Win the adventure of a lifetime!

UNITED24 - Make a charitable donation in support of Ukraine!

Homeland Security


II. The U.S. Experience with Influenza Over the Last Century


Seasonal influenza epidemics occur every winter and are estimated to cause some 30,000-40,000 deaths in the United States alone, primarily in young children, the elderly, and others with underlying medical conditions. Several times in the last century new subtypes of influenza have swept through the human population—which has little or no immunity to them—and caused global pandemics.

Preparation for influenza pandemics is shaped in large part by the experiences of the pandemic of 1918-19, when 40-100 million people perished worldwide, and the swine flu “fiasco” of 1976, when 45 million Americans were vaccinated for a virus that never spread beyond a tiny cluster. In 2005, concerns that the highly lethal avian H5N1 virus could precipitate an influenza pandemic led to significant invest­ ments and improvements in Federal preparedness, although significant transmission among humans has fortunately not occurred to date. Based on available information, the influenza pandemics most analogous to the current 2009-H1N1 outbreak may be those of 1957 and 1968, in which the death rates were two- to four-fold higher than normal.

The main lessons from these experiences are that vigorous preparation and action can save lives, but that it is critical to maintain situational awareness and flexibility as a pandemic unfolds.


Seasonal influenza epidemics occur every winter, peaking between December and February, and are estimated to cause 30,000 to 40,000 deaths in the United States alone, primarily in children under age 2 and adults over age 65, and more than 250,000 hospitalizations per year. The economic impact of seasonal influenza is estimated at $37 billion each year.

Since 1977, two influenza A virus subtypes and one influenza B subtype have circulated each winter. Seasonal influenza viruses undergo frequent mutations that can cause small changes in proteins nec­ essary for entry into human cells, allowing them a measure of protection against immune responses, even in people who were infected with prior strains. (Influenza viruses are classified by these proteins, called hemagglutinin [H] and neuraminidase [N]; see Box 1A in Chapter 1.) This genetic variability means that people experience repeated influenza infections over their lives and vaccine formulations must be updated nearly every year.

At irregular intervals, new subtypes of influenza burst on the scene and sweep through the human population, which has no significant immunity to them. Such global pandemics appear to occur three to four times per century. In the twentieth century, pandemics were caused by new variants of influenza

A virus in 1918 (H1N1), 1957 (H2N2), and 1968 (H3N2). These pandemics varied in severity, for reasons related both to the level of pre-existing immunity in the human population and to the genetic makeup of the virus. In spite of extraordinary scientific advances in understanding influenza viruses, they remain highly unpredictable.

Responses to new pandemics should be informed by historical experience. The severe pandemic of 1918–19 offers some lessons about the benefits of rapid action, and the swine flu vaccination campaign of 1976 is instructive about the risk of an overly aggressive response to an unproven threat. Over the past five years, the emergence of human cases of a highly pathogenic avian H5N1 influenza has stimulated unprecedented pandemic planning efforts.

1918–19 Influenza Pandemic

The 1918–19 pandemic was the worst natural calamity of the twentieth century, with an estimated mor­ tality worldwide of 40–100 million lives. In the United States, between 500,000 and 750,000 perished at a time when the U.S. population was one-third its current size. In contrast to seasonal influenza, mortality was especially high among previously healthy young adults. In cities that adopted early measures of “social distancing,” such as cancelling public gatherings and closing schools, the epidemic appeared to have spread more slowly and reached a lower peak incidence.

What is most informative about the 1918–19 pandemic for current planning purposes is its pattern of spread. A first, or spring, wave began in March 1918 and spread unevenly across the United States, Europe, and Asia. Although illness rates were high, death rates in most locations were not significantly above those of seasonal influenza. The spring outbreak was mild enough that the public health and medical communities saw no cause for alarm. However, a second, fall wave spread globally from September to November 1918, with death rates approximately ten-fold higher than in the spring. Cities that responded rapidly by closing schools, churches, and theaters, restricting public gatherings, and otherwise discouraging social interaction appear to have reduced transmission and mortality while the measures were in effect. However, most cities could not sustain these measures, and many experienced the return of influenza as control efforts lapsed. In some places, a third wave occurred in early 1919. Death from pneumonia was a hallmark of the 1918–19 fall and winter waves. The 1918–19 pandemic vividly illustrates what can happen when the public health and medical communities lack knowledge, contingency plans, and effective vaccines or treatments.

1976 Swine Flu “Fiasco”

The events of 1976 serve as an example of a public health response premised only on the “worst case” scenario, which ended up being a false alarm. In January 1976, a novel H1N1 virus first appeared in a group of army recruits at Fort Dix, New Jersey. Four were hospitalized and one died. In March, on the advice of public health experts, President Gerald Ford announced on television that he was asking Congress for $137 million “to inoculate every man, woman, and child in the United States” against swine flu. Within 10 weeks of the launch of the fall vaccination campaign, about 45 million people, or 1 in 4 Americans, had received swine flu immunizations. Public confidence, however, was soon shaken by the deaths of three elderly adults in Pittsburgh soon after they received their swine flu shots. Although such events are expected by chance, local public health officials and the media raised the possibility that the deaths were due to the immunizations. Later reports found Guillain-Barré syndrome, a paralyzing neuromuscular disorder, to be associated with 1976 vaccination at a frequency of approximately 1 per 100,000 vaccinations. With no disease from the swine flu virus having appeared since the outbreak at Fort Dix, even this relatively rare complication was enough to lead to the suspension of the immuniza­ tion program.

The key policy error in 1976 was to bundle all decisions (e.g., make the vaccine, immunize everyone, make a Presidential announcement) into a single “go” or “no-go” decision, with no provision for the monitoring of the situation and continual reconsideration of policy directions based on new evidence. The experience of 1976 highlights the challenge of coordination horizontally across different agencies of the Federal Government; vertically across the various levels of government (Federal, state, local); among public officials and health professions and institutions; and between the public and private sec­ tors. The 1976 swine flu immunization program highlighted other lessons, including the importance of communication to the public, the long-term need to preserve credibility, and the need for preparations relating to vaccine liability insurance anticipation of coincident deaths in a mass immunization program, the potential impact of vaccine side effects, and the role of chance.

In applying these lessons to present circumstances, it is worth noting a number of crucial differences between then and now. Among them: (1) the current 2009-H1N1 is continuing to spread, unlike the single, self-quenching outbreak at Fort Dix; (2) a wider array of interventions, including antiviral medica­ tions, is available; (3) more sophisticated characterization and surveillance systems for circulating viral strains are in place; (4) the Federal Government has a more complex structure and a larger number of relevant agencies and officials; (5) the media are vastly more varied and operate on a continuous news cycle; and (6) widespread international travel contributes to accelerated transmission around the world.

Other Pandemics

The two other influenza pandemics in the last century also provide insight into the current situation. The so-called “Asian Flu” of 1957 appeared in the United States in late spring. Small outbreaks occurred over the summer, but transmission accelerated in the late summer through early fall, peaking in October before vaccine supplies were widely available. Public health authorities learned an important lesson about the potential value of early protection against influenza. After a lull following the October peak, there was a smaller upsurge in transmission in early 1958, in which the elderly were disproportionately affected. Mortality in 1957–58 was high—an estimated 70,000 deaths in a population of about 170 million. Although death rates were highest among the elderly, about 30 percent of all deaths occurred in those under age 65.

The mildest of the twentieth century pandemics occurred in 1968, with an estimated 34,000 deaths in a population roughly two-thirds of today’s; death rates were highest in the elderly, but about half of all deaths occurred in people under age 65. The 1968 virus emerged first in Hong Kong in mid-1968 and appeared in the United States in September, but did not peak until December 1968/January 1969. One reason for the slower spread and reduced death toll during the 1968 pandemic may be that the virus (H3N2) shared some similarities with the virus that was already circulating (H2N2), so the population may have been partially immune.

In the decade following each of the twentieth-century pandemics, seasonal epidemics continued and excess deaths in younger age groups remained elevated above normal seasonal levels. These recent pandemics illustrate that the timing of peak pandemic activity may be earlier than that of normal flu season, but unpredictable in that younger age groups suffer more during pandemics than during sea­ sonal influenza outbreaks, and that the impact of new strains on these younger groups persists into subsequent seasons.

Avian Flu

In Hong Kong in 1997, a highly pathogenic avian H5N1 virus was found to have infected large numbers of poultry and a small number of humans. Following initial control by extensive slaughter of poultry flocks, the virus disappeared, only to reappear in 2003–2004. The virus was felt to be a potential pan­demic threat because, although the rate of bird-to-human transmission was low and person-to-person spread was rare, the mortality rate was over 60 percent.

The possibility that H5N1 could acquire the ability to transmit efficiently between humans and thereby start a new and severe pandemic spurred major pandemic-planning efforts at the state, Federal, and global levels. Beginning in 2005, the Federal Government undertook a number of initiatives to address this threat, including: (1) developing a “National Strategy for Pandemic Influenza” to guide the pre­ paredness efforts of Federal departments and agencies, state and local authorities, businesses, and the public; (2) requesting that Congress appropriate $7.1 billion to establish a domestic stockpile of antiviral medications and pre-pandemic vaccine and to significantly expand domestic influenza vac­ cine production capacity; (3) developing guidance on pandemic influenza preparedness for the public and a broad spectrum of stakeholders outside of the Federal Government; (4) establishing policies to guide the pandemic response in areas such as border management and prioritizing allocation of pre- pandemic vaccine; and (5) creating the International Partnership on Avian and Pandemic Influenza to facilitate global surveillance and preparedness actions.

These and other efforts in response to the H5N1 threat have informed and guided many of the actions undertaken in response to the 2009-H1N1 outbreak to date.

Lessons for Fall 2009

Given the concern about avian influenza and awareness of the catastrophic results of the 1918–19 pan­ demic, much of the effort for pandemic planning has been directed toward responding to an extremely severe pandemic. This worst-case-scenario planning has led to improvements in the efficiency of vaccine production and testing, stockpiling of antiviral drugs, and other measures that will be valuable in the fall, if used appropriately. However, unless the severity of the 2009-H1N1 influenza increases markedly, it is unlikely that community mitigation on the scale envisioned for a more severe pandemic will be required. On the other hand, as described in the next chapter, it is already clear that the current pandemic is no false alarm (as in 1976) and has the potential to cause serious health consequences, especially in relatively young age groups and in individuals with certain pre-existing medical conditions.

While the features of 2009-H1N1’s next wave cannot be accurately predicted, history teaches us that the most effective responses will be achieved by advanced planning, knowledgeable judgments about the range of possible events, continued situational awareness about the pandemic, and flexibility in thinking and decision making.

Join the mailing list