In 1918, an influenza virus jumped from birds to humans and killed an estimated 50 million to 100 million people in a world with less than a quarter of today’s population. Dozens of mammals also became infected.
Now we are seeing another onslaught of avian influenza. For years it has been devastating bird populations worldwide and more recently has begun infecting mammals, including cattle, a transmission never seen before. In another first, the virus almost certainly jumped recently from a cow to at least one human — fortunately, a mild case.
While much would still have to happen for this virus to ignite another human pandemic, these events provide another reason — as if one were needed — for governments and public health authorities to prepare for the next pandemic. As they do, they must be cautious about the lessons they might think Covid-19 left behind. We need to be prepared to fight the next war, not the last one.
Two assumptions based on our Covid experience would be especially dangerous and could cause tremendous damage, even if policymakers realized their mistake and adjusted quickly.
The first involves who is most likely to die from a pandemic virus. Covid primarily killed people 65 years and older, but Covid was an anomaly. The five previous pandemics we have reliable data about all killed much younger populations.
The 1889 pandemic most resembles Covid (and some scientists believe a coronavirus caused it). Young children escaped almost untouched and it killed mostly older people, but people ages 15 to 24 suffered the most excess mortality, or deaths above normal. Influenza caused the other pandemics, but unlike deaths from seasonal influenza, which usually kills older adults, in the 1957, 1968 and 2009 outbreaks, half or more deaths occurred in people younger than 65. The catastrophic 1918 pandemic was the complete reverse of Covid: Well over 90 percent of the excess mortality occurred in people younger than 65. Children under 10 were the most vulnerable, and those ages 25 to 29 followed.
Any presumption that older people would be the chief victims of the next pandemic — as they were in Covid — is wrong, and any policy so premised could leave healthy young adults and children exposed to a lethal virus.
The second dangerous assumption is that public health measures like school and business closings and masking had little impact. That is incorrect.
Australia, Germany and Switzerland are among the countries that demonstrated those interventions can succeed. Even the experience of the United States provides overwhelming, if indirect, evidence of the success of those public health measures.
The evidence comes from influenza, which transmits like Covid, with nearly one-third of cases transmitted by asymptomatic people. The winter before Covid, influenza killed an estimated 25,000 here; in that first pandemic winter, influenza deaths were under 800. The public health steps taken to slow Covid contributed significantly to this decline, and those same measures no doubt affected Covid as well.
So the question isn’t whether those measures work. They do. It’s whether their benefits outweigh their social and economic costs. This will be a continuing calculation.
Such measures can moderate transmission, but they cannot be sustained indefinitely. And even the most extreme interventions cannot eliminate a pathogen that escapes initial containment if, like influenza or the virus that causes Covid-19, it is both airborne and transmitted by people showing no symptoms. Yet such interventions can achieve two important goals.
The first is preventing hospitals from being overrun. Achieving this outcome could require a cycle of imposing, lifting and reimposing public health measures to slow the spread of the virus. But the public should accept that because the goal is understandable, narrow and well defined.
The second objective is to slow transmission to buy time for identifying, manufacturing and distributing therapeutics and vaccines and for clinicians to learn how to manage care with the resources at hand. Artificial intelligence will perhaps be able to extrapolate from mountains of data which restrictions deliver the most benefits — whether, for example, just closing bars would be enough to significantly dampen spread — and which impose the greatest cost. A.I. should also speed drug development. And wastewater monitoring can track the pathogen’s movements and may make it possible to limit the locations where interventions are needed.
Still, what’s achievable will depend on the pathogen’s severity and transmissibility, and, as we sadly learned in the United States, how well — or poorly — leaders communicate the goals and the reasons behind them.
Specifically, officials will confront whether to impose the two most contentious interventions, school closings and mask mandates. What should they do?
Children are generally superspreaders of respiratory disease and can have disproportionate impact. Indeed, vaccinating children against pneumococcal pneumonia can cut the disease by 87 percent in people 50 and older. And schools were central to spreading the pandemics of 1957, 1968 and 2009. So there was good reason to think closing schools during Covid would save many lives.
In fact, closing schools did reduce Covid’s spread, yet the consensus view is that any gain was not worth the societal disruption and damage to children’s social and educational development. But that tells us nothing about the future. What if the next pandemic is deadlier than 1957’s but as in 1957, 48 percent of excess deaths are among those younger than 15 and schools are central to spread? Would it make sense to close schools then?
Masks present a much simpler question. They work. We’ve known they work since 1917, when they helped protect soldiers from a measles epidemic. A century later, all the data on Covid have actually demonstrated significant benefits from masks.
But whether to mandate masks is a difficult call. Too many people wear poorly fitted masks or wear them incorrectly. So even without adding in the complexities of politics, compliance is a problem. Whether government mask mandates will be worth the resistance they foment will depend on the severity of the virus.
That does not mean that institutions and businesses can’t or shouldn’t require masks. Nor does it mean we can’t increase the use of masks with better messaging. People accept smoking bans because they understand long-term exposure to secondhand smoke can cause cancer. A few minutes of exposure to Covid can kill. Messaging that combines self-protection with communitarian values could dent resistance significantly.
Individuals should want to protect themselves, given the long-term threat to their health. An estimated 7 percent of Americans have been affected by long Covid of varying severity, and a re-infection can still set it off in those who have so far avoided it. The 1918 pandemic also caused neurological and cardiovascular problems lasting decades, and children exposed in utero suffered worse health and higher mortality than their siblings. We can expect the same from the next pandemic.
What should we learn from the past? Every pandemic we have good information about was unique. That makes information itself the most valuable commodity. We must gather it, analyze it, act upon it and communicate it.
Epidemiological information can answer the biggest question: whether to deploy society-wide public health interventions at all. But the epidemiology of the virus is hardly the only information that matters. Before Covid vaccines were available, the single drug that saved the most lives was dexamethasone. Health officials in Britain discovered its effectiveness because the country has a shared data system that enabled them to analyze the efficacy of treatments being tried around the country. We have no comparable system in the United States. We need one.
Perhaps most important, government officials and health care experts must communicate to the public effectively. The United States failed dismally at this. There was no organized effort to counter social media disinformation, and experts damaged their own credibility by reversing their advice several times. They could have avoided these self-inflicted wounds by setting public expectations properly. The public should have been told that scientists had never seen this virus before, that they were giving their best advice based on their knowledge at the time and that their advice could — and probably would — change as more information came in. Had they done this, they probably would have retained more of the public’s confidence.
Trust matters. A pre-Covid analysis of the pandemic readiness of countries around the world rated the United States first because of its resources. Yet America had the second-worst rate of infections of any high-income country.
A pandemic analysis of 177 countries published in 2022 found that resources did not correlate with infections. Trust in government and fellow citizens did. That’s the lesson we really need to remember for the next time.
John M. Barry, a scholar at the Tulane University School of Public Health and Tropical Medicine, is the author of “The Great Influenza: The Story of the Deadliest Pandemic in History.”
The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: letters@nytimes.com.
Follow the New York Times Opinion section on Facebook, Instagram, TikTok, WhatsApp, X and Threads.
