We Need a Covid-19 Vaccine—Let’s Get It Right the First Time

The flu shot campaigns of 1976 and 2009 offer key lessons for how (and how not) to distribute, monitor, and communicate about vaccines. But will anyone listen?
a scientists working with H1N1 vaccine
A laboratory technician works on the Influenza A (H1N1) virus vaccine production line.Photograph: Oliver Bunic/Getty Images

On a rainy night in January 1976, a batch of new Army recruits training at Fort Dix, New Jersey, were sent on a 5-mile march. The next day, one of the recruits, Private David Lewis, collapsed with severe pneumonia. Lewis died, and a whole swath of his platoon fell ill with chest congestion and fever: almost 200 men, 13 of whom had to be hospitalized. January is within flu season, and military physicians assumed that the flu had somehow made its way onto the base—a problem for the group and a tragedy for the dead soldier, but not unexpected.

Tests upended that thinking. The soldiers did have the flu, but among some of them, at least, the virus that was causing their illness was not the common strain that was circling the globe that year. It was instead an unfamiliar virus to which almost no one had immunity. It was among the strains of flu designated H1N1, and it was genetically related to a flu epidemic that some people in medicine at the time were old enough to still remember: the world-spanning, millions-killing pandemic of 1918.

person lathering hands with soap and water
Plus: What it means to “flatten the curve,” and everything else you need to know about the coronavirus.

The discovery of what came to be known as the 1976 swine flu electrified the country. Before the end of that March, President Gerald Ford declared the US would vaccinate “every man, woman, and child” in the United States against it. Congress appropriated emergency funds. Manufacturers rushed to make a new vaccine formula. By Thanksgiving, almost 45 million Americans, a quarter of the population at the time, received the new shot. Ford led the way: He was photographed receiving it in the Oval Office on October 14.

But unlike 1918, this time there was no pandemic. The cases among the soldiers were a spark that did not catch. And by the time that became clear, more than 500 people out of that 45 million had come down with an extremely rare condition, a paralysis called Guillain-Barré syndrome. Thirty-two of them died.

President Ford receives a swine flu inoculation from his White House physician, Dr. William Lukash. Courtesy of Gerald R. Ford Library

The events of 1976 had a profound effect on the US public health system. Congress held hearings for months. The director of the CDC (then called the Center for Disease Control) was fired. The rush to counter the apparent threat came to be seen as a mistake, and the possibility of a pandemic came to seem so unlikely that it took another 27 years before the federal government drafted a plan to respond to one.

“That campaign cost the government a great deal of credibility,” says Howard Markel, a physician and historian of epidemics who is director of the University of Michigan's Center for the History of Medicine. “It created for many years a Chicken Little response by government officials: They were afraid to act too quickly. Yet the thing about epidemics is, when they begin, you have to act quickly, without a lot of data.”

The pandemic that didn’t happen in 1976 did arrive 33 years later, when a different flu strain—another H1N1, but not the virus of 1976 or 1918—swept the world. It arose after the flu season should have ended, starting in April 2009 with a cluster of cases in Mexico, California, and Texas. By June, the World Health Organization declared the new strain was causing a pandemic. Eventually, more than 60 million people were infected just in the United States, and an estimated 203,000 people died worldwide.

Yet that response was troubled, too. A new vaccine was ginned up to respond, and though there were no obvious adverse reactions, there were significant stumbles in organizing manufacturing, and in getting the new formula out where it was needed most.


The flaws in these past campaigns matter, because they are among the largest, fastest emergency vaccination efforts to occur in the United States in the lifetimes of people making policy and practicing science today. Covid-19 isn’t influenza, but it is a pandemic, and it too is triggering a rapid search for a vaccine that could end in giving millions of shots to those who are vulnerable. So the lessons learned in 1976 and 2009 are important—especially since some of their mistakes are being made again.

Health experts see one in particular being repeated right now: Letting politicians, instead of scientists, be the spokespeople for what the country needs to do. “Politicians always want to overpromise, and then you’re at risk of underdelivering when the time comes,” says William Schaffner, a physician and professor of infectious diseases at Vanderbilt University School of Medicine, who in 1976 had just joined the Vanderbilt faculty after serving as a CDC disease detective. “You should always aim to do it the other way—underpromise and overdeliver—because then you’re a hero.”

Analyses written after the 1976 vaccination campaign identified a key problem: The entire project, from making a vaccine to delivering it, was created via a single decision in March 1976, immediately after the Fort Dix cases were identified. Harvey Fineberg, a physician and now president of the Gordon and Betty Moore Foundation, who cowrote an after-action report on the vaccination campaign for the incoming administration of President Jimmy Carter, told the World Health Organization in an interview in the midst of the 2009 pandemic that the “single ‘go or no-go’ decision” doomed the 1976 project.

Instead, he said, the administration could have commissioned the manufacturers to make the vaccine, but held off on administering it until it was clear that the number of cases was increasing and the new shot was needed. “Separate what needs to be done to prepare for future decisions from reaching conclusions and announcing them, before relevant information is at hand,” he said.

Today, some researchers are concerned the need to do something about coronavirus could create a similar rush. “It's likely there will be enormous pressure, both political and social, to get a vaccine into production quickly once it looks good in trials,” warns Stuart Blume, an emeritus professor of science at the University of Amsterdam and author of Immunization: How Vaccines Became Controversial.

Annual flu shots don’t need to go through clinical trials every time they are adjusted for each year’s flu strain, because the backbone of the vaccine was already clinically tested and licensed and doesn’t change. That might create an expectation that a Covid-19 vaccine can also arrive rapidly—an expectation that cannot be met, because there is no licensed coronavirus vaccine. “One thing that concerns me is whether the pressure to get it out quickly won't reduce the commitment of testing it in a variety of population groups that could react differently,” Blume adds, “whether it's age groups or ethnic groups or pregnant women.”

That caution comes in part from reactions to some of the formulas in which the quickly created 2009 pandemic flu vaccine was delivered. A small number of children in Sweden and Finland developed narcolepsy, a neurological disorder in which people fall asleep unpredictably during their waking hours, after receiving a version of the 2009 vaccine containing an adjuvant, an additional component that increased the immune system’s reaction. (That version was not used in the 2009 US campaign and is not sold in the US now.) Several months later, the government of Australia reported that up to 1 child out of 100 had fevers and convulsions after receiving a different vaccine formula that contained the 2009 strain, a reaction that other national campaigns could not identify in their data. \

Those reactions were statistically very rare, and to some extent remain mysterious. In the United States, for instance, Guillain-Barré occurs several thousand times every year, in people who have received immunizations as well as people who have not (but have experienced some other recent immune system-challenging event, such as an infection). That the cases occurred is not in dispute, but their association with vaccination was made after the cases occurred—so there is still some debate over whether they were caused by the vaccines or coincidental given how widely the vaccine was distributed. Nevertheless, the 1976 experience in particular taught health planners to set up sensitive surveillance systems before beginning the 2009 campaign, and that may have made a difference—because, though there was a rise in Guillain-Barré cases during the 2009 swine flu, it was one-tenth what it was in 1976.

Vaccines save lives and reduce illnesses and, if the Covid-19 pandemic proceeds as it has the past few months, any new vaccine against it could be in high demand. Which might put health authorities into a position that planners faced in 2009: trying to figure out who moves to the front of the line.

Flu vaccine manufacturing is scattered around the world; not every country contains a plant. Nations fortunate enough to house one have committed to the WHO to share their products equitably worldwide. But in 2009, Australia and Canada effectively nationalized their production, keeping back doses meant for other countries until they were sure their own citizens were taken care of.

“I am wondering what kind of coordination there is going to be if we do have a vaccine that becomes available for coronavirus,” says Samantha Vanderslott, a health policy expert and social science researcher at the Oxford University Vaccine Group. “Looking back at H1N1, it was very much individual countries working with suppliers to make sure they had the stocks they needed.”

Patchy supply isn’t only going to be a problem internationally. In the United States, administering public health is a state power. That means, among other things, that each state government gets to decide how to put shots into arms, even when a national vaccination program has been declared.

In 2009, that led to 50 different methods of distribution: at doctors’ offices, via pharmacists, in special mass-vaccination clinics, through the pre-existing networks that states maintain for giving childhood vaccines, and more. At the same time, to handle inevitable scarcity, states set slightly different standards for who ought to receive that vaccine first—leading to situations in which people who lived near a state line, and weren’t their own state’s priority group, could see people just like them getting the shot in the state next door.

“What’s really, really hard is those first weeks or months when there is enormous demand and there just isn’t enough to go around,” says Kelly Moore, a physician and epidemiologist who ran Tennessee’s 2009 immunization campaign within its health department and now chairs the WHO’s Immunization Practices Advisory Committee. “How do you get it to the people who are occupationally or medically at-risk? And how do you make prioritization decisions and adhere to them without letting them get in the way of administering the vaccine with appropriate speed?” Moore knows this from experience: The first vaccine shipment she received in 2009 contained just 60,000 doses to cover a state population of more than 6 million.

The channels by which a vaccine gets distributed in the US aren’t likely to be reorganized wholesale before a coronavirus vaccine arrives—so it’s critical that state governments begin laying plans now for how to share the vaccine fairly, equip the providers who will administer it, and develop the information systems to track who receives it. There isn’t a lot of institutional memory for this; Moore estimates that, out of the 64 immunization program managers now working in states, cities and territories, only a handful held those jobs in 2009.

“These programs can be hard to communicate to the public,” she says. “We’re fortunate that we have 18 months to work on this, and not four months like we had in 2009.”

Read all of our coronavirus coverage here.

The need for clear communication is the biggest lesson of past pandemic responses. During the 2009 campaign, the CDC asked Dr. David Sencer, who directed the agency during the 1976 episode and lost his job over it, to head up a real-time assessment of how the pandemic response was going. Sencer created an instant think tank called “Team B” and populated it with epidemiologists, virologists, and logistics experts. Markel was a member. He recalls that every day, the team spent hours on a conference call, reviewing whatever data or reactions had surfaced and recommending course corrections. Sencer sent their daily assessment straight to the leadership of the CDC.

In an analysis written a few months before he died in 2011, Sencer explained how much of a difference that made. In the entire 1976 campaign, he wrote, the government held only a single press conference, conducted by the White House and used to launch the vaccination program. When the vaccine was withdrawn, the White House put out only a press release.

In 2009, in contrast, the CDC was allowed to be the campaign’s public face. Acting director Dr. Richard Besser, who was succeeded by director Dr. Thomas Freiden in May 2011, gave near-daily public briefings: 61 over the course of the pandemic, according to an internal agency accounting. “The message was reinforced at all levels of government that this would be a science-led and science-driven response,” Sencer wrote.

In the current crisis, the CDC has not held a press conference since March 9.

Looking ahead to next year—the earliest, conservatively speaking, that a vaccine could be researched and tested sufficiently to be efficacious and safe—there’s a long list of tasks to accomplish. These will include not just developing the formula itself but negotiating manufacturing agreements, establishing monitoring for rare events, setting up distribution networks, and making sure the public understands what will happen and when. For each of those tasks, the experiences of 1976 and 2009 offer lessons, from knowing when it’s crucial to move quickly to recognizing when it would be smarter to pause.

The unanswered question is whether those lessons will be listened to. “The last, awful act of any pandemic,” Markel says, “is amnesia.”

More From WIRED on Covid-19