UpToDate
About Us | News from UpToDate | Contact Us | Help
white circle Log Out
New Search
Patient Info
What's New
Calculators
CME 160.5
My Account
Back to Search Results for "influenza"
Epidemiology of influenza
Find in TopicFind Patient InfoPatient Print This TopicPrint Email This TopicEmail
Epidemiology of influenza
Author
Raphael Dolin, MD
Section Editor
Martin S Hirsch, MD
Deputy Editor
Anna R Thorner, MD
Disclosures
All topics are updated as new evidence becomes available and our peer review process is complete.
Literature review current through: Dec 2012. | This topic last updated: Oct 4, 2012.
INTRODUCTION - Influenza occurs in distinct outbreaks of varying extent every year. This epidemiologic pattern reflects the changing nature of the antigenic properties of influenza viruses, and their subsequent spread depends upon multiple factors, including transmissibility of the virus and the susceptibility of the population. Influenza A viruses, in particular, have a remarkable ability to undergo periodic changes in the antigenic characteristics of their envelope glycoproteins, the hemagglutinin and the neuraminidase.
Influenza hemagglutinin is a surface glycoprotein that binds to sialic acid residues on respiratory epithelial cell surface glycoproteins. This interaction is necessary for the initiation of infection. After viral replication, progeny virions are also bound to the host cell. Neuraminidase cleaves these links and liberates the new virions; it also counteracts hemagglutinin-mediated self-aggregation entrapment in respiratory secretions.
Among influenza A viruses that infect humans, three major subtypes of hemagglutinins (H1, H2, and H3) and two subtypes of neuraminidases (N1 and N2) have been described. Influenza B viruses have a lesser propensity for antigenic changes, and only antigenic drifts in the hemagglutinin have been described.
The epidemiology of influenza, including morbidity and mortality, will be reviewed here. The clinical manifestations, complications, diagnosis, prevention, and treatment of this infection are discussed separately; the epidemiology of pandemic H1N1 influenza ("swine influenza") and avian influenza are also presented elsewhere. (See "Clinical manifestations of seasonal influenza in adults" and "Antiviral drugs for the prevention and treatment of seasonal influenza in children" and "Seasonal influenza vaccination in adults" and "Diagnosis of seasonal influenza in adults" and "Treatment of seasonal influenza in adults" and "Prevention of seasonal influenza in adults" and "Epidemiology of pandemic H1N1 influenza ('swine influenza')" and "Epidemiology, transmission, and pathogenesis of avian influenza".)
DEFINITION OF ANTIGENIC SHIFTS AND DRIFTS - Major changes in the envelope glycoproteins, the hemagglutinin and the neuraminidase, are referred to as antigenic shifts, and minor changes are called antigenic drifts. Antigenic shifts are associated with epidemics and pandemics of influenza A, whereas antigenic drifts are associated with more localized outbreaks of varying extent.
ANTIGENIC SHIFTS - Influenza viruses have a segmented genome that can result in high rates of reassortment among viruses coinfecting the same cell. Reassortment between animal and human viruses may result in the emergence of pandemic strains (table 1) [1,2]. Such reassortment events led to the emergence of the viruses that caused the pandemics of 1957, 1968, and 2009. (See 'Pathogenesis' below.)
Reassortment between different influenza viruses does not always result in a pandemic. As an example, a novel reassortant strain of influenza A H1N2 viruses appeared in humans in the 2001 to 2002 season in North America, Europe, the Middle East, and Southeast Asia, and caused seasonal epidemics [3-5].
There was concern that reassortment between human and avian influenza viruses might have occurred when avian influenza A/H5N1 infections were detected in humans in Hong Kong in March 1997 at the time of an extensive outbreak of avian influenza A/H5N1 in poultry. However, such a reassortant virus was not found. Nevertheless, the potential for such reassortment remains a concern, given the high mortality of avian influenza A/H5N1 infections in humans. (See "Epidemiology, transmission, and pathogenesis of avian influenza".)
Many historians agree that the first recognition of an influenza pandemic was in 1510 [6]. In more recent history, the 1918 influenza was particularly severe and widespread.
Pandemic of 1918 - The extremely severe and extensive pandemic of 1918 and 1919 (swine influenza or Spanish influenza) was associated with the emergence of antigenic shifts in both the hemagglutinin (H1) and the neuraminidase (N1) of influenza A [7]. The pathogenicity of this virus has been well-characterized. (See 'Pathogenesis' below.)
Other pandemics
In 1957, the shift to H2 and N2 resulted in a severe pandemic.
In 1968, an antigenic shift occurred that involved only the hemagglutinin (from H2N2 to H3N2); the resulting pandemic was less extensive than that seen in 1957 [8].
In 1977, an influenza A virus emerged that had shifted to H1N1. The resulting pandemic affected primarily young individuals who lacked preexisting immunity to H1N1 (ie, those born after H1N1 viruses had last circulated from 1918 to 1957).
Since 1977, A/H1N1 and A/H3N2 subtypes along with influenza B viruses have frequently circulated at the same time [9].
The emergence of a novel H1N1 human-swine-avian reassortant virus in March 2009 in North America resulted in a pandemic, and the pandemic H1N1 virus has continued to circulate since then. (See "Epidemiology of pandemic H1N1 influenza ('swine influenza')".)
H3N2 variant influenza - Since July 2011, the United States Centers for Diseases Control and Prevention (CDC) has reported over 300 cases of H3N2 variant influenza A infection caused by reassortment of swine-origin H3N2 influenza A viruses and 2009 pandemic H1N1 influenza A viruses, most of which have occurred since July 2012 [10-17]. The H3N2 variant influenza virus contains the M gene from 2009 pandemic H1N1 influenza A virus, which may confer increased transmissibility to and among humans compared with other swine-origin influenza viruses [14]. However, sustained human-to-human transmission of this strain has not been observed. The majority of cases in 2012 have been reported in Indiana and Ohio, but cases have also been reported in several other states. Almost all patients reported direct or indirect contact with swine, and the majority attended fairs where swine were present. Most patients have had mild illness. Updated case counts can be found on the CDC's website.
The CDC has issued recommendations for the prevention of transmission of H3N2 variant influenza, which include frequent handwashing and avoidance of contact with pigs that appear ill [12]. In addition, individuals at high risk for influenza complications (eg, individuals <5 or ?65 years of age, pregnant women, and individuals with certain chronic medical conditions) should consider avoiding exposure to pigs and swine barns, especially if sick pigs have been identified. Further recommendations can be found on the CDC's website.
The viruses are susceptible to oseltamivir and zanamivir, but resistant to amantadine and rimantadine. (See "Treatment of seasonal influenza in adults", section on 'Choice of antiviral drug'.)
The seasonal influenza vaccine is not expected to provide protection against these strains. (See "Seasonal influenza vaccination in adults", section on 'Vaccine design'.)
The World Health Organization and the CDC, in conjunction with other agencies, have decided to refer to swine-origin influenza viruses identified in humans as "variant" viruses and to denote them with a "v" (eg, H3N2v) [10].
ANTIGENIC DRIFTS - Between the years of antigenic shifts, antigenic drifts have occurred almost annually and have resulted in outbreaks of variable extent and severity. Outbreaks due to antigenic drifts are usually less extensive and severe than the epidemics or pandemics associated with antigenic shifts. Antigenic drifts are believed to result from point mutations in the RNA gene segments that code for the hemagglutinin or the neuraminidase; they are thought to occur sequentially as the virus spreads through a susceptible population [18]. Changes in the hemagglutinin that result in antigenic shifts are of such great magnitude that they cannot be accounted for by point mutations alone.
PATHOGENESIS - The extremely severe and extensive pandemic of 1918 and 1919 resulted in 50 to 100 million deaths worldwide and was exceptional in the high death rates that were seen among healthy adults aged 15 to 34 years [19,20]. A similarly high death rate has not occurred in this age group in either prior or subsequent influenza A pandemics or epidemics.
The pathogenicity of the hemagglutinin of the 1918 pandemic virus was directly demonstrated in a mouse model using genetic recombination techniques [21,22]. Researchers constructed influenza viruses using hemagglutinin alone or hemagglutinin with neuraminidase from the pandemic strain. Both recombinant viruses led to widespread infection of the lungs, suggesting that hemagglutinin conferred enhanced pathogenicity in mice. Furthermore, these recombinant viruses could induce high levels of chemokines and cytokines, resulting in inflammatory cell infiltration and severe hemorrhage that were characteristic of the illness seen during the pandemic.
Scientists used reverse genetics to create an influenza virus with all eight gene segments of the 1918 pandemic strain in order to study its virulence in animal models [23]. After infection, the 1918 strain produced 39,000 times more virus particles in the lungs of mice compared with more contemporary H1N1 influenza strains. Furthermore, the ability of an influenza virus to replicate in the absence of protease is thought to be a critical determinant of pathogenicity in animal models; the 1918 strain was able to replicate equally well in the absence or presence of trypsin in vitro.
Scientists have fully sequenced the entire genome of the 1918 strain, which has given insight into the origins of the virus [24]. In the pandemics of 1957 and 1968, two to three gene segments from avian strains combined with the circulating human strain to form a reassortant virus. The influenza virus that caused the 2009 pandemic was caused by a quadruple reassortment of two swine strains, one human strain, and one avian strain. In contrast, sequence and phylogenetic analyses of the 1918 virus genome suggest that it was derived wholly from an ancestor that originally infected birds and adapted to humans. Some amino acid changes identified in the 1918 strain are also seen in H5N1 and H7N7 avian viruses that have caused human fatalities. (See "Epidemiology, transmission, and pathogenesis of avian influenza".)
CHARACTERISTICS OF INFLUENZA OUTBREAKS - Influenza outbreaks have a seasonal distribution and characteristic time course. Factors influencing the extent and severity of an outbreak are less clear. Two or three different influenza strains typically circulate concurrently in a given influenza season [25].
One factor that may influence which influenza strains will predominate during an influenza season is the seroprotection rate among children during the preceding season. In a study in which children and adults were tested for antibodies to locally circulating influenza strains during three consecutive influenza seasons (2006-2007, 2007-2008, and 2008-2009), the lowest rates of seroprotection in children (but not in adults) coincided with the dominant influenza subtype during the following winter epidemic [26]. The authors suggest that the protection rate in children is important for shaping future epidemics because children are prolific disseminators of respiratory virus infections.
Seasonality - Outbreaks of influenza occur almost exclusively during the winter months in the northern and southern hemispheres (which occur at different times of the year). It is highly unusual to detect influenza A viruses at other times, although individual infections and even outbreaks have been reported during the warm weather months.
Travelers to tropical regions should be reminded that influenza may occur throughout the year in the tropics [27]. In addition, summertime outbreaks of influenza have occurred on cruise ships in the northern and southern hemispheres. Repeat vaccination is not necessary in those who received routine vaccination at the appropriate time in the previous fall or winter. Volume of airline travel has also been linked to transnational spread of influenza infection [28]. (See "Immunizations for travel", section on 'Influenza vaccine' and "Seasonal influenza vaccination in adults" and "Seasonal influenza vaccination in children", section on 'Travelers'.)
How influenza A virus persists between outbreaks remains poorly understood. It is possible that sporadic cases of viral infection at other times are caused by influenza but not diagnosed as such or that virus is imported from geographically distant sites, in which outbreaks are occurring, by the travel of infected individuals.
Time course of an outbreak - Influenza A outbreaks typically begin abruptly, peak over a two to three week period, and last for two to three months [25,29]. In most outbreaks, the earliest indication of influenza activity is an increase in febrile respiratory illnesses in children, followed by increases in influenza-like illnesses in adults. Increases in absenteeism from work and school are usually later manifestations of outbreaks. (See "Clinical manifestations of seasonal influenza in adults".)
Most outbreaks have attack rates of 10 to 20 percent in the general population, but rates can exceed 50 percent in pandemics [30]. Extraordinarily high attack rates have been reported in institutionalized and semiclosed populations.
Factors determining the severity of an outbreak - The factors that determine the extent and severity of outbreaks are not fully understood. The susceptibility of the population, as determined by the prevalence of antibodies to circulating virus, clearly plays a major role. Some outbreaks cease when a large pool of susceptible individuals is no longer present in the population. However, some outbreaks appear to end when a large pool of susceptible individuals still exists. It has been suggested that influenza viruses may differ in "intrinsic virulence" such as their efficiency of transmission or their ability to cause symptomatic infection.
The severity of influenza outbreaks depends partly on the strain(s) of influenza virus circulating among the total population. H3N2 influenza A viruses usually cause the most severe disease, followed by influenza B viruses, and with H1N1 influenza A viruses causing the least severe disease [31-33]. However, among young children, H1N1 influenza infections may be more severe than in older children and adults. Between 1976 and 2007, the average mortality rates in the United States for the 22 seasons during which H3N2 influenza A was a prominent strain were 2.7 times higher than for the nine seasons that it was not the major strain [34]. Seasons in which both H3N2 influenza A viruses and influenza B viruses cocirculate cause higher rates of hospitalization than otherwise expected [31].
ANTIBODY RESPONSE TO THE 1918 PANDEMIC STRAIN - In a study of 32 individuals born in or before 1915, all had neutralizing antibody responses to the H1N1 influenza strain that caused the 1918 pandemic, even nine decades after its occurrence [35]. Seven of eight individuals tested had circulating B cells that secreted antibodies that bound hemagglutinin (HA) from the 1918 pandemic influenza strain. Monoclonal antibodies that were generated from the B cells of three separate donors had potent neutralizing activity against the 1918 strain and bound to its HA protein with high affinity. They also cross-reacted with the genetically similar HA of a 1930 swine H1N1 influenza strain, but did not cross-react with HAs of more contemporary human influenza viruses.
MORBIDITY AND MORTALITY IN ADULTS - Between 1976 and 2007, annual influenza-associated deaths from respiratory and circulatory causes (including pneumonia and influenza) in the United States ranged from 3349 to 48,614, and the annual rate of influenza-associated deaths ranged from 1.4 to 16.7 deaths per 100,000 persons [34]. Due at least in part to high attack rates, the morbidity caused by influenza in the general population is substantial. Among adults, increased rates of morbidity and mortality are associated with advanced age and with underlying comorbidities [36]:
Influenza epidemics generally disproportionately affect elderly persons, with the highest rates of morbidity and mortality in this group [31-33,37-40]. In a study of the National Hospital Discharge Survey database, hospitalization rates for pneumonia increased by 20 percent from 1988-1990 to 2000-2002 for patients aged 65 to 85 years [38]. In addition, the risk of death during a hospitalization was 50 percent higher if the diagnosis was pneumonia compared with 10 other common reasons for admission in the elderly population. The risk of pneumonia in this age group is increased in patients with comorbid conditions, such as chronic cardiac and pulmonary diseases or diabetes [37,38].
Excess hospitalizations for patients with chronic diseases who acquire influenza infection range from approximately 20 to more than 1000 per 100,000 individuals, with the highest rates occurring in those less than five and more than 64 years of age. Similar findings were noted in a retrospective cohort study of women under the age of 65 with and without chronic medical conditions [41]. Rates of hospitalization for acute cardiopulmonary events and mortality were higher during the influenza season and the presence of other comorbidities increased the risk of hospitalization and death.
Influenza vaccination was associated with a decrease in hospitalizations for cardiac disease and cerebrovascular disease among a large cohort of patients 65 years and older from three managed care groups compared to members who were not vaccinated [42]. The mortality rate from all causes was also significantly lower among the vaccinated group.
