Support Provided ByLearn More
Body + BrainBody & Brain

1918 Flu: Expert Q&A

On November 27, 2006, Terrence Tumpey answered selected viewer questions about the 1918 flu virus, its recent revivification, and the implications of bringing it back to life.


Receive emails about upcoming NOVA programs and related content, as well as featured reporting about current events through a science lens.

Q: What do you see as the greatest benefit from reviving this virus? Romig Middle School, Anchorage, Alaska

Support Provided ByLearn More
Terrence Tumpey: The influenza pandemic of 1918-19 killed an estimated 20 to 50 million people worldwide, many more than the subsequent pandemics of the 20th century. The biological properties that confer virulence to pandemic influenza viruses are poorly understood. Reconstructing the virus has allowed us to conduct research to better understand how the individual genes of the 1918 pandemic virus contribute to the disease process. This research will provide important insights into the basis of virulence, not only for the 1918 pandemic strain but also for the currently circulating influenza A viruses. This kind of information will enable us to devise appropriate strategies for early diagnosis, treatment, and prevention, should a similar pandemic virus emerge. Additionally, such research will provide us with general principles with which we can better design antivirals and other interventions against all influenza viruses with enhanced virulence.

Q: It appears that the strain of bird flu everyone worried about has not materialized. What are the odds/chances that it will mutate to a more deadly, easily spread strain in the next 5-10 years? Is there some other more likely deadly viral candidate? Rick Carreiro, San Jose, California

Tumpey: Many scientists believe it is only a matter of time until the next influenza pandemic occurs. Influenza pandemics occurred three times in the past century—in 1918-19, 1957-58, and 1968-69—and were caused by different strains or subtypes of influenza. Therefore, there are many avian influenza subtypes that are considered potentially pandemic strains, including the H5N1 subtype virus commonly called "bird flu." The timing and severity of the next pandemic cannot be predicted, so scientists are unable to determine the odds/chances that any avian influenza strain will mutate to a more deadly, easily spread strain. Future research will help us understand whether or not the H5N1 subtype virus has the potential to become a strain that can easily spread from human to human.

Q: Is there a reasonable chance that a vaccine could be developed as a consequence of your research that could be employed in the event that the virus were to be released into the public? F. Joseph Peabody, Provo, Utah

Tumpey: The reconstruction of the 1918 virus will enable CDC to pursue development of a vaccine candidate that would provide optimal protection against this or similar viruses. There has been some progress on the evaluation of vaccines against 1918 influenza virus in animal models. Testing of influenza vaccines in animals is the first step in the evaluation or pre-clinical testing of vaccines. Previously, it was shown that vaccines containing the 1918 protein hemagglutinin (or HA) or other subtype H1 HA proteins were effective in protecting mice. In fact, the current influenza vaccine that Americans receive in the fall provided some level of protection against the 1918 virus in mice. More recently, in collaboration with Dr. Gary Nabel of the Vaccine Research Center, NIH, a 1918 HA DNA vaccine was evaluated for its ability to protect mice from a lethal challenge with the reconstructed 1918 virus. Mice challenged with 1918 virus were completely protected from death (see Kong et al., 2006; Proc Natl Acad Sci USA, 103:15987).

Q: What led you to undertake this investigation? Who came up with the idea? Thomas Edison High School, San Antonio, Texas

Tumpey: This investigation that led up to the successful reconstruction of the influenza A (H1N1) virus responsible for the 1918 "Spanish flu" pandemic actually started in 2001, and the investigations and ideas were a joint effort among scientists from CDC, Mount Sinai School of Medicine, the Armed Forces Institute of Pathology, and Southeast Poultry Research Laboratory. CDC scientists reconstructed the 1918 virus by using a common technique called reverse genetics. The natural emergence of another pandemic virus is considered highly likely by many experts, and therefore novel insights into pathogenic mechanisms of the virus could contribute to the development of prophylactic and therapeutic interventions needed to control pandemic viruses.

Q: My mother was at Linden Hall, a private school in Pennsylvania during the 1918 flu epidemic. She and two other girls did not get the flu and were made to deliver food to the door of the ill girls. My mother would never get a flu shot and never got the flu. She believed she was naturally immune. Is this possible? Janice Vansant, Hopewell, New Jersey

Tumpey: There is no medical evidence that humans are naturally immune to influenza. It is possible that humans may experience subclinical infections in which there are little or no overt symptoms of disease. Subclinical infections would still result in the development of immunity and protection against subsequent influenza infections.

Q: Was the 1918 flu strain any more virulent than other modern-day strains, or did that seem to be the case because of poor health care and ignorance? Martha, Zolfo Springs Elementary School, Zolfo Springs, Florida

Tumpey: The pandemic influenza virus of 1918-19 killed up to 50 million people worldwide, including an estimated 675,000 deaths in the United States. The pandemic's most striking feature was the unusually high death rate among healthy adults aged 15 to 34 years, which lowered the average life expectancy in the United States by more than 10 years. The current research evidence suggests that the 1918 virus was more lethal than modern-day influenza strains. It was found to be 100 times more lethal in mice as any other human influenza virus and resulted in the death of mice as early as 3-5 days after they had been infected. The first influenza was not discovered until 1930. Thus, in 1918-1919 people did not understand what was causing the deaths and therefore did not know how to take precautions to prevent influenza.

Q: If a person survived the 1918 flu, is he or she immune to later strains of flu? I was born Feb. 24, 1919, so I was in the womb most of the year 1918. I know of another man who was born in Feb. 1919, and he is still living. Would samples of my blood for analysis contribute to the ongoing research on that virus? William Conrad Lawson, Stoneville, North Carolina

Tumpey: In previous serology studies, we found that survivors of the 1918 influenza pandemic had antibodies that neutralized the reconstructed 1918 virus. This finding was consistent with data testing historical blood samples demonstrating that survivors of the 1918 pandemic had antibodies that neutralized classic swine (1918-like) influenza virus. We also found that individuals born after 1918 possessed cross-reactive antibodies to the 1918 virus. This is most likely due to the fact that 1918-like viruses circulated among humans until 1957. Individuals born after 1962 have very little to no antibody levels to the 1918 virus.

Q: Do we know yet why the 1918 flu was so deadly to young, otherwise healthy people? Do we have answers about whether surviving a deadly flu in the early 1890s was a factor? Anne Brennan, Vancouver, British Columbia, Canada

Tumpey: Your question is a good one and hopefully one that we will be able to answer with future research. In 1918, two age groups had an unusually high case fatality rate as a result of influenza: the very young (less than one year) and healthy young adults (ages 15-35). A third group made up of individuals over the age of 65 invariably has higher attack rates during influenza epidemics. If you were to plot a graph of the mortality rate by age group, it would result in what has been called a W-shaped mortality curve in 1918, whereas we usually see a U-shaped mortality curve for influenza, as the virus tends to attack the very young and the elderly. In order to fully answer your question, we need pre-1918 virus samples to understand what circulated before the 1918 pandemic.

I believe that the high mortality W-shape for the 1918 virus resulted from the combined effects of (1) the intrinsically greater virulence of the virus and (2) the fact that it struck a population that was mostly immunologically naïve. Therefore, if young healthy adults (ages 15-35) had good H1N1 immunity (as we do now through vaccination) in the years 1918-1919, then we probably would not have seen so many excess deaths in that age group. There is some belief that an H1-like virus was circulating before 1889 that provided protection in 1918 among individuals over 35. Without this prior H1 exposure of the population ages 35-65, the case fatality rate for the 1918 pandemic may have followed a different mortality curve, with many more adults dying or becoming seriously ill.

Q: I've heard two arguments on how the H1N1 virus jumped to humans. One argument states that it jumped directly from birds to humans. The other argument states that pigs were a mixing vessel between the avian virus and a currently circulating human virus. Which argument, in your opinion, is correct and why? Tim Borger, Phoenix, Arizona

Tumpey: The identification of the direct precursor of the 1918 pandemic virus will enable us to better answer such questions. Thus, we need the genetic sequence of pre-1918 virus samples to answer this question with certainly. Looking at the evolution of human and swine H1N1 influenza strains, the available evidence suggests that the direction of transmission was most probably from humans to pigs. There are numerous examples of human influenza A virus strains infecting swine since 1918, but swine influenza strains have been isolated only sporadically from humans.

Q: How do you know that you have the exact sequence of genes that made the Spanish flu virus of 1918? Anonymous, Jefferson, Ohio

Tumpey: The identity of virus genes in the 1918 rescued virus was confirmed by reverse transcription polymerase chain reaction and sequence analysis. The sequence matched the sequence analyzed in archived lung tissue samples prepared during autopsies of patients who died from severe pneumonia in September 1918.

Q: My dad died at age 67 from rigid Parkinson's disease about 50 years after the 1918 flu epidemic, in which he had a case of the flu and survived. We were told that his form of Parkinson's was common among 1918 flu survivors and was thought to be related to the 1918 flu virus. Is this correct? Thank you. Bill St. James, Florence, Massachusetts

Tumpey: At this time, we have no evidence of the 1918 virus going to the brain of humans, nor have we detected the virus in the brains of 1918 virus-infected animals. There is no sound evidence that the 1918 virus was related to Parkinson's disease.

Q: A coworker of mine thinks the 1918 influenza deaths may be overstated because the real causes of some deaths were embarassing. He says imagine a family in which someone has died of syphillis. It would have been more acceptable to say he got the flu than to admit the truth. Perhaps some doctors were even persuaded to put influenza on the death certificates when they shouldn't have. Does this notion have any credibility? Rick Wiegmann Koshko, Ottawa, Illinois

Tumpey: There is no evidence to suggest that this form of record keeping took place in 1918. For a host of reasons it is difficult to be precise on the exact numbers of individuals that died in the years 1918-1919. That is why you see such a broad range of deaths reported (20-50 million). Influenza was not a reportable disease in 1918-1919, as the first influenza strain would not be isolated until 1930. Actually, the culprit was widely held to be a bacterium called Pfeiffer's bacillus, which added to the confusion at the time. Furthermore, during the 1918 pandemic, not every country had the same level of record keeping, making it difficult to obtain exact numbers of individuals that died as a result of pneumonia.

Q: I have read that this flu was the result of immunization of the troops with a strong vaccine concocted to control rampant typhoid. The troops spread Spanish flu as they returned home after the war, and people at home were immunized to protect them from the returning soldiers. What information do you have about this? Diane, Nanaimo, British Columbia, Canada

Tumpey: Although it was likely that WWI troops contributed to the spread of the 1918 influenza strain, there is no evidence that the 1918 influenza pandemic started as a result of immunization of troops. During the 1918-1919 pandemic there was no influenza vaccine available.

Q: Webster describes a cross-reactive, T-cell-mediated immunity induced by H9N2 viruses providing protection from H5N1 virus infection in chickens. The findings showed that prior infection with H9N2 virus having highly homologous internal genes provided protection from signs of disease caused by highly pathogenic avian H5N1 influenza viruses.

In Austria, swans have been found with humoral H5 antibodies. The question is, how could these swans survive, overcome H5N1 infection, and develop antibodies?

Would you say that there might be underlying the same mechanism of cell-mediated cross-protecting immunity, which could have moderated the disease in the swans, and if so, what strains could be involved in this process? Bauch MD, Munich, Germany

Tumpey: The avian influenza H5N1 virus, which emerged in 1997, reemerged in 2003 and is spreading among birds across Europe, Asia, and Africa and has resulted in more than 250 laboratory-confirmed human infections, with an approximate 58 percent fatality rate. Research suggests that currently circulating strains of H5N1 viruses are becoming more capable of causing disease in wild birds, including swans, than were earlier H5N1 viruses. In fact, close contact with and de-feathering of infected wild swans was considered to be the most plausible source of exposure to influenza H5N1 virus infections in the Republic of Azerbaijan, February-March 2006 (Salyan cluster), and H5N1 strains deposited at the State Collection of Viruses of the Russian Federation were isolated from the cloacal/tracheal swabs of sick mute swans. So H5N1 viruses continue to evolve and adapt to species other than chickens and ducks.

As a result, more research is needed in birds to characterize the role of immunity in the protection against virulent H5N1 viruses. The understanding of avian immunology is lagging far behind from what is known in mammals, and scientists do not fully understand how some swans survive H5N1 virus infection. Furthermore, it is unknown if the immune protection is mediated by antibody and/or cell-mediated immunity (CMI) to H5N1 virus infection. Traditionally, the immune response has been divided into the humoral (antibody) and cell-mediated response. Cell-mediated immunity is specific immunity mediated by T lymphocytes and has been suggested to be an important factor to the development of protection in chickens vaccinated against viral diseases. The subsets of T lymphocytes: CD4 helper cells and CD8 cytotoxic cells (CTL) constitute the principal cells of the CMI response. For influenza, CD8+ CTL play a crucial role in controlling infectious virus from the lungs of mice. In fact, influenza virus-specific CTL generated through vaccination or introduced by adoptive transfer in animals lead to a more rapid viral clearance and recovery of the host and protection from death. There is limited data on CTL responses to influenza viruses in birds, and it is generally believed that determination of the immune effector(s) mediating infection and even vaccine protection is crucial for the design of improved vaccines.

Q: Why did the 1918 flu die out? Did everyone susceptible die? Frankly, I have never understood how an epidemic can end short of infecting almost everybody.

Also, if the bird flu from an infected human can't infect another human, how did it get into the human in the first place?

Absolutely excellent show. My wife and I were impressed with how much material we learned, and how well you presented it. Thanks. And, of course, we want more. Christopher West, Purchase, New York

Tumpey: There is some belief that an H1-like virus was circulating before 1889 that may have provided protection in 1918 among individuals over 35 years of age. That would partially explain why humans had different susceptibilities to influenza virus infection in 1918-1919. As far as the 1918 influenza virus waning over time, 1918-like viruses circulated among humans until 1957 and then reemerged again in 1977. It is thought that influenza pandemic strains have a gradual decrease or decline in strength due to the host adaptation process (the process in which the virus changes in order to replicate better in a particular host). The immune response of humans will provide pressure to cause influenza strains to mutate and some of the mutations are thought to weaken the virus.

For H5N1 virus infections it has been demonstrated that all viral gene segments were found to be of avian origin, indicating that the H5N1 human cases in Hong Kong were the result of multiple independent transmissions of the virus from birds to humans. To date, avian influenza H5N1 viruses have not acquired human influenza virus genes and thankfully lack the ability to spread efficiently from human to human. However, the repeated transmission of H5N1 viruses from infected poultry to humans could increase the likelihood of the emergence of an avian-human reassortant virus and/or an avian H5N1 virus that has acquired the essential molecular trait(s) needed for efficient and sustained transmission among a serologically naïve human population.

Q: What do you mean a virus needs a special gene to get freed from the cell? Doesn't a bacteriophage (active virus), after taking over the cell's normal functions, start "mass production" of virus parts, and when they are assembled, the cell bursts because of the overwhelming of viruses? Anonymous in 6th grade, New Jersey

Tumpey: For some viruses, such as influenza, the assembly of the virus occurs on the inside of the cell membrane, and as the process is completed, the virus moves through the cell wall and attaches on the surface. Neuraminidase is a major surface structure protein of the influenza virus that is an essential enzyme for the spread of the virus throughout the respiratory tract. It enables the virus to escape the host cell and infect new cells. You can imagine that once an influenza virus particle "buds" from the surface of the infected cell that it would stick to sialic acid using its hemagglutinin (H or HA). The HA is the other major surface structure protein that enables the virus to attach itself to a cell in the human airway. So it is believed that the major function of viral neuraminidase is at the final stage of infection when NA cleaves sialic acid from cell surface and influenza virus, allowing the virus release from infected cells.


  1. Did I hear correctly that all flu comes from birds? If so, why would any flu be singled out and called "the bird flu?"
  2. Was the footage shown really shot from inside the restricted area of the lab?
  3. How much time did it take from conception of the idea to recreate the 1918 flu to the time it was realized?
  4. Is your life at risk every time you handle the flu, and how often, on average, is that?
  5. Just a comment: if a movie is EVER made, you so should be played by Matt Damon.

Michelle K., Colorado Springs, Colorado

Tumpey: "Bird flu" is a phrase similar to "pig flu" or "human flu," and it generally refers to an illness caused by any of many different strains of influenza. Because the one type of influenza subtype (also called, "avian influenza H5N1") is in the spotlight, it has been most often referred to as "bird flu."

This investigation that led up to the successful reconstruction of the influenza A (H1N1) virus responsible for the 1918 "Spanish flu" pandemic actually started in 2001, and the investigations and ideas were a joint effort among scientists from CDC, Mount Sinai School of Medicine, the Armed Forces Institute of Pathology, and Southeast Poultry Research Laboratory. The work described in the show was done using stringent biosafety and biosecurity precautions that are designed to protect workers and the public from possible exposure to this virus (for example, from accidental release of the virus into the environment). Biosafety Level 3 or Animal Biosafety Level 3 practices, procedures, and facilities, plus enhancements that include special procedures (discussed in the next question below), are recommended for work with the 1918 strain.

Q: It seems to me that to establish exactly what it is within the 1918 H1 hemagglutinin that is responsible for its deadliness, you would need to be able to mutate some precise RNA amino acids within the 1918 H1N1 that when then injected in mice do not kill them. Has this been done? Nad Nathan, Madison, Wisconsin

Tumpey: No, this has not been done yet, but you raise an important area of research that needs to be pursued in the future.