John Barry, Historian and Author
Quote: “What we learn from history is that we do not learn anything from history,” George Bernard Shaw.
Numbers: No one really knows how many people died, but the first contemporary estimates put deaths at 21 million worldwide. That estimate is often still quoted, but it was certainly far worse than that. MacFarlane Burnett spent much of his life studying influenza, and he concluded that the worldwide death toll (from the 1918–1919 pandemic) was a minimum of 50 million, possibly 100 million. He may well have been right.
If you adjust for population, that could be 300 million today. In three weeks, it killed more people than AIDS has in 24 years. In the U.S., a reasonable estimate starts at 500,000 deaths. Personally, I am inclined toward a higher estimate of 700,000.
There was 15–53% morbidity. In San Antonio, over 90% of households had at least one family member with the illness. Most of the deaths occurred in healthy adults. In Army camps, the death toll was routinely over 5%, sometimes as much as 10%, and in some communities as high as 30%.
A human face: One physician at Camp Devens outside Boston wrote a colleague:
“These men start with what appears to be an ordinary attack of LaGrippe or Influenza, BUT when brought to the Hosp. they very rapidly develop the most vicious type of Pneumonia that has ever been seen. SOON YOU see the Cyanosis extending from their ears and spreading all over the face, until it is hard to distinguish the coloured men from the white.”
“It is only a matter of a few hours then until death comes... It is horrible. One can stand it to see one, two or twenty men die, but to see these poor devils dropping like flies... We have been averaging about 100 deaths per day... Pneumonia means in about all cases death... It takes special trains to carry away the dead. For several days there were no coffins and the bodies piled up something fierce. It beats any sight they ever had in France after a battle... Good By old Pal, God be with you till we meet again.”
That letter reflected the typical experience in 1918 army camps. It was sure that these men would end up in death. It was a horrible sight to look upon because the cyanosis was such that most physicians had never seen it before.
A big problem was you never knew if you were going to have the lethal kind, or normal flu. Regardless, it was gruesome. There were many with broken ribs from violent coughs. Air pockets also formed outside the lungs, and when you’d move these people it would sound like popping bubble wrap. Nose bleeds were common, less common was bleeding from mouth, and even bleeding from the eyes and ears.
In 1957 during the Asian Flu pandemic, 25% of all deaths were directly from viral pneumonia. In 1918, there were probably more than that. Antibiotics, while useful, are not necessarily a solution if the cause of pneumonia is viral.
There were enough anecdotal reports from reliable observers that prove some deaths were extraordinarily rapid, occurring in 24 hours or less. While most, even from acute respiratory distress symptoms, would succumb four to six days after their first symptoms, for others death would occur in two to three weeks.
Context: We were at war. It was the first total war. It was the first time the government tried to fully control the public. The pandemic was known as Spanish flu, but we know it did not start in Spain. It was called Spanish Flu because Spain was not at war, so there was a free press. Most of Europe had a censored press. The U.S. press was more open, but still the new Sedition Act was passed. This was a law that made the Patriot Act look like it was written by the ACLU (American Civil Liberties Union).
To enforce the law, the Justice Department issued badges to the “American Protective League” (APL), who was identified as Secret Service. Within a year, the attorney general said 200,000 APL members were operating in 1,000 communities.
In Chicago, a “flying squad” of league members and police trailed, harassed, and beat members of the International Workers of the World (IWW). In Arizona, league members and vigilantes locked 1,200 IWW members and their “collaborators” into box cars and left them on a siding across the state line in the New Mexico desert. Throughout the country, the league's American Vigilance Patrol targeted “seditious street oratory,” sometimes calling upon the police to arrest speakers for disorderly conduct, sometimes acting more directly. Everywhere, the league spied on neighbors, investigated “slackers” and “food hoarders,” and demanded to know why people didn't buy—or didn't buy more—liberty bonds.
At the same time, states outlawed the teaching of German. In Providence, Rhode Island, the Providence Journal carried a banner warning that “every German or Austrian in the United States unless known by years of association should be treated as a spy.” The problem was that the single largest ethnic group in the United States was German. 1918 was a time when there was a well-run propaganda machine. The government created the Committee for Public Information (CPI). It was intended to create one white-hot mass of determination. A quote by Arthur Bullard (who directed the CPI's Russian Division) summarizes the tenets of propaganda at the time, “Truth and falsehood are arbitrary terms... There is nothing in experience to tell us that one is always preferable to the other... There are lifeless truths and vital lies... The force of an idea lies in its inspirational value. It matters very little if it is true or false.”
Part of the propaganda machine included the energetic direction of Mississippi newspaper editor George Creel, an advisor to President Wilson. Under Creel, the CPI churned out national propaganda through diverse media. Creel organized the “Four-Minute Men,” a virtual army of volunteers who gave brief speeches wherever they could get an audience. There were 150,000 of the Four-Minute Men, who would propagandize for “four minutes” prior to any kind of public gathering. These men were charged with creating fear in the population because it was “an important element to be bred in the civilian population. It is difficult to unite a people by talking only on the highest ethical plain. To fight for an ideal, perhaps, must be coupled with thoughts of self-preservation.”
There was intimidation that was enforced by the law; in fact a U.S. Congressman went to jail for 15 years for violating the law. Several journalists and editors were jailed too.
The 1918–1919 influenza pandemic swept across the U.S. during a time when patriotism was more important than truth. Thus, intimidation and propaganda were part of the communication culture.
A pandemic meets the propaganda machine: During this time at least 25% of doctors and nurses were in the military. Additionally, national, local, and state officials all operated in the same way. At best, they communicated half-truths, or even out-right lies. As terrifying as the disease was, the officials made it more terrifying by making little of it, and they often underplayed it. Local officials said things like “if normal precautions are taken, there is nothing to fear” but then they would close all businesses.
“Worry kills more people than the disease itself,” a Chicago public health official was quoted as saying. Other quotes were: “Don’t get scared,” and “The so-called Spanish influenza is nothing more or less than old fashioned grippe.”
Communication was rarely honest, because honesty would hurt morale. One of the first newspapers that started telling the truth in Milwaukee saw its editor jailed, so they stopped telling the truth. In Philadelphia, after a public health official finally closed all public gatherings and public funerals, the newspaper said, “This is not a public health measure.”
There was a lot of cognitive dissonance. People heard from authorities and newspapers that everything was going fine, but at the same time, bodies were piling up. Imagine your spouse lying dead in bed for six to eight days. There were coffin shortages. The dead were piled up where they died. There were police going around asking people to “bring out their dead.”
Panic was incipient. Even if there was war, the war was removed from us. The fear was so great that people were afraid to leave home or talk to one another. Everyone was holding their breath, almost afraid to breathe, for fear of getting sick.
Army camps were shut down. People didn’t want to let soldiers return home. People were afraid to have any intimacy with their community and loved ones. Nobody would bring food in or come to visit.
In the hills of Kentucky, there were hundreds of cases. People were panic stricken and many were starving, not even for lack of food, but for fear of going near each other.
In Phoenix, a rumor started that dogs carried influenza, and people were shooting their pets. They’d give their pets to the police to kill them, because they couldn’t kill them themselves.
At that time, the economy was controlled. The national government did little to help. The Public Health Service (PHS) was given $1 million, but did next to nothing and ended up returning most of the money.
There were four groups (local charitable organizations) that did help. This was actually the American Red Cross’ “golden age.” At that time, Red Cross groups were headed by people of great prominence. They begged people to come in to volunteer. They took out a paid half-page advertisement that said: “The safety of this country demands that all patriotic available nurses, nurses' aids [sic] or anyone with experience in nursing place themselves at once under the disposal of the Government... Graduate nurses, undergraduates, nurses' aids, and volunteers are urged to telegraph collect at once... to their local Red Cross chapter or Red Cross headquarters, Washington, D.C.”
Liberty Loan groups were in communities trying to help. This was one extraordinarily organized group. They had a “captain” for every block and every large apartment building. Other private sector organizations were horizontally integrated so that everyone talked to each other. In many ways, it was better than today.
Despite these efforts, absenteeism remained a problem. At the L.H. Shattuck Company, 45.9% of the workers stayed home. At the George A. Gilchrist shipyard 54.3% stayed home. At Freeport Shipbuilding 57% stayed home. At Groton Iron Works 58.3% stayed home. Even in an industry that was crucial to the war effort, absenteeism was high. Fear and terror was akin to the Black Plague.An internal American Red Cross report concluded, “A fear and panic of the influenza, akin to the terror of the Middle Ages regarding the Black Plague, [has] been prevalent in many parts of the country.”
One report noted that not a single automobile was moving on the street in Manhattan. In Philadelphia, a medical student was able to count cars on the way in and said the city was dead. It seemed at the time like civilization could easily disappear from the face of the earth.
Is this an over-statement of how the pandemic affected our country? Probably; there was one woman I met who had no recollection of the pandemic. However, it is important to note that society worked better in places where authorities told the truth. San Francisco was an example. In San Francisco authorities took out advertisements that said: “Wear a mask and save your life.” Masks probably didn’t help, but people trusted government more and they organized to feed people. San Francisco just seemed to keep operating better.
In 1918, Departments of Public Health were often referred to as Departments of Public Assurance.
Take away message: False reassurance is the worst thing you can do. Don’t withhold information, because people will think you know more. Tell the truth— don’t manage the truth. If you don’t know something, say why you don’t know, and say what you need to do to know. Drown people with the truth, rather than withhold it.
Richard Hatchett:
- The previous speaker set the context well. It is important to understand that it was a special time in our history. WWI was at a crisis. It wasn’t clear that the war was about to end. Spanish Flu showed up right at the time that the German lines were breaking.
- At that time it was the equivalent of “9-11” happening every day. For example, assassination attempts against Lenin were on page 12, ahead of this story…the Czar had been executed.
- The pressure to maintain the war effort made so many other things lose importance.
- The overall case fatality rates were extremely high in younger populations. Background fatality rates for children were quite different than today.
- 1918 case fatality rates in one-to-four year-olds is equivalent to what it would take now for 15–16 years of fatalities. This relates to a huge surge. It is unfathomable, really.
- Public messaging: Disease was so widespread, nothing good would result from public action. Fear was more fearsome to the officials than the disease.
- By the time they closed schools in Baltimore, it was too late. It was important to do something about it, right away (e.g., need to not have a lag at all in taking action).
- To put this in perspective, the median time between onset and time of death was eight to 10 days. Additionally, the interpretation of data was delayed.
- Remember that at that time, nothing was done to interfere with public morale.
John Barry:
- Many times public health officials knew the truth but did not tell it. This was the case in Philadelphia.
- Boston had already exploded in chaos several weeks earlier.
- In many cases they were just plain lying.
- Newspapers were pretty much unbelievably bad sources.
- Newspapers were writing about it when it was far away, but went silent when deaths came to their own city.
- The attitude of authorities was: “This isn’t happening, don’t worry about it.”
Q: A significant amount of data show that pregnant women died disproportionately. Of course, back then many women gave birth in the home. Can you tell us how the pandemic affected pregnant women?
A: (Hatchett) There was certainly an additional strain on pregnancy.
- Maternal mortality rates peaked in 1918 with 916 per 100,000 births. That means one per 1,000 births. The multiplying factor is enormous. There is not, however, a lot of specific data about pregnancy during that time.
- We’ve run several flu exercises to look at a number of different issues, including pregnancy, because back then hospitals were different than today. In these exercises, we have run scenarios from crisis to Katrina-like problems. In all of these exercises a key issue has been “transparent communication.” Today, we have the opportunity to frame how the public perceives what is happening. What seems to be the problem early on is they aren’t able to shift rules of engagement for behavior, so we miss opportunities to act until it is too late.
- Today, I think, as opposed to back in 1918, we don’t have as much of a problem with misinformation as the lack of transparent information. We also have a situation in which institutions will need to operate under different sets of rules in the case of treatment protocols, including those for pregnancy.
A: (Barry) I want to emphasize that it is not likely that public health officials would tell outright lies. However, it is just as important not to hold back information. I was just at an exercise last week and a public health official said, “The last thing I want to do is have a press conference.” This is the wrong approach. You really need to get out in front right away. For example, you could get out and say we know it is in the Orient, but we don’t know if this individual is the first case, so we need to start getting ready right now. Don’t wait. We need to worry about being too conservative. I can’t emphasize enough how much you need to get out in front of it.
Q: While there were no definite countermeasures known to be effective, could you say that even anecdotally there would be better outcomes in places that had better communication?
A: (Barry) The key is trust. It is when people feel totally alienated and isolated that the society breaks down. Telling the truth is what held society together.
- Today we should look at the nonpharmaceutical interventions and emphasize them early on.
- There is a lot of work being done to determine if the risk communication provided actually improved social issues.
A: (Hatchett) There is a tremendous amount of wishful thinking that the virus won’t come here. In 1918 the shaping of the cognitive environment varied dramatically from location to location. In some places elected officials and public officials locked arms on some things. Some tentative evidence shows that social distancing interventions did help. However, Baltimore is a case example of how to do it wrong (e.g., not close the schools). In Baltimore there were fights between elected officials and public health officials. Another example of doing it wrong was Pittsburgh. In Pittsburgh, the Mayor actually told the public to ignore the public health officials.
Q: Can you expand on the issue of mortality and morbidity for healthcare workers?
A: (Barry) There is no definitive data. I can say that often death notices for doctors and nurses showed up in tiny print in the newspapers. In many places the rates were high; morbidity over 50%. Logic suggests that social distancing measures helped, but I am not convinced that those interventions had an effect that was anything more than random. One can’t really generalize, because morbidity and mortality rates varied so much from community to community.
A: (Hatchett) There is no aggregate data on physician and nurse deaths. However, I too wouldn’t be surprised if it wasn’t over 50%.
Exercise: Note cards were turned in from audience members to collect thoughts and concerns from the group. A summary of these audience-generated questions and issues can be found in Appendix D of this document.
Dr. Richard Hatchett, Associate Director for Radiation Countermeasures Research and Emergency Preparedness at the National Institute for Allergy and Infectious Diseases
Note about the speaker: While most people are involved in preparing for a pandemic in one capacity or another, Dr. Hatchett has devoted the last five years to public health emergency preparedness and a considerable amount of attention to disease containment, particularly in an environment where there may not be adequate stockpiles of medications.
Navy midshipmen learning example: In Trafalgar, Lord Nelson was outnumbered and outgunned, so he split his fleet and came alongside the French and Spanish line to achieve his comparative advantage because his ships were much more maneuverable. In doing so, he destroyed the French and Spanish fleet. Although he died in battle, he achieved through tactics and maneuvers what he could not achieve through brute force. This can be applied to our context today for our analysis of 1918. Context is important because environment and circumstances change. What may work in one time and place may not work in another.
Another military example: General Grant advanced on Vicksburg very quickly. In Virginia, he developed a completely different strategy…to grind down Lee’s Army. He used totally different strategies in each campaign because he was in completely different environments. In both cases he faced foes with different sets of preoccupations. The point is that Grant understood his opponents. He devised the correct strategy for each campaign. He did not try to fight the previous battle of Vicksburg in Virginia. Fighting the previous war is often a good recipe for disaster if used to fight the next war. This could be applied to creating strategies of approach for planning for a pandemic today.
To take what was done in 1918 and apply that to today is to learn to deal with handling a pandemic in our environment and communities. Historical perspective is critical for deriving plans for today. For example, deriving generalizations from a close reading of the historic record is beneficial because:
- a poorly mitigated pandemic will overwhelm medical resources;
- in the absence of prior planning and an agreed upon plan of action, conflicts may emerge that may retard the emergency response;
- figuring out ways to gathering information in a timely fashion during difficult circumstances will be crucial;
- it is better for political leaders to be truthful rather than minimize what is happening; and
- public health leaders need to realize that you can’t make everyone happy in a pandemic.
Another question that policy makers will want answered from history may be whether or not schools should be closed. Framed in the 1918 experience, that leads to a series of other questions:
- Is there evidence that closing schools made a difference?
- Is the timing of school closings important?
- If it made a difference in 1918, would it make a difference today, given the changes in demographics?
Today, we have a much more developed sense of critical infrastructure and interdependency between the structures than our grandparents did. Our grandparents:
- had a clear concept of the war effort,
- knew what industries fit into the war effort,
- had a developed notion of public health and medical infrastructure, and
- had more advanced local level concepts, which have been eroded over the last several decades.
How the 1918 pandemic affected infrastructure:
- Infrastructure was less complex than today.
- There was less dependence on just-in-time delivery of key resources and materials.
- There were more natural firewalls, as opposed to cascading firewalls.
- National power grids did not exist.
- Americans were not dependent upon foreign oil.
- Food distribution was more extended.
- Communities were more self-sufficient and more insulated from each other.
Today, extended networks create the potential for cascading failures, but also increase the resiliency for critical infrastructure against local failures. Today, we should not focus on the critical first weeks; instead, we should focus on the aggregate effect over time.
It is important to note that sectoral failures did not occur in 1918. So, while the October draft was postponed for about three weeks and productivity in shipyards was compromised, the war effort continued. The war effort continued even when New York City reported a 40% decline in productivity in shipyards on October 18th.
Los Angeles introduced interventions during the first few weeks of the epidemic, much earlier than New York, so they did not suffer the types of problems New York experienced. Baltimore newspapers published appeals to the public to minimize the use of telephones, but even at the height of the pandemic the phone system did not fail. Chicago attempted to estimate absenteeism. They estimated absenteeism averaged about 5% over a two-month period, not quite double the normal average. Absenteeism peeked on October 22. However, the aggregate impact of absenteeism was significantly reduced. Economic impacts were modest, below 5%, when averaged out. A major recession did not ensue. Retail sales did decline in October, but they rebounded in November. New York City transit use, the Dow Jones, and business failures were indiscernible and modest when compared to the volatility of the period. Because of the short duration of the pandemic and human resiliency, it was characterized as a hit-and-run disease that only produced brief slowdowns.
John Barry:
- Courts did not meet in every community.
- It is important to understand how quickly the pandemic moved in military camps (two and a half weeks) and in the civilian community (five to six weeks). The peak was a matter of days.
- There were three waves in 1918. The first wave was widely missed because symptoms were mild, with little complications (four deaths). The second wave came six months after the first wave was noticed, and it was lethal. It rolled across the country and around the world in 8 days to 3 weeks.
- Communication was very important in terms of the first wave. Adequate attention was not paid regarding the complexity the first wave represented. A first wave is very complicated in terms of messaging.
- Communities were more resilient in 1918. In Philadelphia, a manufacturer of street cars was asked to make coffins when the city ran out of coffins. That wouldn’t happen today due to specialization.
- The economic impact would be much more severe today.
- Dependence on transportation is enormously different today.
Richard Hatchett:
- The velocity of the 1918 pandemic did not compare with those in 1957 and 1968. The 1918 pandemic moved much faster through communities.
- It is important to understand all pandemics are different. In 1918, cities that were comparatively successful in mitigating the first wave were pressured to relax interventions. When relaxed, cities experienced that it never really went away.
- The key is early implementation of all interventions; however, once implemented, the trigger has been pulled, and cities needed to do a good job of managing messages to keep the public on board.
- It is tricky for communities to organize human behavior and maintain public compliance.
Q: It has been a fascinating discussion about sectoral value. It is surprising that there would be so little economic impact when so many of the workers in the prime of their productive years died. How many people in their prime actually died?
A: (Hatchett) Excess mortality was 0.4%, which implies a case fatality rate of 1.75 to 2.0% and there was a cumulative morbidity of 18–20%. Although a lot of deaths were in the productive population, there were children and seniors who died as well. Estimates would be about 0.2% for the general population.
Q: In terms of not affecting government, how is that so?
A: (Barry) Richard [Hatchett], was talking systematically about structure. There were not a lot of government functions in 1918, and there was a more robust public health system. While many courts did close down, police departments did continue to function.
Q: In terms of transportation, communities were isolated from each other in 1918. Now we don’t have that. Have any prospective pieces of work been done?
A: (Hatchett) Cities were more isolated from each other. The pattern of the spread of the virus in 1918 was related to troop movements. There has been a considerable amount of work on modeling from Department of Transportation (DOT) looking at connectivity of major urban centers to estimate how infectious diseases would spread now. In recent work looking at real world data, the geographic diffusion process is driven more by work-related travel. The ability to make definitive statements will be limited by caveats.
A: (Barry): Remember, we don’t need airplanes to spread influenza. Work is being done on the impact of border crossings to delay entry by two weeks. I do not believe the potential benefits outweigh the clear negative impacts of shutting down the borders…the economic impacts would be significant. Look at the impact of SARS [Severe Acute Respiratory Syndrome]. It was pretty intense, but only for a relatively short period of time.
Q: What is your idea of community members and businesses purchasing antiviral doses? Is it the role of business?
A: (Hatchett) CDC and HHS have been engaging businesses on how to handle influenza. As opposed to privately stockpiling antivirals, creating liberal leave policies would be more useful. Some companies (privately held) have made the decision that it is in their best interest to keep employees with sick family members out of the work place.
A: (Barry) Keep in mind that the H5N1 virus can become resistant to Tamiflu, especially the more it is used as a prophylactic. The more this happens, the more rapidly a resistant strain will emerge. It may help some, but we don’t know how much. It is not a magic bullet that many people would hope it to be. I do not personally believe that H5N1 will be the next pandemic; having said that, yes, I do have a personal cache of Tamiflu.
Q: Is there a Plan B aside from Tamiflu?
A: (Barry) I have heard that Relenza may be better as a prophylactic. I know that there is a lot of work and research going into developing and producing antivirals.
Q: There is a tremendous emphasis on preparedness for pandemic influenza, but there seems to be the potential for a wide range of other public health emergencies. Do we risk public confidence if the next public pandemic is different?
A: (Hatchett) Certain aspects of pandemic preparedness are specific to an influenza threat, but a lot of what we’re doing has the potential for broad application. In terms of dealing with a respiratory virus, a lot of what we’re doing will serve us.
Exercise: Note cards were turned in from audience members to collect thoughts and concerns from the group. A summary of these audience-generated questions and issues can be found in Appendix D of this document.
