The History of Polio

News reports about this disease tend to mix up three different terms: that it has disappeared from a region, that it has been certified, or that it has been eradicated. These three words operate on very different scales, yet news headlines often blur them together into one.

Elimination is when a disease stops producing new cases within a specific geographic area. That status only holds as long as prevention efforts continue. Eradication is when that same milestone is achieved across the entire globe—a point where we can finally let down our guard and stop routine interventions. That is why no single country can achieve eradication on its own.

Certification belongs to yet another layer. It is the formal process in which an independent commission verifies that "this region has met the criteria." For a region to be certified, there must be zero transmission of wild poliovirus across every country in that region for 3 consecutive years, and disease surveillance throughout those 3 years must meet certification standards. Five of the World Health Organization's six regions have received this certification: the Americas in 1994, the Western Pacific in 2000, Europe in 2002, South-East Asia in 2014, and Africa in 2020.

Here, we need to make another distinction. What these five regions received was certification for the "wild type." The vaccine-derived viruses we will discuss later sit on a completely separate track, and they continued to emerge in several countries even after certification. Furthermore, wild poliovirus itself splits into three types. The Global Commission for the Certification of Poliomyelitis Eradication declared Type 2 eradicated on September 20, 2015, and Type 3 on October 24, 2019. Following that 2019 declaration, the only wild type that has not been declared eradicated is Type 1.

Therefore, the sentence "polio has been eradicated" cannot be written just yet. Even on August 25, 2026, the International Health Regulations Emergency Committee unanimously concluded that the international spread of the virus remains a Public Health Emergency of International Concern. On top of that, these terms are used differently across documents. The elimination and eradication we just separated mark how far — a defined area, or the whole planet. But the same two words are also used to mark how much. A 2010 review paper calls the status where indigenous wild poliovirus is gone elimination, and reserves eradication for when vaccine-derived viruses are gone as well. World Health Organization documents attach them in reverse, applying eradication to the wild type and elimination to vaccine-derived strains. Two different distinctions are sharing the same pair of words. So what you should take away from this story is not which word is right, but the habit of checking each time whether a statement refers only to the wild type or includes vaccine-derived strains as well.

This disease was not new. Michael Underwood recorded clinical observations of it in 1789, and Jakob Heine classified it as a distinct condition in 1840. Even ancient Egyptian artwork depicts a person leaning on a staff with a withered leg.

Yet around 1880, something strange happened. Outbreaks of "infantile paralysis" began to be reported one after another across Scandinavian countries and the United States. They appeared at nearly the same time in Europe and the United States, while no outbreaks were reported in other regions around that time. Why did this ancient disease suddenly become an epidemic then, and in those specific countries?

A 2010 review paper summarized what it calls a "hypothetical explanation." When sanitation was poor, infants encountered this virus between 6 and 12 months of age. At that stage, maternal antibodies were still present in their bodies, so even if an infection occurred in the gut, the virus could not travel through the bloodstream to the nerves. They gained immunity without paralysis and moved on. But as personal hygiene and water and sewage systems improved, the age of first exposure was pushed beyond 12 months. By then, maternal antibodies had waned.

The paper offered three lines of evidence: outbreaks emerged first in countries with advanced hygiene and public health, spreading to other countries over 60 years; once outbreaks took hold, the age distribution of patients gradually shifted upward over many years; and in countries that had not yet seen outbreaks, there was no gap between an infant losing maternal antibodies and developing its own. Between 1947 and 1953, a study in Casablanca compared two population groups living in the same city. The paralysis rate in one group was twenty times higher, and patients in that group were mostly older children and young adults. In the other group, patients were mostly infants.

Still, we must pause here. The authors themselves described this evidence as "circumstantial." What is more, the same paper makes an additional distinction. Data are mixed on whether the proportion of infections leading to paralysis also rose in older age brackets. Some records show no sign of that rate increasing between ages 3 and 14, and data from one major city showed no rise in incidence during that span either. Other data do indicate that case fatality rates increased with age. So the accurate statement is not that "poor hygiene spread the disease," but rather that "there is an explanation suggesting outbreaks became visible as hygiene improved."

When a major outbreak struck New York City in 1916, the only available measures were isolating the infected, shutting down public gatherings, and disinfecting. More than 2,000 people died. Although a virus had already been identified as the cause in 1908, there was no vaccine to prevent it and no medicine to cure it. In 1921, a 39-year-old American politician developed a fever accompanied by ascending paralysis and was diagnosed with polio. His name served as the public face of the disease for a long time, but in 2003, a paper evaluated the probabilities across eight symptoms and reported that six pointed toward a different condition. Which it actually was remains unsettled to this day.

What happens to people exposed to this virus falls into four paths. Showing no symptoms at all is the most common, accounting for about 72%. A mild illness that brings a fever for a few days accounts for about 24%, meningitis without paralysis makes up 4%, and progressing all the way to paralysis is usually under 1%.

Don't confuse two numbers here. The proportion of infected individuals who develop paralysis and the proportion of paralyzed individuals who die are completely different figures. The former is under 1%, while the latter is 5 to 10% according to the World Health Organization. That happens when the breathing muscles stop working.

The timeline of symptoms also overlaps. From infection to the onset of mild symptoms takes 3 to 6 days, and it takes 7 to 21 days for paralysis to begin. The virus is shed from the throat for 1 to 2 weeks, and in stool for several weeks. The US Centers for Disease Control and Prevention notes: "Even people who do not have symptoms can pass the virus to others and cause illness."

This single line is the heart of the story. For any disease to be targeted for eradication, three conditions must be met: the pathogen must not hide in hosts other than humans, there must be a way to identify infected individuals, and there must be a means to stop transmission. Poliovirus meets the first condition. Humans are its only host. It also meets the third condition, because there are two vaccines. Where things break down is the second condition.

This is where polio parts ways with smallpox. Anyone who caught smallpox broke out in an unmistakable rash, and no one went around spreading it without symptoms. So finding one patient revealed the chain of transmission, and ring vaccination—vaccinating only the circle of people around that patient—worked. Polio is not like that. Behind a single paralyzed individual, far more invisible infections can be lurking underneath, so wetting just the visible circle around one point can hardly catch up to where the virus has already spread. Tracing contacts one by one carries little power here, because the trail itself cannot be seen. That is why outbreak response leans toward running two or three rounds of vaccination campaigns across the entire community.

That completely changes how we confirm that "this region is free of the disease." Instead of counting patients, health workers search for and test every single case of sudden-onset, floppy paralysis in children under 15 years old. Surveillance is considered thorough only if at least 2 cases per 100,000 people are found, even for paralysis arising from non-polio causes. In at least 80% of those cases, two stool specimens must be collected more than 24 hours apart, kept on ice, and delivered intact to an accredited laboratory within 14 days of paralysis onset. On top of that, environmental surveillance samples sewage to search for the virus. The global laboratory network comprises more than 150 facilities, investigating over 100,000 cases of acute flaccid paralysis and collecting more than 8,000 sewage samples every year.

Why this surveillance is necessary became clear in Israel in 2013. The wild-type virus entered an area with high injectable-vaccine coverage and circulated for several months without producing a single case of paralysis. That reality was uncovered only through sewage surveillance.

There was no treatment to restore paralyzed nerves back then, and there is none today. When the path to a cure is blocked, efforts split in two directions: helping patients hold on while paralyzed, or preventing them from catching the disease in the first place. During the first half of the 20th century, energy was poured into the former first.

In 1928, doctors at Boston Children's Hospital approached engineer Philip Drinker. There was simply nothing they could do for children whose breathing muscles had become paralyzed. Drinker, together with his colleague Louis Shaw, built an airtight chamber that enclosed the patient's entire body. By raising and lowering the air pressure inside the tank, the chest expanded and contracted on its own. The first human-sized device was powered by two vacuum cleaners.

Though it was called the "iron lung," it was not a device that replaced the lungs. It was a machine that took over the work of the breathing muscles. Nor did it cure the disease. A 1986 paper cited this machine as a prime example of "halfway technology." It was an article examining the true cost of technologies that merely prop up consequences without eliminating the underlying cause.

Its limits became clear in Copenhagen in 1952. Between July and December of that year, more than 2,700 people were hospitalized; 866 suffered paralysis, and among them, 316 experienced respiratory or bulbar paralysis. The hospital had only one tank ventilator and six cuirass respirators that strapped onto the chest. The shortage of machines was an issue, but an even bigger problem was the medical team's judgment that this method did not clear carbon dioxide effectively.

On August 26, anesthesiologist Bjørn Ibsen performed a tracheotomy on a 12-year-old patient and switched to pumping air directly into the lungs. With no machines to do the job, people stepped in. Every day, 250 medical students worked in shifts, squeezing ventilation bags by hand. By the time the epidemic ended, papers record that roughly 1,500 people had logged a total of 165,000 hours of manual bagging. The mortality rate for patients with respiratory paralysis plunged from 87% to around 40% within days, and dropped to 11% within a year.

On August 1 of the following year, Ibsen opened a ward at Copenhagen Municipal Hospital where specialists from multiple fields cared for patients together. Modern intensive care is widely considered to have begun right here. It was not created to cure polio, but was the result of redesigning how to hold on in the face of an incurable disease.

The path toward prevention opened in 1949. That year, Enders, Weller, and Robbins succeeded in culturing the virus in non-neural human tissue. Until then, simply obtaining enough virus to produce a vaccine had been an immense challenge in itself.

On April 12, 1955, the results of a massive field trial for the inactivated injectable vaccine were announced, and it was licensed on the very same day. And then, an accident occurred that year. Batches made by one manufacturer were distributed with virus that had not been fully inactivated. Paralysis struck 79 vaccinated individuals, 105 of their family members, and 20 community contacts. An investigation confirmed live virus in 7 out of 17 batches. On May 7, the U.S. Surgeon General recommended suspending vaccinations.

The official record of what went wrong is understated: "The formaldehyde inactivation and safety testing procedures used were not adequate." Following the incident, regulatory authority over biological products was transferred to a newly established division within the National Institutes of Health, and later moved to the Food and Drug Administration in 1972. What this crisis left behind was not a verdict on the vaccine itself, but an answer to where the systems of testing and licensing ought to reside.

A few years later, a different vaccine emerged. It was an oral vaccine, administered by dropping weakened live virus into the mouth. Massive trials took place in the Soviet Union involving 20,000 people in 1958 and 10 million in 1959, leading to the licensing of a monovalent version in 1961 and a trivalent version covering all three serotypes in 1963.

Here, a common misconception must be cleared up. These two are neither a chronological sequence nor a contest of winners and losers. They simply belong in different places. First, where they build immunity within the body differs. The injection produces antibodies in the bloodstream, giving it powerful strength to prevent paralysis. However, if an individual who has never encountered live virus receives only the shot, almost no immunity forms in the gut. If given the oral vaccine later, they become infected and shed the virus in their stool much like an unvaccinated child. Because the oral vaccine multiplies in the intestines, it creates mucosal immunity there, thereby reducing the amount shed upon re-encounter. The silent transmission in Israel described in the previous chapter was precisely the setting where this difference revealed itself.

Second, their impact on others is different. A person who takes the oral vaccine sheds vaccine virus from the throat for a few days and in stool for 4 to 6 weeks, which can spread to household members and even confer immunity on those who have not been vaccinated. Third, the method of administration and storage are different. The oral vaccine is two drops in the mouth, while the shot must be injected into muscle. To keep the oral vaccine for a long time, it must be frozen and is very sensitive to heat. The injectable vaccine can be kept between 2 and 8 degrees for 3 years, but it must never be frozen.

Fourth, the accompanying risks differ. The oral vaccine carries the risk of vaccine-associated paralysis. As the attenuated virus multiplies in the gut, it can regain neurovirulence, causing paralysis in the vaccinated individual or their contacts. In countries using the oral vaccine, this is estimated at 3.8 per 1 million births. For this reason, in 2000, after wild poliovirus ceased to circulate, the United States switched to an all-inactivated injection schedule, and no cases of vaccine-associated paralysis have been reported since. There were two reasons for the change: it was deemed unacceptable to continue bearing that burden in the absence of wild poliovirus, yet vaccination had to continue because the risk of importation from abroad remained. The vaccine listed on South Korea's immunization schedule is also the injectable form.

In 1988, the World Health Assembly resolved to eradicate polio globally. That year, over 350,000 cases of paralysis were being reported annually across more than 125 countries. The target deadline was the year 2000.

That deadline came and went. Still, the numbers dropped. Regional certifications followed one by one, each certifying that "wild poliovirus transmission had not been detected in the region for three consecutive years, while surveillance met required standards during that time." The Western Pacific Region, which includes South Korea, received its certification in 2000. That came more than a decade after the country's last patient was reported in 1983.

The wild strains were tackled serotype by serotype, too. The last paralysis case caused by wild type 2 occurred in India in October 1999, and certification of its eradication was declared on September 20, 2015. The last case of type 3 was recorded in Nigeria in November 2012, with the declaration following on October 24, 2019. The gap of more than a decade between the last patient and the declaration is what stands out in this timeline. That time was not spent merely waiting to see if the disease lingered; it was spent proving that the surveillance described in the previous chapter was truly airtight.

The scale of this effort kept expanding. Today, over 1 billion doses of vaccine are administered each year to 370 million children across more than 40 countries. Vaccinators and volunteers number 20 million. Since the initiative began, an estimated 20 million people are walking today who would otherwise have been paralyzed.

Then in 2016, the entire world synchronized a switch of oral vaccines on a coordinated date. They switched from a formula containing all three serotypes to one containing only types 1 and 3. There were two reasons: wild type 2 had not been seen for a long time, and type 2 was responsible for a heavy burden of vaccine-associated paralysis and vaccine-derived outbreaks. The expectation was that removing type 2 from the vaccine would make that burden disappear.

Yet after the switch, type 2 vaccine-derived outbreaks actually increased. The World Health Organization's vaccine position paper points to several reasons: inadequate responses in places where the virus was newly detected, delayed campaigns, and insufficient coverage with the type 2-only response oral vaccine. While wild strains were being wiped out one by one, a different front was quietly growing on the side.

The virus in the oral vaccine is a weakened live virus. In a person with a healthy immune system, it clears within 6 to 8 weeks. The trouble is that during that period, the person sheds the virus in their stool, and in places with poor sanitation, it can spread to others.

Where vaccination rates are high, even if it spreads, it meets people who are already immune and soon stops. Places with low vaccination coverage are a different story. If enough children remain unvaccinated, this virus keeps passing from person to person. As it circulates for usually twelve to eighteen months, its weakened traits unravel, and it regains the power to damage nerves. That is how vaccine-derived poliovirus comes to be. So this is less something that arises from the vaccine itself, and more something born of the gaps where vaccination is patchy.

This front was confirmed between 2000 and 2001 on the island of Hispaniola. When scientists sequenced its genes, they discovered that a virus originating from the vaccine had been circulating among people. There were three reasons it was detected so late: the technology to read genetic sequences rapidly had only become available around that time, a global laboratory network had been built over the preceding decade or so, and above all, this phenomenon only happens in places where vaccination rates are very low.

The numbers have to be tracked on two separate lines. In 2020, there were 1,113 reported cases of vaccine-derived paralysis, compared to 140 cases of wild type 1 paralysis that same year. The following year, the figures were 691 and 6. If you add the two together or blur them into one, how each side is moving gets erased. In these years, the vaccine-derived tally was larger than the wild-type count.

It also turns up in places that had seen no cases for a long time. In July 2022, an unvaccinated adult in Rockland County, New York, was paralyzed by vaccine-derived type 2 poliovirus. It was the first confirmed paralytic case in the United States since 2013. Genetically linked virus was subsequently detected in wastewater across neighboring counties. Out of 1,076 samples collected from Rockland and 12 surrounding counties, 89 tested positive for type 2.

Efforts also turned toward fixing the tool itself. A new type 2 oral vaccine, genetically tweaked at several sites so its weakened state would not easily unravel, was placed on the World Health Organization Emergency Use Listing in November 2020. It was cleared only for outbreak response and not approved for routine immunization. On February 13, 2026, a version produced by a second manufacturer passed WHO prequalification, expanding available supplies.

Yet the number does not drop to 0. From March 2021 to September 2023, while about 700 million doses were administered, seven new outbreaks were confirmed across six countries, turning up 61 cases of paralysis and 39 wastewater samples. The U.S. Centers for Disease Control and Prevention notes this rate as a provisional estimate of 1 case per 100 million doses, alongside the rate for the previous type 2 oral vaccine, which was 1 case per 10 million doses. It is a figure that has shrunk, but not disappeared.

This is how things stand based on the World Health Organization Emergency Committee statement on August 25, 2026. In 2025, 52 cases of wild poliovirus type 1 paralysis were reported across the entire year: 21 cases in Afghanistan and 31 cases in Pakistan. In 2026, 4 cases were reported through April 30.

The same statement has a separate line for vaccine-derived cases. In 2025, there were 238 cases across 30 countries, and in 2026, 32 cases across 12 countries through April 30. Laying these two lines side by side makes it clear why this program is running two endgames at the same time.

The curve does not move in only one direction, either. In 2024, 99 cases of wild type 1 were reported: 74 cases in Pakistan and 25 cases in Afghanistan. A few years earlier, both countries had dropped to their lowest levels in history, and transmission was even interrupted for short periods. The committee pointed to a few border corridors connecting the two countries and one major city as the places where wild virus transmission persists.

Signals are also surfacing from places that had not seen a case in a very long time. In October 2025, wild type 1 was detected in sewage in Germany. Around the same time, vaccine-derived viruses were found in wastewater across several European countries. In a sense, the sewage speaks up before paralysis ever appears in people. That is why the reminder that being "certified" does not mean you can let your guard down appears again and again throughout this program's documents.

Since 2014, the international spread of this virus has been designated a Public Health Emergency of International Concern, and its status is reviewed every three months. The August 2026 review concluded once again that the emergency must stay in place. The global goals laid out by the committee consist of two lines: interrupting endemic transmission of wild type 1 within 2026, stopping type 2 vaccine-derived outbreaks by 2028, and then proceeding toward certification for each. They are targets, not a guaranteed schedule.

Let us return to the vaccination schedule we started with. In South Korea, the last patient was reported in 1983, and the Western Pacific Region was certified in 2000. Yet this is why that box still remains on the vaccination chart. As long as seeds still linger over the fence, even the household with a clean yard cannot put down its hoe.

There is no need to try and guess when it will end. Instead, in light of the rules this story has shown us, let's summarize where to look so you can be the first to notice when the board moves.

First, make a habit of reading numbers in two lines. News that simply reports "X cases of polio" erases whether it is wild or vaccine-derived. The two tracks differ in both cause and response, and in some years, vaccine-derived cases were far higher than wild ones. A number collapsed into a single line does not tell you what actually moved.

Second, watch the weight of the words. News that a certain region has been certified and news that the entire globe has eradicated it are completely different statements. What's more, all regional certifications to date have been for the wild strain. And as we saw earlier, that same pair of words is tied to different axes depending on the document. So if the word "eradication" appears one day, instead of just taking the word at face value, check first whether the statement covers both tracks.

Third, watch the work of fixing the tools. The new type 2 oral vaccine was re-engineered so its weakened state does not easily unravel, and by preliminary estimates, new outbreaks are about tenfold lower than with the old one. Whether that figure moves from preliminary to confirmed, and where it settles if it stops short of 0, are the key watching points on this track.

Fourth, consider the sheer scale of confirming that "none" exists. Watch where acute flaccid paralysis surveillance and wastewater surveillance become denser and where they thin out. Rather than news that no paralysis cases have been reported, how many cases of paralysis a region actively finds and tests each year tells us far more. When surveillance is thin, it does not mean "there are no patients"—it means "we failed to find them."

Lastly, just one more thing. This story also features those who ended up on the losing side. The people who decided in 2016 to drop type 2 from the oral vaccine had their reasons at the time. The two risks that serotype created were genuinely heavy. Even in 1955, where the testing procedures were judged insufficient, those procedures were the accepted standard up to that point. It is easy to say those people were wrong after we already know the outcome. Nobody yet knows which side we will be standing on when the next verdict comes in.