HA1 and HA2, which remain associated. The acidic endosomal environment then triggers a conformational rearrangement that exposes the HA2 fusion peptide and brings the viral and endosomal membranes together. NA acts as a receptor-destroying enzyme: it removes terminal sialic acids, helping newly formed virions detach from the cell and preventing their aggregation through HA–receptor interactions. Influenza C and D instead possess a hemagglutinin–esterase–fusion protein, HEF, combining receptor binding, membrane fusion and receptor-destroying esterase activity. (ictv.global)\n\nBeneath the envelope, the matrix protein M1 provides structural organization and connects membrane-associated components with the internal genome. Each RNA segment is associated with multiple copies of nucleoprotein (NP) and a viral polymerase, forming a viral ribonucleoprotein complex, or vRNP. NP supports RNA organization rather than enclosing it in a rigid capsid. In influenza A, the eight segments conventionally encode PB2, PB1, PA, HA, NP, NA, matrix proteins and nonstructural proteins, respectively. Segment number therefore does not equal protein number: alternative RNA splicing and translation permit individual segments to produce several proteins, including M1 and M2 from the matrix segment. (ictv.global)\n\nDuring influenza A entry, the M2 proton channel allows acidification of the virion interior, facilitating separation of M1 from the vRNPs. After HA-mediated membrane fusion, the released complexes undergo further uncoating and enter the nucleus through host nuclear-import machinery. Unlike many RNA viruses that synthesize RNA in the cytoplasm, influenza conducts both transcription and genome replication in the nucleus. Its polymerase is a three-subunit complex: PB1 contains the RNA-synthesis catalytic centre, PB2 binds capped host RNA, and PA contains the endonuclease that cleaves it. (ictv.global)\n\nViral transcription begins through cap-snatching. PB2 captures the 5′ cap of a newly synthesized host RNA, and PA cuts approximately 10–14 nucleotides downstream. The resulting capped fragment becomes a primer that PB1 extends using viral genomic RNA as the template. Structures reported by Stefan Reich and colleagues in 2014 showed how rotation of the PB2 cap-binding domain can direct this primer toward the polymerase active site. Viral mRNA receives a poly(A) tail through repeated copying, or “stuttering,” over a short uridine-rich sequence near the template’s 5′ end; it is then exported for translation by cellular ribosomes. (nature.com)\n\nGenome replication differs from transcription: it does not require a stolen capped primer. The polymerase produces a full-length, positive-sense complementary RNA, or cRNA, which serves as the template for new negative-sense genomic RNA. Newly synthesized RNA associates with NP and polymerase to form ribonucleoprotein complexes. Progeny vRNPs are exported from the nucleus and assembled with membrane-associated viral proteins at budding sites. In influenza A, M2 also contributes to the final membrane-scission step that separates the budding particle from the cell, while NA removes receptors that would otherwise retain or aggregate the released virions. (nature.com)"
    },
    {
      "heading": "Classification of types and subtypes",
      "text": "Influenza A viruses are divided into subtypes according to two surface proteins: hemagglutinin (HA) and neuraminidase (NA). The subtype designation combines their identifying numbers: A(H3N2), for example, has an H3 hemagglutinin and an N2 neuraminidase. The established classification includes 18 HA subtypes, H1–H18, and 11 NA subtypes, N1–N11. These numbers describe antigenic categories, not individual strains; viruses within one subtype can differ genetically and in their recognition by antibodies. Subtypes are further divided into clades and subclades using phylogenetic relationships between viral gene sequences. (cdc.gov) Birds harbour 16 HA subtypes, H1–H16, and nine NA subtypes, N1–N9; the additional H17, H18, N10 and N11 categories were identified in bat-associated viruses rather than avian viruses. (wwwnc.cdc.gov)\n\nThe influenza A subtypes established in seasonal human circulation are A(H1N1) and A(H3N2). Contemporary seasonal H1N1 viruses descend from the virus responsible for the 2009 pandemic and are therefore commonly designated A(H1N1)pdm09, distinguishing them from the former seasonal H1N1 population. (cdc.gov) A subtype label alone does not specify host adaptation: H1N1 and H3N2 viruses also occur in pigs, and animal viruses sharing a subtype designation with human viruses need not belong to the same evolutionary lineage. Likewise, identifying a human infection with an animal-associated subtype does not establish sustained human-to-human circulation. (cdc.gov)\n\nInfluenza B is not classified into numbered HA–NA subtypes. Its two historically circulating lineages are B/Victoria and B/Yamagata, distinguished principally by genetic and antigenic differences in HA. (www-new.cdc.gov) B/Yamagata viruses have not been detected in surveillance after March 2020, whereas B/Victoria continues to circulate. (cdc.gov) In September 2023, the World Health Organization began recommending exclusion of the Yamagata component from seasonal vaccines; its September 2025 recommendations maintained that position. US vaccines for the 2026–2027 season consequently contain two influenza A components and one B/Victoria component. Absence of detection is distinct from definitive proof that a lineage is extinct. (cdn.who.int)\n\nHumans are the principal host of influenza B, but its host range is not exclusively human. Albert Osterhaus and colleagues reported in 2000 the isolation of B/Seal/Netherlands/1/99 from a naturally infected harbour seal; its sequence resembled human viruses circulating four to five years earlier. (pubmed.ncbi.nlm.nih.gov) Influenza C predominantly causes human respiratory infections, generally mild, and does not cause the large seasonal epidemics associated with A and B. (cdc.gov) Natural infection in pigs is also documented: Guo Yuanji and colleagues reported 15 isolates from Chinese pigs sampled in 1981. Influenza D was first isolated from an ill pig in Oklahoma in 2011, but cattle subsequently emerged as its principal reservoir; it has not been established as a cause of human illness. (citeseerx.ist.psu.edu)\n\nInfluenza A has the broadest established host range and is the only type known to cause human pandemics. Its animal reservoirs provide genetically and antigenically diverse viruses that can cross species barriers or exchange genome segments through reassortment, potentially producing viruses unfamiliar to human immunity. Pandemic spread additionally requires efficient transmission between humans. (cdc.gov) Type A also frequently dominates seasons with substantial severe disease, especially A(H3N2)-predominant seasons among older adults. This is a population-level pattern, not an absolute severity ranking: influenza B can also cause severe illness, hospitalization and death. (cdc.gov)"
    },
    {
      "heading": "Transmission and pathogenesis",
      "text": "Influenza spreads through virus-containing respiratory particles released during breathing, talking, coughing and sneezing. Larger droplets can deposit directly on another person’s nasal or oral mucosa, while finer aerosols can remain suspended and be inhaled. These routes overlap rather than having a sharp particle-size boundary, and their relative contributions vary with exposure conditions. Transmission through contaminated surfaces, or fomites, can occur when virus is transferred by the hands to the mouth, nose or possibly eyes, but is considered less common than respiratory transmission. In a 2018 study led by Jing Yan, 142 people with confirmed influenza provided samples during 218 visits; infectious virus was recovered from 39% of valid fine-aerosol cultures. Coughing was not necessary for infectious aerosol production, demonstrating that ordinary breathing can release viable virus into shared air. (cdc.gov)\n\nCrowding increases opportunities for close exposure, while poorly ventilated indoor spaces allow exhaled particles to accumulate. Temperature and humidity also influence transmission, through effects on particle evaporation, virus survival and host respiratory defenses. In 2007, Anice Lowen and colleagues conducted 20 guinea-pig transmission experiments at relative humidities of 20–80% and temperatures of 5, 20 or 30°C. Cold, dry conditions generally favored transmission between separated cages, whereas transmission was blocked at 30°C under the tested conditions. These experiments establish environmental effects in an animal model, not temperature or humidity thresholds that universally determine human transmission. (cdc.gov)\n\nEntry into respiratory cells depends partly on receptor recognition by hemagglutinin. Human-adapted influenza A viruses generally prefer glycans terminating in sialic acid linked to galactose through an \\(\\alpha 2,6\\) bond; avian-adapted viruses generally prefer \\(\\alpha 2,3\\)-linked structures. Human nasal and tracheal epithelia contain abundant \\(\\alpha 2,6\\)-linked receptors, whereas susceptible cells deeper in the lung include cells bearing \\(\\alpha 2,3\\)-linked receptors. Kyoko Shinya and colleagues documented this anatomical distinction in 2006, helping explain how an avian virus can infect human lung tissue without readily spreading between people. Receptor distribution is nevertheless more complex than a simple upper-versus-lower airway division, and receptor preference alone does not establish transmissibility: an H5N1 experimental variant with increased human-type receptor recognition transmitted by direct contact in ferrets but not through respiratory droplets. (pubmed.ncbi.nlm.nih.gov)\n\nInfection involves the epithelium lining the nose, trachea and bronchi, with extension into the lower respiratory tract in some cases. Viral replication injures epithelial cells and can cause loss of cilia, weakening the transport of mucus and trapped particles out of the airways. In a 2021 human-airway-on-a-chip study, influenza infection produced ciliary loss, disrupted intercellular junctions and compromised epithelial barrier function. Lower-airway infection can also involve alveolar epithelial cells, as demonstrated in experiments with primary human cells infected with H1N1 and H5N1 viruses. Such involvement provides a cellular basis for primary viral pneumonia, whose clinical consequences are addressed separately. (nature.com)\n\nSymptoms reflect both local tissue injury and host inflammation, rather than direct viral damage alone. In a 1998 experimental study, Frederick Hayden and colleagues followed 19 volunteers infected with influenza A/Texas/36/91 (H1N1). Nasal levels of the cytokine interleukin-6 and interferon-alpha peaked on day 2 and correlated with viral titers, body temperature, mucus production and symptom scores. Interleukin-8 peaked later, on days 4–6, and correlated with lower-respiratory symptoms. These observations connect inflammatory signaling with the timing of illness without implying that every symptom is attributable to a single mediator. (pmc.ncbi.nlm.nih.gov)\n\nFor seasonal influenza, the incubation period is usually about two days, with a range of one to four days. Viral shedding can begin approximately one day before symptoms and commonly continues for five to seven days after illness begins. Infectiousness is generally greatest during the first three symptomatic days; young children and immunocompromised people may remain contagious longer. Infection without recognized symptoms can also occur, and such individuals may transmit virus. Presymptomatic shedding therefore allows transmission before illness prompts a change in behavior. (cdc.gov)"
    },
    {
      "heading": "Clinical features and complications",
      "text": "Uncomplicated influenza typically begins abruptly with fever, chills, muscle aches (myalgia), headache, pronounced fatigue, a dry cough, and sore throat; nasal congestion or discharge may also occur. Most acute symptoms resolve within 3–7 days, although cough and malaise can persist for more than two weeks, particularly in older adults and people with chronic lung disease. Fever is not universal and may be absent in older or immunosuppressed people. Young children more often have nausea, vomiting, or diarrhea alongside respiratory symptoms, while frail older adults may present with confusion, behavioral changes, or loss of appetite. Compared with a common cold, influenza generally has a more sudden onset, more prominent systemic symptoms, and more severe aches and exhaustion; colds more often feature sneezing and nasal congestion. These patterns overlap, however, and symptoms alone cannot reliably distinguish influenza from other respiratory infections. (cdc.gov)\n\nPneumonia is a major severe complication and may result directly from influenza virus infection, from bacterial infection, or from both. Primary viral pneumonia can progress rapidly, with worsening breathlessness, diffuse pulmonary infiltrates, and respiratory failure. Secondary bacterial pneumonia may produce a biphasic course: initial improvement followed by renewed fever, productive cough, and pleuritic chest pain. Important bacterial pathogens include Streptococcus pneumoniae, Staphylococcus aureus—including methicillin-resistant strains—and Haemophilus influenzae. Viral destruction of respiratory epithelial cells contributes to loss of airway defenses. Either form of pneumonia can progress to acute respiratory distress syndrome, in which widespread lung injury severely impairs oxygen exchange, or to sepsis, shock, and multiorgan failure. These complications can be fatal. Less severe respiratory complications include sinusitis and middle-ear infection; children may also develop croup or bronchiolitis. (cdc.gov)\n\nInfluenza can exacerbate asthma, chronic obstructive pulmonary disease, heart failure, and diabetes, including precipitating diabetic ketoacidosis. Cardiac complications include myocardial infarction, myocarditis (inflammation of heart muscle), and pericarditis (inflammation of the surrounding sac). A 2020 study of 80,261 adults hospitalized with laboratory-confirmed influenza in the United States during the 2010–2011 through 2017–2018 seasons found that 11.7% had a coded acute cardiovascular event; acute heart failure and acute ischemic heart disease were the most frequent categories. These figures describe hospitalized patients, not the risk among all influenza infections. Neurological complications include seizures, encephalopathy, and encephalitis, while muscle complications include myositis and rhabdomyolysis. Reye syndrome, involving acute brain dysfunction and liver injury, is a rare complication associated with influenza and salicylate exposure in children. (cdc.gov)\n\nRisk of severe disease is elevated in children younger than five years, especially those younger than two, and adults aged 65 years or older. Other high-risk groups include pregnant people and those within two weeks postpartum; residents of long-term care facilities; people with chronic pulmonary, cardiovascular, kidney, liver, metabolic, hematological, or neurological disorders; and people with immune suppression caused by disease or medication. Extreme obesity, defined in adults by a body mass index of at least \\(40\\ \\mathrm{kg/m^2}\\), is also a recognized risk factor. Nevertheless, severe complications and deaths can occur in otherwise healthy people of any age. (cdc.gov)\n\nClinical markers of severe or progressive illness include difficulty breathing, rapid breathing, bluish lips or face, persistent chest pain, confusion or reduced alertness, seizures, markedly reduced urine output, and severe weakness. In children, chest-wall retractions indicate increased work of breathing, while absent tears, dry mouth, and no urine for eight hours are signs of dehydration. Severe muscle pain may make a child refuse to walk. Fever or cough that returns or worsens after improvement can indicate a complication, including secondary pneumonia; worsening of an underlying chronic condition is another important marker of progression. The absence of fever does not exclude severe influenza, particularly in older or immunosuppressed patients. (cdc.gov)"
    },
    {
      "heading": "Diagnosis",
      "text": "Clinical diagnosis combines the pattern of acute respiratory illness with evidence of influenza circulation, but symptoms alone do not reliably distinguish influenza from other respiratory infections. The World Health Organization surveillance definition of influenza-like illness, standardized in 2014, specifies an acute respiratory infection with measured fever of at least \\(38\\,^\\circ\\mathrm{C}\\), cough, and onset within the preceding ten days. This definition identifies a surveillance population rather than confirming the cause of illness or capturing every influenza infection. Laboratory testing establishes a more specific diagnosis; multiplex molecular assays can distinguish influenza A and B from other pathogens, including SARS-CoV-2 and respiratory syncytial virus. (who.int)\n\nReverse-transcription polymerase chain reaction (RT-PCR) detects influenza viral RNA in respiratory specimens. Reverse transcription produces DNA from the RNA template, after which amplification detects selected viral sequences. Depending on their targets, assays distinguish influenza A from B and identify seasonal A(H1N1)pdm09 or A(H3N2) subtypes; a test reporting “influenza A” does not necessarily establish its subtype. Laboratory molecular assays generally return results within approximately 45 minutes to several hours. Their high sensitivity makes them central to laboratory confirmation, although detection of RNA does not by itself demonstrate viable, infectious virus. Specimen timing and location remain important: upper-respiratory samples collected within four days of illness onset generally provide the best yield. In severe lower-respiratory disease, an upper-respiratory sample may be negative while a lower-respiratory specimen remains positive. Poor collection, inappropriate storage, or delays in processing can also produce false negatives. (cdc.gov)\n\nRapid influenza diagnostic tests detect viral antigens, rather than amplifying genetic material, and usually produce results in 10–15 minutes. Their sensitivity is substantially lower than that of molecular assays: CDC summaries describe approximately 50–70% sensitivity and 95–99% specificity, although performance varies by device and newer reader-based tests can perform better. False negatives occur when antigen concentrations fall below the detection threshold, particularly with late sampling or inadequate specimens. A negative result therefore does not exclude influenza. Predictive value also changes with prevalence: during intense influenza circulation, negative antigen results are less reassuring, whereas positive results are more likely to be false positives when circulation is low. Rapid molecular assays are a distinct point-of-care technology, typically returning results in 15–30 minutes with approximately 90–95% sensitivity. Thus, “rapid test” describes turnaround time, not a uniform diagnostic method or level of accuracy. (cdc.gov)\n\nViral culture recovers replicating virus and supplies isolates for detailed characterization, but its slower turnaround limits its role in immediate diagnosis. Shell-vial culture can yield results in one to three days, while conventional cell culture generally requires three to ten days. Culture supports antigenic testing and the characterization of strains considered for vaccine inclusion. Sequencing complements these methods by determining viral gene sequences, identifying genetic groups, and detecting mutations associated with altered viral properties or antiviral resistance. Phylogenetic comparisons place specimens alongside previously sampled viruses, supporting surveillance rather than merely confirming infection in an individual. Since 2014, CDC has used next-generation sequencing to expand whole-genome characterization and detect variation that older consensus-sequencing methods could miss. Genetic findings are interpreted alongside antigenic and other laboratory data. (cdc.gov)\n\nSerological testing measures the host antibody response and is principally used in research and public-health investigations. A single serum specimen cannot reliably establish recent influenza infection. Retrospective confirmation instead uses paired acute and convalescent specimens collected approximately three to four weeks apart, with a fourfold or greater rise in strain-specific antibody titre supporting recent infection. Because this interval extends beyond the period when an acute diagnostic result is needed, serology is unsuitable for routine clinical decision-making, although it can document infections retrospectively in epidemiological investigations. (cdc.gov)"
    },
    {
      "heading": "Immune response and immunity",
      "text": "Infected respiratory epithelial cells initiate innate immunity by detecting viral RNA. The cytoplasmic sensor RIG-I recognizes features including a 5′-triphosphate end and signals through the mitochondrial adaptor MAVS, activating transcription factors that induce interferons and inflammatory mediators. Type I interferons, principally IFN-α and IFN-β, and type III interferons, IFN-λ, bind receptors on infected and neighboring cells. Their JAK–STAT signaling pathways activate interferon-stimulated genes whose products inhibit viral entry, protein synthesis, and replication. IFN-λ acts especially at epithelial barriers. Single-cell experiments published in 2019 showed that infected and uninfected bystander epithelial cells participate differently in this response, with IFN-λ1 expression spreading through paracrine signaling. Influenza A’s NS1 protein counteracts several antiviral signaling mechanisms. (pmc.ncbi.nlm.nih.gov)\n\nResident macrophages and dendritic cells also detect infection; endosomal TLR7 recognizes viral single-stranded RNA, while activation of the NLRP3 inflammasome promotes maturation of IL-1β and IL-18. These and other cytokines, including IL-6 and tumor necrosis factor, coordinate inflammation and recruitment of additional immune cells. Natural killer cells help eliminate infected cells, and macrophages remove cellular debris. This response has a dual role: it restricts infection and supports tissue repair, but excessive or prolonged inflammatory activity can damage the respiratory epithelium and impair gas exchange. Dendritic cells carrying viral antigens migrate to draining lymph nodes, linking early detection to adaptive immunity. (pmc.ncbi.nlm.nih.gov)\n\nActivated B cells differentiate into antibody-secreting plasma cells and memory cells. Neutralizing antibodies against hemagglutinin (HA) are particularly important: antibodies to its globular head can block receptor binding, whereas some antibodies to its more conserved stem inhibit membrane fusion. Antibodies against neuraminidase can restrict viral release and spread rather than primarily preventing entry. Protection therefore cannot be reduced to a single antibody specificity. In a 2016 human H1N1 challenge study, Matthew Memoli and colleagues found that baseline anti-HA and anti-neuraminidase responses provided complementary information about protection, with neuraminidase-inhibiting antibodies independently associated with reduced disease measures. (pubmed.ncbi.nlm.nih.gov)\n\nCell-mediated immunity recognizes infected cells rather than free virions. CD8-positive T cells detect viral peptides presented by major histocompatibility complex class I molecules and kill infected cells through cytotoxic mechanisms. Many recognized peptides derive from relatively conserved internal proteins, including nucleoprotein and matrix protein M1, allowing some recognition across influenza A subtypes. CD4-positive T cells support antibody production and other antiviral responses. In a study published in 2013, Saranya Sridhar and colleagues followed 342 healthy adults through the United Kingdom’s 2009 pandemic waves. Among infected participants lacking cross-reactive neutralizing antibodies, higher pre-existing responses to conserved CD8 T-cell epitopes correlated with milder illness. Such responses can limit disease without necessarily preventing infection. (nature.com)\n\nProtection against reinfection is often strain-specific because prominent antibody targets on the HA head vary between viruses. Changes in those targets can reduce recognition despite persisting immunological memory. Protection after vaccination also declines over time, but influenza immunity is not uniformly short-lived: antibody concentrations, memory-cell persistence, and protection against a changed virus are distinct properties. In 2008, Xiaocong Yu and colleagues studied 32 survivors of the 1918 pandemic, aged 91–101. All had serum antibodies reactive with the reconstructed 1918 virus, and memory B cells isolated from some participants yielded neutralizing antibodies, demonstrating that components of influenza-specific memory can persist for decades. (cdc.gov)\n\nPrevious infection or vaccination thus shapes both the speed and specificity of subsequent responses. Early exposures can establish immune imprinting, favoring responses to previously encountered antigens. A 2016 analysis by Katelyn Gostic and colleagues associated childhood exposure to HA group 1 viruses, including H1 and H2, with protection against severe H5N1 disease, and exposure to group 2 H3 viruses with protection against severe H7N9 disease. Its statistical models estimated approximately 75% protection against severe infection and 80% against death associated with matching HA-group imprinting. These were population-level estimates for the studied zoonotic infections, not evidence of complete protection against infection or identical effects for every seasonal strain. (pmc.ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Antigenic drift and antigenic shift",
      "text": "Antigenic drift is the progressive alteration of influenza surface antigens through point mutations in the genes encoding hemagglutinin (HA) and neuraminidase (NA). It occurs in both influenza A and B viruses. Mutations that alter these proteins can reduce recognition by antibodies generated against earlier strains, allowing partial escape from existing immunity. Closely related variants often retain substantial cross-protection; drift therefore does not imply complete loss of immunity. Nevertheless, accumulated changes can make previously infected or vaccinated people susceptible again, helping sustain recurrent annual epidemics. This is also a principal reason for periodically updating seasonal vaccine composition. (cdc.gov)\n\nGenetic change and antigenic change are not equivalent: mutations differ greatly in their effects on antibody recognition, and a single amino-acid substitution at an important HA site can have a larger effect than several substitutions elsewhere. In 2013, Björn F. Koel and colleagues showed that major antigenic changes in human A(H3N2) viruses between 1968 and 2003 were caused mainly by substitutions at only seven HA positions immediately adjacent to the receptor-binding site. Thus, although drift involves ongoing sequence evolution, its measurable antigenic consequences can be punctuated rather than uniformly gradual. (pubmed.ncbi.nlm.nih.gov)\n\nAntigenic shift is an abrupt, substantial change in influenza A antigens encountered by humans. A central mechanism is reassortment: when two influenza viruses infect the same cell, progeny can inherit different combinations of their eight RNA genome segments. This exchanges whole segments rather than merely adding point mutations within a segment. Reassortment can introduce an antigenically unfamiliar HA, with or without a different NA, into a virus capable of infecting humans. However, reassortment is not synonymous with shift: exchanging segments does not necessarily produce a major antigenic change. A pandemic additionally requires efficient, sustained human-to-human transmission; antigenic novelty alone is insufficient. Shift can also involve an animal-origin virus entering human circulation without a newly occurring reassortment event. (cdc.gov)\n\nThe 2009 pandemic illustrates why shift need not introduce a new subtype designation. The emerging virus was H1N1, as was one of the previously circulating seasonal viruses, but its animal-origin antigens were substantially different. Six genome segments came from a North American swine-virus lineage and two from a Eurasian swine-virus lineage; the former already contained genes with swine, avian and human ancestry. Its emergence therefore represented a major antigenic discontinuity within the H1N1 subtype rather than a change from one H-number to another. (archive.cdc.gov)\n\nAntigenic change is measured by comparing viruses’ reactions with reference antisera. In the hemagglutination-inhibition (HI) assay, antibodies block HA-mediated binding and agglutination of red blood cells. The titer is the reciprocal of the greatest serum dilution that still prevents agglutination. Using the same antiserum, a lower titer against a test virus than against the reference virus indicates reduced recognition. CDC considers viruses antigenically similar when their HI titers differ by no more than two twofold dilution steps—a fourfold difference. Neutralization assays provide complementary measurements of antibodies’ ability to inhibit infection, particularly for A(H3N2) viruses with altered red-cell-binding properties. (cdc.gov)\n\nSequence-based analysis identifies amino-acid substitutions, assigns genetic groups through phylogenetic comparisons, and tracks segment ancestry. These data are interpreted alongside antibody assays because sequence similarity alone does not establish antigenic similarity. Antigenic cartography converts patterns of serum–virus reactivity into maps of antigenic relationships. In 2004, Derek J. Smith and colleagues mapped A(H3N2) evolution from 1968 to 2003, showing that antigenic evolution was more punctuated than genetic evolution and that some genetic changes had disproportionately large antigenic effects. (cdc.gov)"
    },
    {
      "heading": "Seasonal epidemiology and burden",
      "text": "In temperate regions, influenza epidemics usually occur during winter: approximately November–March in the Northern Hemisphere and May–September in the Southern Hemisphere. These are broad seasonal windows rather than fixed boundaries. Tropical and subtropical regions have more varied patterns, including rainy-season peaks, two annual peaks, or continuing transmission throughout the year. A 2013 analysis by James Tamerius and colleagues, using laboratory-confirmed surveillance data from 78 locations, identified two broad climatic settings associated with epidemics: cold–dry conditions at higher latitudes and humid–rainy conditions in many tropical locations. Thus, a single winter-based calendar does not describe global influenza epidemiology. (journals.plos.org)\n\nTemperature and humidity can influence viral survival and the behavior of respiratory particles, while school attendance and indoor crowding alter contact opportunities. In 2007, Anice Lowen and colleagues experimentally demonstrated that airborne transmission between guinea pigs was more efficient at 5°C than at 20°C; at 20°C, transmission was particularly efficient at relative humidities of 20–35%, but absent at 80% in those experiments. These results establish environmental effects in an animal model, not universal thresholds for human epidemics. A subsequent particle-dynamics study showed how humidity affects evaporation, settling and viral inactivation. Climatic associations nevertheless differ geographically, and rainy-season peaks cannot simply be attributed to the same cold–dry mechanism observed in temperate winters. (journals.plos.org)\n\nThe World Health Organization estimates approximately one billion seasonal influenza cases annually, including 3–5 million severe illnesses, and 290,000–650,000 influenza-associated respiratory deaths. These figures describe different outcomes: severe illness is not synonymous with hospital admission, and respiratory mortality excludes some non-respiratory consequences. Hospitalization estimates therefore require their own definitions and population denominators. For example, a systematic review and modelling study of children younger than five estimated that, in 2018, influenza was associated with 109.5 million illness episodes, 10.1 million acute lower respiratory infection episodes and 870,000 hospital admissions worldwide. The hospitalization estimate had an uncertainty range of 543,000–1,415,000 and applies specifically to influenza-associated lower respiratory infections in this age group. (who.int)\n\nBurden estimates are not simple counts of positive laboratory tests. Many infections receive no testing, and influenza may no longer be detectable when a patient dies from a subsequent complication. A Danielle Iuliano and colleagues’ global mortality study, published in 2018, fitted time-series models to respiratory death records and influenza surveillance data from 33 countries, then extrapolated to countries without comparable data. This approach estimates influenza-associated excess mortality above an expected baseline. Hospitalization models also adjust surveillance counts for testing frequency and diagnostic sensitivity; changes in these inputs can revise historical estimates. For the United States’ 2017–2018 season, improved information on testing reduced the estimated degree of under-detection and consequently lowered the final burden estimates. (pubmed.ncbi.nlm.nih.gov)\n\nSevere outcomes are concentrated among young children, older adults, pregnant people and people with chronic disease or immunosuppression, although their relative contributions differ between populations. In the United States during 2017–2018, adults aged 65 or older accounted for an estimated 66% of influenza hospitalizations and 83% of deaths. Geographic inequalities are especially pronounced in childhood mortality: WHO reports that 99% of deaths among children younger than five with influenza-related lower respiratory infections occur in developing countries. Comparisons between countries must distinguish underlying disease burden from differences in testing, access to hospital care and completeness of death registration. (who.int)\n\nYear-to-year variation can be substantial. The predominantly A(H3N2) United States season of 2017–2018 produced an estimated 41 million illnesses, 710,000 hospitalizations and 52,000 deaths. In contrast, during October 2020–May 2021, laboratory-confirmed influenza hospitalization incidence in the US surveillance network was only 0.8 per 100,000 population, compared with 62.0–102.9 in the preceding four seasons. This exceptional suppression coincided with COVID-19-related changes in masking, distancing, travel and other behavior; their individual contributions could not be separated. Such laboratory-confirmed surveillance rates should not be treated as directly equivalent to modelled national hospitalization totals. (archive.cdc.gov)"
    },
    {
      "heading": "Pandemics in history",
      "text": "Influenza pandemics occur when an influenza A virus that is antigenically unfamiliar to much of the population acquires the ability to spread efficiently between people. The 1957 and 1968 pandemics introduced new hemagglutinin subtypes, H2 and H3, through reassortment between avian and human viruses. A new subtype designation is not essential, however: the 2009 pandemic virus was H1N1, like an already circulating seasonal virus, but differed sufficiently in its antigens that seasonal vaccination offered little cross-protection. Population susceptibility therefore depends on antigenic differences and previous exposure, not simply on the H–N label. (archive.cdc.gov)\n\nThe 1918–1919 pandemic, caused by influenza A(H1N1), infected an estimated 500 million people, approximately one-third of the contemporary world population. A widely cited estimate places worldwide deaths at about 50 million; CDC gives approximately 675,000 deaths in the United States. These are retrospective estimates rather than complete recorded counts. Successive waves occurred in spring 1918, autumn 1918, and early 1919, with the autumn wave particularly lethal. Its distinctive age pattern included high mortality among adults approximately 20–40 years old, producing a “W-shaped” age-specific mortality curve alongside the usual vulnerability of infants and older people. There was no effective influenza vaccine and no antibiotic treatment for secondary bacterial infections. Responses consequently relied on isolation, quarantine, hygiene measures, and restrictions on public gatherings, applied unevenly between communities. (archive.cdc.gov)\n\nThe 1957–1958 pandemic, historically called “Asian flu,” involved A(H2N2). Its virus acquired three avian-origin gene segments, including those encoding H2 hemagglutinin and N2 neuraminidase. It emerged in East Asia in early 1957, was reported in Hong Kong in April, and reached coastal United States cities that summer. CDC estimates approximately 1.1 million deaths worldwide and 116,000 in the United States. Unlike 1918, deaths in the United States were concentrated predominantly among older adults, although infection rates were highest among schoolchildren and adults aged 40 or younger. Laboratory surveillance enabled rapid identification of the antigenic change. Maurice Hilleman, then at the Walter Reed Army Institute of Research, helped characterize the new virus and initiate development of a strain-specific vaccine; military and civilian authorities organized immunization programmes during 1957. (archive.cdc.gov)\n\nThe 1968 pandemic, historically called “Hong Kong flu,” was first detected in Hong Kong in July and reached the United States by September. The A(H3N2) virus acquired two avian-origin gene segments, including a new H3 hemagglutinin, while retaining the N2 neuraminidase of the preceding H2N2 virus. Previous N2 exposure appears to have moderated disease, although it did not prevent worldwide spread. CDC estimates approximately one million deaths globally, with most excess deaths among people aged 65 or older. Timing varied geographically: the United States experienced substantial mortality in 1968–1969, whereas several western European countries had greater mortality in the following season. The United States Military Commission on Influenza obtained virus strains and recommended a new vaccine in July 1968; Canadian trials during winter 1968–1969 likewise evaluated updated vaccines, but supplies were limited. (cdc.gov)\n\nThe 2009–2010 pandemic involved a swine-origin A(H1N1)pdm09 virus identified in California in April 2009 amid outbreaks in North America. The World Health Organization declared a pandemic on June 11, 2009. A modelling study estimated 151,700–575,400 associated deaths worldwide during the virus’s first year, substantially fewer than in 1918; approximately 80% occurred in people younger than 65. Some older adults possessed cross-reactive antibodies from earlier H1N1 exposure, contributing to the younger age distribution of illness. The United States response included expanded laboratory surveillance, antiviral distribution, and a dedicated vaccination campaign whose first doses were administered on October 5, 2009. Large quantities of vaccine did not become available until late November, after the autumn peak of illness in the United States. WHO announced the transition to the post-pandemic period on August 10, 2010. (cdc.gov)"
    },
    {
      "heading": "Animal reservoirs and zoonotic spillover",
      "text": "Wild aquatic birds, especially ducks, geese and shorebirds, form the principal natural reservoir of avian influenza A viruses. Many low-pathogenicity viruses replicate in the intestinal tract without causing conspicuous illness and pass between birds through fecally contaminated water. This reservoir maintains viral diversity and provides opportunities for transmission into domestic poultry. Farms and live-bird markets bring susceptible birds into close contact, allowing amplification and exchange of viruses among species. “Highly pathogenic avian influenza” (HPAI) describes pathogenicity in poultry, particularly chickens; it does not directly predict disease severity in humans or every wild-bird species. Unlike many low-pathogenicity infections, HPAI H5N1 can cause systemic disease and extensive mortality in wild birds. (wwwnc.cdc.gov)\n\nSwine can acquire influenza A viruses from other pigs, humans and birds. If different viruses infect the same cell, their segmented genomes can undergo reassortment, producing progeny with combinations of gene segments from different parental viruses. This susceptibility underlies the description of pigs as “mixing vessels,” although pigs are not a necessary intermediary for avian-to-human transmission: H5N1 and H9N2 have crossed directly from poultry into humans. Horses and dogs also maintain their own influenza viruses, while avian viruses occasionally infect terrestrial and marine mammals. A zoonotic infection therefore represents a crossing of the species barrier, not necessarily the establishment of a virus capable of sustained transmission in its new host. (cdc.gov)\n\nThe goose/Guangdong lineage of avian influenza H5N1 was identified in domestic waterfowl in southern China in 1996. In Hong Kong in 1997, poultry outbreaks accompanied the first recognized human H5N1 infections: 18 laboratory-confirmed cases, including six deaths. The local outbreak was controlled, but related viruses persisted in birds. From 2003, H5N1 spread widely through Asian poultry populations and subsequently into Africa, Europe and the Middle East. By September 6, 2026, the World Health Organization had recorded 1,003 laboratory-confirmed human cases from 25 countries since 2003, including 479 deaths—a reported case-fatality proportion of 47.8%. This proportion concerns detected cases, not all infections; reported illness has ranged from asymptomatic or mild infection to fatal disease. Most human infections have followed exposure to infected animals or contaminated environments. (archive.cdc.gov)\n\nDiversification of the H5 hemagglutinin gene and reassortment produced several descendant lineages and neuraminidase combinations. Clade 2.3.4.4b H5N1 became predominant across much of Asia, Africa, Europe and the Middle East by late 2021 and reached North American wild birds that year; United States poultry outbreaks followed in February 2022. Its spread has involved both wild and domestic birds and repeated spillovers into mammals, including domestic cats, wild carnivores and seals. Detections extended to the Antarctic peninsula and sub-Antarctic islands during the 2023–2024 austral summer. These events distinguish a geographically widespread animal epidemic from isolated introductions into individual farms or human patients. (archive.cdc.gov)\n\nOn March 25, 2024, H5N1 infection was reported in United States dairy cattle, marking the first documented outbreak in this host. Experimental and farm investigations demonstrated infection of mammary-gland tissue, high concentrations of infectious virus in milk and reduced milk production. Evidence indicates that contaminated milk and milking procedures are important routes of cattle-to-cattle transmission, rather than efficient respiratory spread alone. By March 1, 2026, 1,088 dairy herds in 19 states had tested positive, and genetic analyses indicated at least four independent introductions into cattle. Dairy-worker infections established an additional animal-to-human pathway; documented cases included conjunctivitis and mild respiratory illness, with virus isolated from the eyes of an infected Texas worker. (nature.com)\n\nRisk assessments distinguish occupational exposure from risk to the wider population. The joint FAO–WHO–WOAH assessment published on May 18, 2026, classified the global public-health risk from circulating HPAI A(H5) viruses as low, while assessing infection risk for occupationally or frequently exposed people as low to moderate, depending on local conditions and protective measures. WHO’s October 2, 2026 update reported no evidence of sustained human-to-human transmission associated with recent A(H5) infections and judged such transmission currently unlikely. Thus, repeated mammalian infections and farm-worker cases document opportunities for spillover, but do not themselves demonstrate efficient transmission between humans. (who.int)"
    },
    {
      "heading": "Prevention and vaccines",
      "text": "Seasonal influenza vaccines principally induce antibodies against hemagglutinin (HA), the viral surface protein used to enter host cells. Inactivated vaccines contain noninfectious viral material produced by growing influenza viruses in embryonated chicken eggs or mammalian cell cultures, followed by inactivation and purification. They are generally injected intramuscularly; several products are licensed from six months of age. Live attenuated vaccines instead contain weakened, cold-adapted viruses administered as a nasal spray, with replication restricted by their temperature-sensitive properties. In the United States, this formulation is licensed for ages 2–49 years and is not used during pregnancy or in immunocompromised people. Recombinant vaccines contain purified HA produced in insect cells using a baculovirus expression system, without growing influenza virus or using eggs. The US recombinant product Flublok is licensed from nine years of age. (cdc.gov)\n\nMessenger RNA vaccines deliver genetic instructions for HA in lipid nanoparticles, allowing recipient cells to produce the antigen. On August 5, 2026, the US Food and Drug Administration approved Moderna’s MFLUSIVA for adults aged 50 years and older. Its indication for those aged 65 and older received accelerated approval based on immune responses, with continued approval potentially dependent on confirmation of clinical benefit. This platform replaces viral propagation with RNA synthesis; seasonal reformulation changes the RNA sequences encoding the selected HA proteins while retaining the delivery platform and manufacturing process. (fda.gov)\n\nVaccination programmes differ in their target populations. The World Health Organization (WHO), in its 2022 position paper, identified health workers, older adults, pregnant individuals and people with underlying conditions as priority groups, while retaining children as an important target according to national disease burden and resources. The United States has recommended annual vaccination for everyone aged six months or older, with rare exceptions, since the 2010–2011 season. For adults aged 65 or older, US guidance preferentially identifies high-dose inactivated, adjuvanted inactivated or recombinant vaccines. High-dose formulations contain 60 micrograms of HA per component, compared with 15 micrograms in standard-dose injections; an adjuvant enhances the immune response without increasing the antigen dose. Protective antibodies generally develop approximately two weeks after vaccination. (who.int)\n\nWHO convenes vaccine-composition consultations twice annually: in February for the following northern-hemisphere season and September for the southern-hemisphere season. Experts assess the genetic and antigenic properties of circulating viruses, their geographic spread, and evidence about immune recognition; national regulators subsequently determine licensed formulations. Selection precedes distribution by several months to accommodate production and testing. Seasonal trivalent vaccines include A(H1N1)pdm09, A(H3N2) and B/Victoria components. WHO recommended removing B/Yamagata in September 2023, and US vaccines returned to trivalent formulations for 2024–2025. Egg-based and non-egg-based formulations can use different representative strains: for example, the 2026–2027 US H3N2 components are A/Darwin/1454/2025-like and A/Darwin/1415/2025-like viruses, respectively. (iris.who.int)\n\nVaccine effectiveness varies with antigenic matching, viral subtype, recipient age and health, vaccine formulation, and time since vaccination. When vaccine and circulating viruses are well matched, vaccination has reduced medically attended influenza by approximately 40–60%; this figure is not a fixed estimate for every season or outcome. Antigenic drift after strain selection can reduce recognition by vaccine-induced antibodies. Growth in eggs can also select HA changes, particularly in H3N2 viruses, that make vaccine antigens less representative of circulating viruses. Cell-based and recombinant production avoid egg adaptation, but do not eliminate mismatch or declining immunity. (cdc.gov)\n\nPandemic preparedness combines candidate vaccine viruses, advance manufacturing arrangements and stockpiles against potentially pandemic subtypes. In May 2024, the US Biomedical Advanced Research and Development Authority arranged to fill approximately 4.8 million doses of stockpiled A/Astrakhan H5N8 vaccine into vials. Adjuvants such as MF59 and AS03 support antigen-sparing, increasing the number of doses obtainable from a given antigen supply. Audenz, an adjuvanted H5N1 vaccine initially approved in 2020 for people aged six months or older at increased exposure risk, uses two doses 21 days apart. Advance production can shorten deployment delays, whereas producing a newly matched vaccine still requires manufacturing, testing and regulatory review: WHO’s August 2009 account estimated five to six months from identification and isolation of a novel strain to initial approved supplies using conventional production. (aspr.gov)"
    },
    {
      "heading": "Treatment and antiviral drugs",
      "text": "Neuraminidase inhibitors act against influenza A and B by blocking the viral enzyme that removes terminal sialic acid residues, thereby restricting the release and spread of newly formed virions. Oseltamivir is administered orally, zanamivir by inhalation, and peramivir intravenously. For uncomplicated influenza, standard treatment comprises five days of oseltamivir or zanamivir, or a single infusion of peramivir. Inhaled zanamivir is contraindicated in patients with underlying airway disease, including asthma, because of bronchospasm risk. The older adamantanes, amantadine and rimantadine, block the M2 ion channel involved in viral uncoating and act only against influenza A; widespread resistance has made them unsuitable for seasonal treatment in the United States. (ncbi.nlm.nih.gov)\n\nBaloxavir marboxil, approved in Japan and the United States in 2018, targets viral transcription rather than virion release. Its active metabolite, baloxavir acid, inhibits the cap-dependent endonuclease in the polymerase acidic (PA) subunit. This prevents “cap-snatching,” in which capped fragments of host messenger RNA provide primers for viral RNA synthesis. Baloxavir is administered as a single oral dose for uncomplicated influenza. In Frederick Hayden and colleagues’ 2018 randomized controlled trial, median symptom duration was 53.7 hours with baloxavir and 80.2 hours with placebo, and was similar with baloxavir and oseltamivir. Baloxavir reduced viral load more rapidly than oseltamivir, but this virological advantage did not produce faster symptom resolution. (pubmed.ncbi.nlm.nih.gov)\n\nAntiviral benefit is greatest when treatment begins within 48 hours of symptom onset. Evidence for uncomplicated influenza most consistently demonstrates modest shortening of illness: Joanna Dobson and colleagues’ 2015 meta-analysis of nine oseltamivir trials, involving 4,328 adults, found a 25.2-hour reduction among participants with laboratory-confirmed influenza. It also reported fewer lower respiratory complications requiring antibiotics and fewer hospital admissions. However, Tom Jefferson and colleagues’ 2014 review found a 16.8-hour symptom reduction in adults, no reduction in admissions, and limitations in the documentation of pneumonia outcomes. Oseltamivir increased nausea and vomiting. Evidence for reduced mortality in hospitalized patients is principally observational: Stella Muthuri and colleagues’ 2014 analysis of hospitalized patients with pandemic H1N1 associated neuraminidase-inhibitor treatment with lower mortality, but such associations remain vulnerable to confounding. (pubmed.ncbi.nlm.nih.gov)\n\nRecommendations distinguish severe disease from uncomplicated outpatient illness and differ between authorities. US CDC guidance prioritizes hospitalized patients, those with severe or progressive illness, and people at increased complication risk, including young children, older adults, pregnant patients, and people with chronic disease or immunosuppression. Treatment in these groups is not delayed for laboratory confirmation; hospitalized or severe cases may benefit even when treatment begins after 48 hours. Oral or enterically administered oseltamivir is the preferred hospitalized-patient treatment. The World Health Organization’s 2024 guideline conditionally recommends oseltamivir for severe influenza and baloxavir for non-severe influenza at high risk of progression, but recommends against oseltamivir for non-severe influenza. These differences reflect assessments of benefit magnitude, adverse effects, and evidence certainty. (cdc.gov)\n\nResistance can arise through spontaneous mutations or selection during treatment. In A(H1N1)pdm09, the neuraminidase substitution H275Y markedly reduces susceptibility to oseltamivir and peramivir, while susceptibility to zanamivir can remain intact. Baloxavir resistance commonly involves PA substitutions I38T, I38M, or I38F; clinical trials have documented treatment-emergent variants and viral rebound. Surveillance therefore combines sequencing of resistance-associated sites with functional testing: neuraminidase-inhibition assays assess neuraminidase inhibitors, while cell-based plaque- or focus-reduction assays assess baloxavir. This distinction matters because a genetic marker or reduced laboratory susceptibility does not invariably establish clinical treatment failure. CDC reported in June 2026 that nearly all recently circulating seasonal influenza A and B viruses remained susceptible to neuraminidase inhibitors and baloxavir, while nearly all seasonal influenza A viruses remained resistant to adamantanes. (cdc.gov)"
    },
    {
      "heading": "Surveillance and public health response",
      "text": "The Global Influenza Surveillance and Response System (GISRS), coordinated by the World Health Organization (WHO), has operated since 1952. National Influenza Centres, designated by health ministries and recognized by WHO, analyse specimens throughout the year and send representative clinical samples and virus isolates to WHO Collaborating Centres for detailed genetic and antigenic characterization. This division of work connects national detection with international assessment of circulating strains and emerging threats; its results also inform seasonal vaccine composition. FluNet, launched in 1997, publishes weekly country-level virological data, including detections by type and subtype. FluID complements these results with epidemiological surveillance of respiratory illness, allowing WHO to interpret virus circulation alongside disease activity. (who.int)\n\nSentinel surveillance uses selected outpatient practices and hospitals linked to laboratories rather than attempting to test every infection. WHO’s standardized influenza-like illness (ILI) definition requires an acute respiratory infection with measured temperature of at least \\(38^\\circ\\mathrm{C}\\), cough, and onset within the preceding ten days. Severe acute respiratory infection (SARI) surveillance uses a similar definition, permits a history of fever, and requires hospitalization. These syndromes are not specific to influenza: laboratory testing identifies the proportion attributable to influenza viruses. Consistent case definitions and sampling support comparison of trends over time and between countries; they are not intended to capture all cases. (who.int)\n\nNon-pharmaceutical responses act on several opportunities for transmission. Handwashing removes contamination from hands; respiratory etiquette, such as covering coughs and sneezes with tissues or an elbow, limits dispersal of respiratory secretions; staying home when ill reduces contacts during symptomatic infection. Policies supporting absence from school or work make isolation practicable, while remote meetings and changes to workplace contact patterns can reduce exposure. Evidence differs by intervention and setting. In a 12-week cluster-randomized trial conducted in 60 Cairo elementary schools in 2008, Maha Talaat and colleagues combined twice-daily handwashing with hygiene education. Their 2011 report described 47% fewer laboratory-confirmed influenza episodes and 40% fewer ILI-related absences in intervention schools. The study was unblinded, used rapid antigen tests, and did not establish whether the effect persisted beyond the observation period. (cdc.gov)\n\nSchool closures reduce contacts among pupils, but timing and behaviour outside school influence their effects. The United States’ 2017 community-mitigation guidelines distinguished early, coordinated dismissals intended to interrupt transmission from reactive closures caused by extensive absenteeism. They cited a four-day reactive closure in Kentucky that did not reduce community ILI, and little effect from closures of 559 Michigan schools during the autumn 2009 pandemic wave. Closure decisions also involve interrupted education, childcare needs and parental work absence. WHO’s guideline published on 23 April 2026 evaluates public-health and social measures through systematic reviews and the GRADE framework, explicitly incorporating effectiveness, feasibility, equity and socioeconomic consequences rather than treating every measure as appropriate for every outbreak. (cdc.gov)\n\nNational pandemic plans specify coordination arrangements, surveillance and risk-assessment capacity, communication, continuity of essential services, and response operations. WHO’s 2017 risk-management guidance incorporated lessons from the 2009 pandemic; its 2018 simulation-exercise manual describes testing plans and using identified weaknesses to revise them. Animal and human surveillance are linked through collaboration among WHO, the Food and Agriculture Organization and the World Organisation for Animal Health. Their joint work monitors zoonotic influenza and assesses emerging viruses; WHO’s Tool for Influenza Pandemic Risk Assessment separately considers the likelihood of sustained human-to-human transmission and the potential impact of a pandemic. Investigations of suspected human spillover include animal-exposure histories and contact tracing to identify possible onward transmission. (who.int)\n\nUnder the International Health Regulations (2005), human infection with a novel influenza A subtype is internationally notifiable. States Parties must notify WHO immediately, within 24 hours, of a laboratory-confirmed recent human infection with such a virus; symptomatic illness is not required for notification. Reporting therefore does not depend on detecting a large outbreak or demonstrating sustained transmission. For example, Mexico’s national IHR focal point notified WHO on 23 May 2024 of the first laboratory-confirmed human A(H5N2) infection reported globally, initiating international information sharing alongside the national investigation. (who.int)"
    },
    {
      "heading": "Research history and key discoveries",
      "text": "In 1933, Wilson Smith, Christopher Andrewes and Patrick Laidlaw at Britain’s National Institute for Medical Research isolated the agent of human influenza using ferrets. They passed patients’ throat washings through filters impermeable to bacteria and introduced the filtrates into ferrets’ noses. The animals developed a febrile respiratory illness that could be transmitted to other ferrets, establishing an experimental system for studying the causative virus. Their report, published in The Lancet on 8 July 1933, provided direct evidence against bacterial explanations of influenza and identified the agent subsequently classified as influenza A. (sciencedirect.com)\n\nPropagation outside infected animals made controlled experiments and vaccine production more practicable. Thomas Francis Jr. and Thomas Magill reported cultivation of human influenza virus in tissue culture in 1935; their subsequent experiments demonstrated continued infectivity and the induction of neutralizing antibodies. These “artificial media” contained living tissue, rather than supporting viral growth independently of cells. During the mid-1930s, Frank Macfarlane Burnet developed propagation in embryonated chicken eggs, using the chorioallantoic membrane and subsequently other inoculation routes. Egg cultivation supplied larger quantities of virus for laboratory investigation and helped establish the production methods used by Francis and Jonas Salk in developing inactivated influenza vaccines, licensed for wider use in 1945. (pubmed.ncbi.nlm.nih.gov)\n\nInvestigation of the 1918 pandemic required recovering genetic information rather than isolating surviving infectious virus. Beginning in 1995, Jeffery Taubenberger and colleagues investigated archived autopsy material; the first viral sequences were reported in 1997. Sources included formalin-fixed lung tissue from American soldiers and frozen lung tissue obtained that year by Johan Hultin from a victim buried at Brevig Mission, Alaska. Reverse transcription and polymerase chain reaction enabled fragmented viral RNA to be analysed, and the coding sequences of all eight genome segments were completed by 2005. Terrence Tumpey and collaborators then reconstructed the virus through reverse genetics under enhanced biosafety-level-3 containment—not by reviving infectious material from the preserved bodies. (pmc.ncbi.nlm.nih.gov)\n\nThe reconstruction enabled experimental comparison with contemporary seasonal H1N1 viruses. In the 2005 study, the reconstructed virus replicated vigorously in human bronchial epithelial cells and caused lethal infection in mice without prior mouse adaptation. Comparisons with viruses carrying different combinations of 1918 and contemporary genes implicated both hemagglutinin and the polymerase machinery in its unusually high virulence. The results supported a multigenic explanation: coordinated activity across the viral genome, rather than a single exceptional mutation, contributed to the phenotype. These experiments identified properties of the pandemic virus but did not make animal-model lethality a direct measure of mortality in human populations. (cmgm-new.stanford.edu)\n\nAtomic-resolution studies established a structural basis for influenza’s surface functions. In 1981, Ian Wilson, John Skehel and Don Wiley determined the hemagglutinin structure at 3 Å resolution, revealing a three-subunit protein with receptor-binding and variable antigenic regions above a helical stem. In 1994, Per Bullough and colleagues resolved a low-pH conformation showing extensive refolding, including displacement of the fusion peptide by at least 100 Å, explaining how acidification activates membrane fusion. Joseph Varghese, Graeme Laver and Peter Colman determined the neuraminidase structure at 2.9 Å in 1983. Its catalytic architecture subsequently enabled structure-guided inhibitor design, exemplified by Mark von Itzstein and colleagues’ 1993 experiments with potent inhibitors of the enzyme and viral replication. (nature.com)\n\nSequencing expanded from individual genes to comparative whole-genome analysis. In 2005, Elodie Ghedin and colleagues reported 209 complete influenza A genomes, comprising 207 H3N2 and two H1N2 isolates and approximately 2.82 million nucleotides. Examining all eight segments exposed genetic diversity and segment exchange that studies restricted to surface-protein genes could miss. Structural analysis likewise extended to larger molecular assemblies: in 2014, Alexander Pflug and colleagues determined the complete bat influenza A polymerase bound to its RNA promoter, locating the PB1 catalytic core, PA endonuclease and PB2 cap-binding domain within one complex. This connected previously separate biochemical findings to an atomic model of RNA synthesis and cap-snatching. (nature.com)"
    },
    {
      "heading": "Economic and social impact",
      "text": "Influenza generates direct medical expenditure and indirect losses of productivity. Direct costs include consultations, diagnostic tests, medicines, emergency care and hospitalization; indirect costs include time away from paid work and time spent caring for sick children or other dependants. Estimates depend on the seasons studied, local prices and wages, and whether premature deaths are assigned a monetary value. A 2018 analysis by Wayan C. W. S. Putri and colleagues estimated the average annual United States burden at US$11.2 billion, comprising US$3.2 billion in direct medical costs and US$8.0 billion in indirect costs. Its underlying estimates included approximately 247,000 hospitalizations and 20.1 million days of lost productivity annually. These were model-based annual averages, not expenditure measured in a single outbreak. (pubmed.ncbi.nlm.nih.gov)\n\nPandemic losses extend beyond healthcare and absenteeism because infection-control measures and changes in consumer behaviour reduce activity in travel, hospitality and other services. World Bank simulations developed in 2006 and 2008 projected losses ranging from 0.7% to 4.8% of global gross domestic product, depending on pandemic severity; these were hypothetical scenarios rather than measured losses from the 2009 pandemic. In Mexico, the 2009 H1N1 outbreak coincided with a recession already driven by the global financial crisis. A subsequent World Bank assessment estimated an additional reduction in economic activity of 0.3–0.5% of GDP, approximately US$2.7–4.5 billion. International tourism revenue in the second and third quarters of 2009 was US$1.5 billion below the corresponding 2008 total, a decline the assessment attributed largely to influenza. (documents1.worldbank.org)\n\nSchools transmit the social costs of outbreaks to households through interrupted teaching and additional childcare requirements. During New York City’s spring 2009 response, officials weighed school closure against parental work loss and disruption of free meals and therapies for pupils with special needs. These services made closure consequential even for children who remained healthy. A United States modelling study published in 2009 estimated that nationwide school closure for four weeks could cost US$10–47 billion and reduce the availability of key healthcare personnel by 6–19%, because some workers would need to care for children. These estimates describe a proposed intervention, not an observed nationwide closure. (wwwnc.cdc.gov)\n\nOutbreaks also place simultaneous pressure on health systems: patient demand rises while illness and caring responsibilities can diminish staffing. During the 2009 pandemic, a Pan American Health Organization expert meeting identified emergency departments and intensive-care units as facing the heaviest patient-care burden. WHO’s African regional office emphasized the additional pressure on services already managing competing health priorities and the importance of maintaining routine care. Hospital capacity therefore involves not only beds and equipment but also the workforce needed to keep influenza treatment and other essential services operating. (paho.org)\n\nFunding debates concern both the value and affordability of annual vaccination. An evaluation restricted to healthcare expenditure can reach a different result from one that includes avoided work absence and caregiving. WHO’s economic-evaluation guidance also identifies delivery costs, seasonal timing and additional doses for previously unvaccinated children as relevant programme expenses. Its 2026 assessment of improved influenza vaccines identifies disease burden, vaccine prices, health-system capacity and implementation costs as determinants of economic value. Consequently, a programme can be cost-effective without being cost-saving or readily affordable within an existing immunization budget. (pmc.ncbi.nlm.nih.gov)\n\nInternational allocation debates concern whether access depends primarily on national purchasing power or on public-health need. WHO’s 2009 pandemic vaccine deployment initiative delivered more than 78 million doses to 77 countries; its report stressed that equitable access required delivery when products were needed. WHO also distributed its three-million-course rapid-response antiviral stockpile to 72 countries within 14 days of the declaration of pandemic phase 4. Adopted on 24 May 2011, the Pandemic Influenza Preparedness Framework linked sharing viruses with pandemic potential to access to vaccines and antivirals. Its mechanisms include contractual product-access commitments from recipients of viral materials and annual industry contributions supporting preparedness, rather than reliance solely on donations negotiated after an outbreak begins. (iris.who.int)"
    },
    {
      "heading": "Open questions and research frontiers",
      "text": "Universal influenza vaccines seek to redirect immunity toward conserved viral structures rather than the rapidly changing head of hemagglutinin (HA). The HA stem, or stalk, contains sites recognized by broadly neutralizing antibodies that act against multiple influenza subtypes. Vaccine designs include stabilized, headless stem proteins and nanoparticles displaying multiple copies of an antigen. Another approach uses sequential chimeric HAs with different unfamiliar heads but a shared stem, repeatedly stimulating stem-reactive B cells. A phase I randomized controlled trial reported by Raffael Nachbagauer and colleagues in December 2020 demonstrated broad, durable, functional stem-directed antibody responses. Structural work published in 2022 identified an additional conserved “anchor” epitope near the viral membrane, expanding the targets available for vaccine design. (nature.com)\n\nClinical studies distinguish broad antibody responses from demonstrated protection against illness. A 2023 phase I study of H1ssF, a stabilized H1 stem displayed on ferritin nanoparticles, enrolled 52 adults aged 18–70 and induced neutralizing antibodies against multiple group 1 influenza A viruses. This breadth does not establish coverage of group 2 viruses, including H3N2, or influenza B. In May 2023, the US National Institutes of Health announced a separate phase I trial of an mRNA-encoded stem-nanoparticle candidate, H1ssF-3928 mRNA-LNP. A group 1 stem-vaccine study published in 2026 likewise reported broad immune responses, but explicitly lacked human efficacy data and an established correlate of protection. Important unresolved endpoints therefore include protection across antigenically distant viruses, durability, and whether measured stem-antibody levels reliably predict clinical benefit. (pubmed.ncbi.nlm.nih.gov)\n\nPredicting which animal virus will cause a pandemic requires separating the ability to infect an individual from the ability to sustain human-to-human transmission. Genomic surveillance can identify reassortment and host-associated changes, but sequence information alone does not determine transmissibility or population susceptibility. The US Centers for Disease Control and Prevention’s Influenza Risk Assessment Tool combines ten elements covering viral properties, population attributes, and viral ecology and epidemiology; it ranks pandemic potential rather than predicting the next pandemic. Computational models can help integrate heterogeneous evidence. A 2024 study tested eleven machine-learning algorithms using data from 717 ferrets infected with 125 influenza A viruses to classify lethality, morbidity, and respiratory-droplet transmission. Such results predict experimental outcomes, not the occurrence or timing of a human pandemic. (cdc.gov)\n\nHost adaptation remains incompletely understood because successful replication depends on compatibility with cellular machinery as well as viral entry. In 2024, Ecco Staller and colleagues resolved structures of H5N1 polymerase associated with human ANP32B, a host protein required for viral genome replication. The structures showed ANP32B helping organize polymerases into a replication platform. Related structural work explained how adaptations such as PB2 E627K can improve compatibility with mammalian ANP32 proteins. The unresolved predictive problem is how these molecular effects combine in different viral genetic backgrounds: evidence of enhanced replication is not, by itself, evidence of efficient transmission between humans. (nature.com)\n\nThe relationship between antiviral immunity and symptom severity is another research frontier. A controlled human H3N2 infection study published in July 2026 analysed 27 volunteers with low strain-specific neutralizing-antibody levels. Twenty-two became infected: eighteen developed mild-to-moderate symptoms and four remained asymptomatic. Symptomatic participants showed earlier and stronger engagement of local and systemic innate immune pathways. These observations indicate that immune activation cannot simply be equated with protection from symptoms; they do not establish predictors of severe disease, because the study involved healthy volunteers and predominantly mild illness. (nature.com)\n\nLong-term consequences require distinguishing persistent effects of infection from pre-existing disease and the consequences of hospitalization. In December 2023, Yan Xie, Taeyoung Choi, and Ziyad Al-Aly reported an observational comparison including 10,985 US veterans hospitalized with seasonal influenza during 2015–2019, with follow-up extending to eighteen months. The influenza cohort experienced substantial post-acute health loss and health-care use. However, participants were predominantly older White men, and the findings cannot be generalized directly to uncomplicated community infections. The frequency, mechanisms, and duration of post-influenza impairment outside hospitalized populations remain unresolved. (eswi.org)"
    }
  ],
  "links": [
    "virus",
    "epidemiology",
    "immune-system",
    "public-health",
    "vaccine",
    "infectious-disease",
    "fever",
    "cough",
    "fatigue",
    "pneumonia",
    "common-cold",
    "covid-19",
    "cell-membrane",
    "genome",
    "rna",
    "messenger-rna",
    "enzyme",
    "protein",
    "antibody",
    "phylogenetics",
    "world-health-organization",
    "inflammation",
    "cytokine",
    "interferon",
    "sepsis",
    "polymerase-chain-reaction",
    "antigen",
    "mutation",
    "innate-immunity",
    "adaptive-immunity",
    "b-cell",
    "t-cell",
    "immunological-memory",
    "cell",
    "randomized-controlled-trial",
    "productivity",
    "health-system",
    "machine-learning"
  ],
  "sources": [
    {
      "title": "Influenza (seasonal)",
      "url": "https://www.who.int/news-room/fact-sheets/detail/influenza-%28seasonal%29%E2%81%A0"
    },
    {
      "title": "Similarities and Differences between Flu and COVID-19",
      "url": "https://www.cdc.gov/flu/about/flu-vs-covid19.html"
    },
    {
      "title": "Cold Versus Flu",
      "url": "https://www.cdc.gov/flu/about/coldflu.html?CDC_AA_refVal=https%3A%2F%2Fwww.cdc.gov%2Fflu%2Fabout%2Fqa%2Fcoldflu.htm"
    },
    {
      "title": "Types of Influenza Viruses",
      "url": "https://www.cdc.gov/flu/about/viruses-types.html?CDC_AAref_Val=%3Cbr%3Ehttps%3A%2F%2Fwww.cdc.gov%2Fflu%2Fabout%2Fviruses%2Ftypes.htm"
    },
    {
      "title": "Influenza (avian and other zoonotic)",
      "url": "https://www.who.int/en/news-room/fact-sheets/detail/influenza-%28avian-and-other-zoonotic%29"
    },
    {
      "title": "Definition & Facts for Viral Gastroenteritis (“Stomach Flu”) - NIDDK",
      "url": "https://www.niddk.nih.gov/health-information/digestive-diseases/viral-gastroenteritis/definition-facts"
    },
    {
      "title": "Glossary of Influenza (Flu) Terms",
      "url": "https://www.cdc.gov/flu/glossary/index.html"
    },
    {
      "title": "Orthomyxoviridae",
      "url": "https://ictv.global/report_9th/RNAneg/Orthomyxoviridae"
    },
    {
      "title": "Proposal for the creation of Influenzavirus D",
      "url": "https://ictv.global/ICTV/proposals/2014.009a-dM.A.v2.InfluenzavirusD.pdf"
    },
    {
      "title": "Enhancement of the Influenza A Hemagglutinin (HA)-Mediated Cell-Cell Fusion and Virus Entry by the Viral Neuraminidase (NA)",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2795206/"
    },
    {
      "title": "Influenza hemagglutinin is spring-loaded by a metastable native conformation",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC24954/"
    },
    {
      "title": "Structures of influenza A and B replication complexes give insight into avian to human host adaptation and reveal a role of ANP32 as an electrostatic chaperone for the apo-polymerase",
      "url": "https://www.nature.com/articles/s41467-024-51007-3"
    },
    {
      "title": "Influenza virus mRNA trafficking through host nuclear speckles",
      "url": "https://www.nature.com/articles/nmicrobiol201669"
    },
    {
      "title": "Influenza virus uses transportin 1 for vRNP debundling during cell entry",
      "url": "https://www.nature.com/articles/s41564-018-0332-2"
    },
    {
      "title": "Structural insight into cap-snatching and RNA synthesis by influenza polymerase",
      "url": "https://www.nature.com/articles/nature14009"
    },
    {
      "title": "Structures of H5N1 influenza polymerase with ANP32B reveal mechanisms of genome replication and host adaptation",
      "url": "https://www.nature.com/articles/s41467-024-48470-3"
    },
    {
      "title": "The host protein CLUH participates in the subnuclear transport of influenza virus ribonucleoprotein complexes",
      "url": "https://www.nature.com/articles/nmicrobiol201662"
    },
    {
      "title": "Influenza Virus M2 Protein Mediates ESCRT-Independent Membrane Scission",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3059587/"
    },
    {
      "title": "Types of Influenza Viruses",
      "url": "https://www.cdc.gov/flu/about/viruses-types.html"
    },
    {
      "title": "Persistence of Highly Pathogenic Avian Influenza Viruses in Natural Ecosystems",
      "url": "https://wwwnc.cdc.gov/eid/article/16/7/09-0389_article"
    },
    {
      "title": "About Influenza A in Animals",
      "url": "https://www.cdc.gov/flu-in-animals/about/index.html"
    },
    {
      "title": "Questions and Answers: Pandemic Influenza",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/pandemic-resources/basics/faq.html"
    },
    {
      "title": "Recommended composition of influenza virus vaccines for use in the 2026 southern hemisphere influenza season: Frequently asked questions",
      "url": "https://cdn.who.int/media/docs/default-source/influenza/who-influenza-recommendations/vcm-sh-2025/b.-26-september-2025-frequently-asked-questions.pdf?sfvrsn=eb94f684_6"
    },
    {
      "title": "Key Facts About Seasonal Flu Vaccine",
      "url": "https://www.cdc.gov/flu/vaccines/keyfacts.html"
    },
    {
      "title": "Influenza B virus in seals",
      "url": "https://pubmed.ncbi.nlm.nih.gov/10807575/"
    },
    {
      "title": "Isolation of a novel swine influenza virus from Oklahoma in 2011 which is distantly related to human influenza C viruses",
      "url": "https://pubmed.ncbi.nlm.nih.gov/23408893/"
    },
    {
      "title": "Cocirculation of Two Distinct Genetic and Antigenic Lineages of Proposed Influenza D Virus in Cattle",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4300623/"
    },
    {
      "title": "2025–2026 Flu Season",
      "url": "https://www.cdc.gov/flu/season/2025-2026.html"
    },
    {
      "title": "How Flu Spreads",
      "url": "https://www.cdc.gov/flu/spread/index.html?trk=article-ssr-frontend-pulse_little-text-block"
    },
    {
      "title": "Infectious virus in exhaled breath of symptomatic seasonal influenza cases from a college community",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5798362/"
    },
    {
      "title": "Taking Steps for Cleaner Air for Respiratory Virus Prevention",
      "url": "https://www.cdc.gov/respiratory-viruses/prevention/air-quality.html"
    },
    {
      "title": "Influenza Virus Transmission Is Dependent on Relative Humidity and Temperature",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2034399/"
    },
    {
      "title": "Avian flu: influenza virus receptors in the human airway",
      "url": "https://pubmed.ncbi.nlm.nih.gov/16554799/"
    },
    {
      "title": "Influenza Virus Receptor Specificity and Cell Tropism in Mouse and Human Airway Epithelial Cells",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC1563738/"
    },
    {
      "title": "In vitro evolution of H5N1 avian influenza virus toward human-type receptor specificity",
      "url": "https://pubmed.ncbi.nlm.nih.gov/22056389/"
    },
    {
      "title": "A human-airway-on-a-chip for the rapid identification of candidate antiviral therapeutics and prophylactics",
      "url": "https://www.nature.com/articles/s41551-021-00718-9"
    },
    {
      "title": "Influenza H5N1 virus infection of polarized human alveolar epithelial cells and lung microvascular endothelial cells",
      "url": "https://pubmed.ncbi.nlm.nih.gov/19874627/"
    },
    {
      "title": "Local and systemic cytokine responses during experimental human influenza A virus infection. Relation to symptom formation and host defense",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC508608/"
    },
    {
      "title": "Clinical Signs and Symptoms of Influenza",
      "url": "https://www.cdc.gov/flu/hcp/clinical-signs/index.html"
    },
    {
      "title": "Signs and Symptoms of Flu",
      "url": "https://www.cdc.gov/flu/signs-symptoms/index.html"
    },
    {
      "title": "Influenza | CDC Yellow Book",
      "url": "https://www.cdc.gov/yellow-book/hcp/travel-associated-infections-diseases/influenza.html"
    },
    {
      "title": "HHS Pandemic Influenza Plan",
      "url": "https://www.cdc.gov/pandemic-flu/media/hhspandemicinfluenzaplan.pdf"
    },
    {
      "title": "Chapter 12: Influenza | Pink Book",
      "url": "https://www.cdc.gov/pinkbook/hcp/table-of-contents/chapter-12-influenza.html"
    },
    {
      "title": "Acute Cardiovascular Events Associated With Influenza in Hospitalized Adults: A Cross-sectional Study",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8097760/"
    },
    {
      "title": "Global Influenza Programme: Surveillance and monitoring",
      "url": "https://www.who.int/teams/global-influenza-programme/surveillance-and-monitoring"
    },
    {
      "title": "Overview of Influenza Testing Methods",
      "url": "https://www.cdc.gov/flu/hcp/testing-methods/index.html"
    },
    {
      "title": "Information on Rapid Molecular Assays, RT-PCR, and other Molecular Assays for Diagnosis of Influenza Virus Infection",
      "url": "https://www.cdc.gov/flu/hcp/testing-methods/molecular-assays.html"
    },
    {
      "title": "Information for Clinicians on Rapid Diagnostic Testing for Influenza",
      "url": "https://www.cdc.gov/flu/hcp/testing-methods/rapidclin.html"
    },
    {
      "title": "Rapid Diagnostic Testing for Influenza: Information for Clinical Laboratory Directors",
      "url": "https://www.cdc.gov/flu/php/laboratories/rapidlab.html"
    },
    {
      "title": "Influenza Virus Genome Sequencing and Genetic Characterization",
      "url": "https://www.cdc.gov/flu/php/viruses/genetic-characterization.html"
    },
    {
      "title": "Innate immunity to influenza virus infection",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4104278/"
    },
    {
      "title": "Innate Immune Response to Influenza Virus at Single-Cell Resolution in Human Epithelial Cells Revealed Paracrine Induction of Interferon Lambda 1",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC6798124/"
    },
    {
      "title": "Evaluation of Antihemagglutinin and Antineuraminidase Antibodies as Correlates of Protection in an Influenza A/H1N1 Virus Healthy Human Challenge Model",
      "url": "https://pubmed.ncbi.nlm.nih.gov/27094330/"
    },
    {
      "title": "Antibody Correlates and Predictors of Immunity to Naturally Occurring Influenza in Humans and the Importance of Antibody to the Neuraminidase",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3633450/"
    },
    {
      "title": "Cellular immune correlates of protection against symptomatic pandemic influenza",
      "url": "https://www.nature.com/articles/nm.3350"
    },
    {
      "title": "Neutralizing antibodies derived from the B cells of 1918 influenza pandemic survivors",
      "url": "https://www.nature.com/articles/nature07231"
    },
    {
      "title": "Potent Protection against H5N1 and H7N9 Influenza via Childhood Hemagglutinin Imprinting",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5134739/"
    },
    {
      "title": "How Flu Viruses Can Change: \"Drift\" and \"Shift\"",
      "url": "https://www.cdc.gov/flu/php/viruses/change.html"
    },
    {
      "title": "Chapter 6: Influenza",
      "url": "https://www.cdc.gov/surv-manual/php/table-of-contents/chapter-6-influenza.html"
    },
    {
      "title": "Substitutions near the receptor binding site determine major antigenic change during influenza virus evolution",
      "url": "https://pubmed.ncbi.nlm.nih.gov/24264991/"
    },
    {
      "title": "Mapping the antigenic and genetic evolution of influenza virus",
      "url": "https://pubmed.ncbi.nlm.nih.gov/15218094/"
    },
    {
      "title": "About Bat Flu",
      "url": "https://www.cdc.gov/flu-in-animals/about/bat-flu.html"
    },
    {
      "title": "CDC H1N1 Flu: Origin of 2009 H1N1 Flu (Swine Flu)",
      "url": "https://archive.cdc.gov/www_cdc_gov/h1n1flu/information_h1n1_virus_qa.htm"
    },
    {
      "title": "Antigenic Characterization",
      "url": "https://www.cdc.gov/flu/php/viruses/antigenic.html"
    },
    {
      "title": "Improving influenza vaccine",
      "url": "https://iris.who.int/bitstream/10665/249599/1/WHO-OHE-PED-GIP-2016.4-eng.pdf"
    },
    {
      "title": "Influenza (seasonal)",
      "url": "https://www.who.int/en/news-room/fact-sheets/detail/influenza-%28seasonal%EF%BC%89"
    },
    {
      "title": "Environmental Predictors of Seasonal Influenza Epidemics across Temperate and Tropical Climates",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3591336/"
    },
    {
      "title": "Influenza Virus Transmission Is Dependent on Relative Humidity and Temperature",
      "url": "https://journals.plos.org/plospathogens/article?id=10.1371%2Fjournal.ppat.0030151"
    },
    {
      "title": "Dynamics of Airborne Influenza A Viruses Indoors and Dependence on Humidity",
      "url": "https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0021481"
    },
    {
      "title": "Global burden of respiratory infections associated with seasonal influenza in children under 5 years in 2018: a systematic review and modelling study",
      "url": "https://stacks.cdc.gov/view/cdc/92503"
    },
    {
      "title": "Estimates of global seasonal influenza-associated respiratory mortality: a modelling study",
      "url": "https://pubmed.ncbi.nlm.nih.gov/29248255/"
    },
    {
      "title": "Estimated Flu-Related Illnesses, Medical Visits, Hospitalizations, and Deaths in the United States — 2017–2018 Flu Season",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/about/burden/2017-2018.htm"
    },
    {
      "title": "Changes in Influenza and Other Respiratory Virus Activity During the COVID-19 Pandemic — United States, 2020–2021",
      "url": "https://www.cdc.gov/mmwr/volumes/70/wr/mm7029a1.htm"
    },
    {
      "title": "1918 Pandemic (H1N1 virus)",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/pandemic-resources/1918-pandemic-h1n1.html"
    },
    {
      "title": "1918 Influenza: the Mother of All Pandemics",
      "url": "https://wwwnc.cdc.gov/eid/article/12/1/05-0979_article"
    },
    {
      "title": "1957-1958 Pandemic (H2N2 virus)",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/pandemic-resources/1957-1958-pandemic.html"
    },
    {
      "title": "Community Mitigation Guidelines to Prevent Pandemic Influenza — United States, 2017",
      "url": "https://www.cdc.gov/mmwr/volumes/66/rr/rr6601a1.htm"
    },
    {
      "title": "Commission on Influenza report on immunization against the 1957 antigenic variant",
      "url": "https://stacks.cdc.gov/view/cdc/72938/cdc_72938_DS1.pdf"
    },
    {
      "title": "1968 Pandemic (H3N2 virus)",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/pandemic-resources/1968-pandemic.html"
    },
    {
      "title": "Influenza Pandemics of the 20th Century",
      "url": "https://wwwnc.cdc.gov/eid/article/12/1/05-1254_article"
    },
    {
      "title": "Influenza Pandemic Periodicity, Virus Recycling, and the Art of Risk Assessment",
      "url": "https://wwwnc.cdc.gov/eid/article/12/1/05-1013_article"
    },
    {
      "title": "Trials of aqueous killed influenza vaccine in Canada, 1968-69.",
      "url": "https://pubmed.ncbi.nlm.nih.gov/4605401/"
    },
    {
      "title": "2009 H1N1 Pandemic (H1N1pdm09 virus)",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/pandemic-resources/2009-h1n1-pandemic.html"
    },
    {
      "title": "Summary of Flu Progress since 2009",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/pandemic-resources/h1n1-summary.htm"
    },
    {
      "title": "The 2009 H1N1 Pandemic: Summary Highlights, April 2009-April 2010",
      "url": "https://archive.cdc.gov/www_cdc_gov/h1n1flu/cdcresponse.htm"
    },
    {
      "title": "Influenza: An Emerging Disease",
      "url": "https://wwwnc.cdc.gov/eid/article/4/3/98-0325_article"
    },
    {
      "title": "Current Situation: Bird Flu in Wild Birds",
      "url": "https://cdc.gov/bird-flu/situation-summary/wildbirds.html"
    },
    {
      "title": "About Influenza in Swine (Pigs)",
      "url": "https://www.cdc.gov/swine-flu/about/influenza-in-swine.html"
    },
    {
      "title": "Emergence of influenza A viruses",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC1088557/"
    },
    {
      "title": "Global Influenza Programme: Human-animal interface",
      "url": "https://www.who.int/teams/global-influenza-programme/avian-influenza"
    },
    {
      "title": "Emergence and Evolution of H5N1 Bird Flu",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/avianflu/communication-resources/bird-flu-origin-infographic.html"
    },
    {
      "title": "Global Human Cases with Avian Influenza A(H5N1), 1997–2026",
      "url": "https://www.cdc.gov/bird-flu/php/surveillance/chart-epi-curve-ah5n1.html"
    },
    {
      "title": "Avian Influenza Weekly Update # 1064: 2 October 2026",
      "url": "https://www.who.int/westernpacific/publications/m/item/avian-influenza-weekly-update---1064--2-october-2026"
    },
    {
      "title": "Updated joint FAO/WHO/WOAH public health assessment of recent high pathogenicity avian influenza A(H5) virus events in animals and people",
      "url": "https://www.who.int/publications/m/item/updated-joint-fao-who-woah-public-health-assessment-of-recent-high-pathogenicity-avian-influenza-a%28h5%29-virus-events-in-animals-and-people"
    },
    {
      "title": "Pathogenicity and transmissibility of bovine H5N1 influenza virus",
      "url": "https://www.nature.com/articles/s41586-024-07766-6"
    },
    {
      "title": "H5N1 clade 2.3.4.4b dynamics in experimentally infected calves and cows",
      "url": "https://www.nature.com/articles/s41586-024-08063-y"
    },
    {
      "title": "How Influenza (Flu) Vaccines Are Made",
      "url": "https://www.cdc.gov/flu/vaccine-process/"
    },
    {
      "title": "Clinical Guidance for Seasonal Influenza Vaccine Safety",
      "url": "https://www.cdc.gov/flu/hcp/vax-summary/vaccine-safety.html"
    },
    {
      "title": "Interim Clinical Considerations for the Use of Seasonal Influenza Vaccines in the United States",
      "url": "https://www.cdc.gov/flu/hcp/vax-summary/acip-recommendations.html"
    },
    {
      "title": "ACIP Recommendations Summary",
      "url": "https://www.cdc.gov/flu/hcp/acip"
    },
    {
      "title": "MFLUSIVA",
      "url": "https://www.fda.gov/vaccines-blood-biologics/vaccines/mflusiva?trk=public_post_comment-text"
    },
    {
      "title": "August 5, 2026 Summary Basis for Regulatory Action - MFLUSIVA",
      "url": "https://www.fda.gov/media/194538/download"
    },
    {
      "title": "WHO issues updated influenza vaccines position paper",
      "url": "https://www.who.int/news/item/01-06-2022-who-issues-updated-influenza-vaccines-position-paper"
    },
    {
      "title": "WHO Preferred Product Characteristics for Next-Generation Influenza Vaccines",
      "url": "https://iris.who.int/bitstream/handle/10665/258767/9789241512466-eng.pdf"
    },
    {
      "title": "Improving the selection and development of influenza vaccine viruses – Report of a WHO informal consultation",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5357705/"
    },
    {
      "title": "Frequently asked questions: Recommended composition of influenza virus vaccines for the 2024–2025 northern hemisphere influenza season",
      "url": "https://cdn.who.int/media/docs/default-source/influenza/who-influenza-recommendations/vcm-northern-hemisphere-recommendation-2024-2025/202402_frequently-asked-questions.pdf"
    },
    {
      "title": "2026-2027 Flu Season",
      "url": "https://www.cdc.gov/flu/season/2026-2027.html"
    },
    {
      "title": "What are the mechanisms of action of the antivirals? - Clinical practice guidelines for influenza",
      "url": "https://www.ncbi.nlm.nih.gov/sites/books/NBK607905/?report=reader"
    },
    {
      "title": "Influenza Antiviral Medications: Summary for Clinicians",
      "url": "https://www.cdc.gov/flu/hcp/antivirals/summary-clinicians.html?CDC_AAref_Val=https%3A%2F%2Fwww.cdc.gov%2Fflu%2Fprofessionals%2Fantivirals%2Fsummary-clinicians.htm"
    },
    {
      "title": "Influenza Antiviral Drug Resistance",
      "url": "https://www.cdc.gov/flu/treatment/antiviralresistance.html"
    },
    {
      "title": "Susceptibility of Influenza Viruses to the Novel Cap-Dependent Endonuclease Inhibitor Baloxavir Marboxil",
      "url": "https://pubmed.ncbi.nlm.nih.gov/30574137/"
    },
    {
      "title": "The active form of the influenza cap-snatching endonuclease inhibitor baloxavir marboxil is a tight binding inhibitor",
      "url": "https://pubmed.ncbi.nlm.nih.gov/33647314/"
    },
    {
      "title": "Baloxavir Marboxil for Uncomplicated Influenza in Adults and Adolescents",
      "url": "https://pubmed.ncbi.nlm.nih.gov/30184455/"
    },
    {
      "title": "Oseltamivir treatment for influenza in adults: a meta-analysis of randomised controlled trials",
      "url": "https://pubmed.ncbi.nlm.nih.gov/25640810/"
    },
    {
      "title": "Oseltamivir for influenza in adults and children: systematic review of clinical study reports and summary of regulatory comments",
      "url": "https://pubmed.ncbi.nlm.nih.gov/24811411/"
    },
    {
      "title": "Effectiveness of neuraminidase inhibitors in reducing mortality in patients admitted to hospital with influenza A H1N1pdm09 virus infection: a meta-analysis of individual participant data",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC6637757/"
    },
    {
      "title": "World Health Emergencies and living guidelines",
      "url": "https://extranet.who.int/prequal/key-resources/documents/2024-joint-unicef-unfpa-who-meeting-world-health-emergencies-and-living"
    },
    {
      "title": "Executive summary - Clinical practice guidelines for influenza",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK607909/"
    },
    {
      "title": "Recommendations for patients with non-severe symptomatic influenza (suspected or confirmed)",
      "url": "https://www.ncbi.nlm.nih.gov/books/n/who378872/pdf/"
    },
    {
      "title": "Global Influenza Surveillance and Response System (GISRS)",
      "url": "https://www.who.int/initiatives/global-influenza-surveillance-and-response-system"
    },
    {
      "title": "National Influenza Centres",
      "url": "https://www.who.int/initiatives/global-influenza-surveillance-and-response-system/national-influenza-centres"
    },
    {
      "title": "Global Influenza Programme: FluNet",
      "url": "https://www.who.int/tools/flunet/"
    },
    {
      "title": "Sentinel sites and specimens",
      "url": "https://www.who.int/teams/global-influenza-programme/influenza-covid19/sentinel-sites-and-specimens"
    },
    {
      "title": "Effects of Hand Hygiene Campaigns on Incidence of Laboratory-confirmed Influenza and Absenteeism in Schoolchildren, Cairo, Egypt",
      "url": "https://wwwnc.cdc.gov/eid/article/17/4/10-1353_article"
    },
    {
      "title": "WHO guideline on public health and social measures for mitigating the risk and impact of epidemic and pandemic influenza",
      "url": "https://www.who.int/publications/i/item/9789240116658"
    },
    {
      "title": "Pandemic Influenza Risk Management",
      "url": "https://www.who.int/publications/i/item/WHO-WHE-IHM-GIP-2017.1"
    },
    {
      "title": "Pandemic Influenza Risk Management",
      "url": "https://www.who.int/docs/default-source/documents/pandamic-influenza-risk-management.pdf"
    },
    {
      "title": "Pandemic Influenza Risk Management",
      "url": "https://cdn.who.int/media/docs/default-source/documents/pandamic-influenza-risk-management.pdf"
    },
    {
      "title": "A practical guide for developing and conducting simulation exercise",
      "url": "https://www.who.int/publications-detail-redirect/9789241514507"
    },
    {
      "title": "A virus obtained from influenza patients",
      "url": "https://pashev.ru/files/smith-1933.pdf.html"
    },
    {
      "title": "History of influenza vaccination",
      "url": "https://www.who.int/news-room/spotlight/history-of-vaccination/history-of-influenza-vaccination"
    },
    {
      "title": "CULTIVATION OF HUMAN INFLUENZA VIRUS IN AN ARTIFICIAL MEDIUM.",
      "url": "https://pubmed.ncbi.nlm.nih.gov/17733215/"
    },
    {
      "title": "STUDIES WITH HUMAN INFLUENZA VIRUS CULTIVATED IN ARTIFICIAL MEDIUM",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2133410/"
    },
    {
      "title": "Frank Macfarlane Burnet 1899-1985",
      "url": "https://science.org.au/our-focus/history-australian-science/fellows-biographical-memoirs/frank-macfarlane-burnet-1899-1985"
    },
    {
      "title": "Discovery and characterization of the 1918 pandemic influenza virus in historical context",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2391305/"
    },
    {
      "title": "Characterization of the Reconstructed 1918 Spanish Influenza Pandemic Virus",
      "url": "https://cmgm-new.stanford.edu/biochem/biochem230/papers2005/week8/1918Bkgd1.pdf"
    },
    {
      "title": "Ask a CDC Scientist: Dr. Terrence Tumpey and the Reconstruction of the 1918 Pandemic Virus",
      "url": "https://archive.cdc.gov/www_cdc_gov/flu/pandemic-resources/1918-commemoration/special-features/ask-a-scientist-terrence-tumpey.htm"
    },
    {
      "title": "Structure of the haemagglutinin membrane glycoprotein of influenza virus at 3 Å resolution",
      "url": "https://www.nature.com/articles/289366a0"
    },
    {
      "title": "Structure of influenza haemagglutinin at the pH of membrane fusion",
      "url": "https://www.nature.com/articles/371037a0"
    },
    {
      "title": "Structure of the influenza virus glycoprotein antigen neuraminidase at 2.9 Å resolution",
      "url": "https://www.nature.com/articles/303035a0"
    },
    {
      "title": "Economic burden of seasonal influenza in the United States",
      "url": "https://pubmed.ncbi.nlm.nih.gov/29801998/"
    },
    {
      "title": "WHO guide on the economic evaluation of influenza vaccination",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5820425/"
    },
    {
      "title": "World Bank Document: Global Development Finance 2009",
      "url": "https://documents1.worldbank.org/curated/en/403051468339681711/pdf/493170PUB0v10G101Official0Use0Only1.pdf"
    },
    {
      "title": "World Bank Document: Progress Report 2010",
      "url": "https://documents1.worldbank.org/curated/en/880721468335983143/pdf/879300PUB0Box30Progress0Report02010.pdf"
    },
    {
      "title": "Pandemic Influenza as 21st Century Urban Public Health Crisis",
      "url": "https://wwwnc.cdc.gov/eid/article/15/12/09-1232_article"
    },
    {
      "title": "Economic Cost and Health Care Workforce Effects of School Closures in the U.S",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2762813/"
    },
    {
      "title": "Pandemic Influenza puts additional pressure on sub Saharan Africa’s health systems",
      "url": "https://afro.who.int/news/pandemic-influenza-puts-additional-pressure-sub-saharan-africas-health-systems"
    },
    {
      "title": "Hospital preparedness checklist for pandemic influenza",
      "url": "https://www.who.int/publications/i/item/hospital-preparedness-checklist-for-pandemic-influenza"
    },
    {
      "title": "Full Value of improved Influenza Vaccine Assessment (FVIVA)",
      "url": "https://www.who.int/teams/immunization-vaccines-and-biologicals/immunization-analysis-and-insights/vaccine-impact-value/full-value-of-improved-influenza-vaccine-assessment-%28fviva%29"
    },
    {
      "title": "Report of the WHO Pandemic Influenza A(H1N1) Vaccine Deployment Initiative",
      "url": "https://iris.who.int/bitstream/handle/10665/44795/9789241564427_eng.pdf"
    },
    {
      "title": "Report of the WHO Pandemic Influenza A(H1N1) Vaccine Deployment Initiative",
      "url": "https://www.who.int/publications/i/item/report-of-the-who-pandemic-influenza-a%28h1n1%29-vaccine-deployment-initiative"
    },
    {
      "title": "Structural and genetic basis for development of broadly neutralizing influenza antibodies",
      "url": "https://www.nature.com/articles/nature11371"
    },
    {
      "title": "A chimeric hemagglutinin-based universal influenza virus vaccine approach induces broad and long-lasting immunity in a randomized, placebo-controlled phase I trial",
      "url": "https://www.nature.com/articles/s41591-020-1118-7"
    },
    {
      "title": "Broadly neutralizing antibodies target a haemagglutinin anchor epitope",
      "url": "https://www.nature.com/articles/s41586-021-04356-8"
    },
    {
      "title": "An influenza hemagglutinin stem nanoparticle vaccine induces cross-group 1 neutralizing antibodies in healthy adults",
      "url": "https://pubmed.ncbi.nlm.nih.gov/37075129/?dopt=Abstract"
    },
    {
      "title": "A group 1 hemagglutinin stem vaccine elicits broad humoral responses against influenza in phase 1/2a study",
      "url": "https://www.nature.com/articles/s41467-026-70396-1"
    },
    {
      "title": "Influenza Risk Assessment Tool (IRAT)",
      "url": "https://www.cdc.gov/pandemic-flu/php/national-strategy/influenza-risk-assessment-tool.html"
    },
    {
      "title": "Machine learning approaches for influenza A virus risk assessment identifies predictive correlates using ferret model in vivo data",
      "url": "https://www.nature.com/articles/s42003-024-06629-0"
    },
    {
      "title": "Innate immune responsiveness predicts enhanced cellular immunity and symptomatic disease after controlled human influenza infection",
      "url": "https://www.nature.com/articles/s41591-026-04483-7"
    },
    {
      "title": "Long-term outcomes following hospital admission for COVID-19 versus seasonal influenza: a cohort study",
      "url": "https://eswi.org/sites/default/files/2024-01/PIIS1473309923006849.pdf"
    }
  ],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}