Spanish flu

• 1. What Is It

  ›

  In July 1918, a newspaper in Freetown, Sierra Leone, the Sierra Leone Weekly News, decided that the Spanish flu needed a stranger name than influenza. Doctors were so puzzled, the paper suggested, that they should borrow a phrase from Exodus 16 and call the disease Man hu, which means 'What is it?'. Nobody then knew that a virus caused influenza. The pandemic that followed ran from 1918 to 1920. It was caused by the H1N1 subtype of influenza A, and it infected an estimated 500 million people, nearly a third of the world. How many of them died is still argued over. This documentary asks why a disease that began elsewhere was blamed on Spain, and where it truly started. It asks why it cut down young adults so often. And it asks what doctors, governments and ordinary families did as the sickness crossed battlefields, oceans and remote islands.

  1448579

• 2. A Misleading Name

  ›

  On the 2nd of June 1918, The Times of London ran a dispatch titled 'The Spanish Epidemic'. A correspondent in Madrid reported over 100,000 victims of an unknown disease, and the dispatch did not yet use the phrase Spanish influenza. Three weeks later the paper observed that everyone now thought of the illness as the 'Spanish' influenza. Three days after that, an advertisement appeared for Formamint tablets, sold to prevent it. When the disease reached Moscow, Pravda announced that 'Ispánka (the Spanish lady) is in town'. Reports from Spain began on the 21st of May, and early reports from both Spain and Kansas called the illness 'three-day fever'.

  Spain was neutral in 1918, and it had no wartime propaganda machine to protect morale. Its newspapers therefore described the epidemic openly, while censors in the fighting nations buried such news. The result was a false impression that the outbreak began in Spain. The silence elsewhere was so effective that Spanish health officials did not realize neighboring countries were suffering too. A Spanish official wrote to the Journal of the American Medical Association in October 1918. He objected that 'this epidemic was not born in Spain, and this should be recorded as a historic vindication.'

  Before 1918, blame for influenza already traveled across borders. During the 1889-1890 pandemic, called the Russian flu, Russians spoke of 'Chinese catarrh', Germans of the 'Russian pest', and Italians of the 'German disease'. In 1918 the habit returned with new variations. The French press first said 'American flu', then switched to 'Spanish flu' rather than antagonize an ally. British soldiers said 'Flanders flu', and German soldiers said 'Flandern-Fieber', both after a battlefield in Belgium. Senegal called it 'Brazilian flu', and Brazil called it 'German flu'. Spain itself used 'French flu' and 'Naples Soldier', the latter taken from a popular song in a zarzuela. Poland spoke of the 'Bolshevik disease', and Russia of the 'Kirghiz disease'. Japan blamed sumo wrestlers returning from Taiwan, and called it 'sumo flu'.

  Other names came from the people living through it. One local name, 'kaapitohanga', meant a disease which passes through like a bullet. Another, 'chipindupindu', meant a disease from seeking to make a profit in wartime. 'Malibuzwe' meant let enquiries be made concerning it. French military doctors preferred 'disease 11', or maladie onze. German doctors downplayed the danger with 'pseudo influenza', while doctors in Africa urged patients to take it seriously with 'influenza vera'. The label 'great influenza epidemic' echoed 'the great war', the common name for World War I before World War II. Children in 1918 even skipped rope to a rhyme adapted from the 1889-90 pandemic, in which 'Influenza' was clipped to 'Enza'.

  In 2015, the World Health Organization published best practices that discourage tying new diseases to culturally significant names. It listed 'Spanish flu' under examples to be avoided. Many authors now write of the 1918-19/20 influenza pandemic instead, a label that says nothing about where it began.

  3334354257

• 3. Searching for Origins

  ›

  Kansas, France, Austria and China each have a champion in the argument over where the Spanish flu began, and the surviving records cannot settle it. Historian Alfred W. Crosby stated in 2003 that the flu began in Kansas. Author John M. Barry pointed in 2004 to a January 1918 outbreak in Haskell County. That outbreak prompted a local doctor, Loring Miner, to warn the editors of Public Health Reports, the journal of the U.S. Public Health Service. A 2018 study led by Michael Worobey examined tissue slides and medical reports. It found the Kansas cases milder, with fewer deaths than the infections in New York City in the same period. Its genetic analysis favored a North American origin without proving one. The structure of the virus's haemagglutinin proteins suggests the virus arose long before 1918, and other work places the key H1N1 reassortment in or around 1915.

  At Étaples in France, a major British troop staging area and hospital camp, virologist John Oxford sees the likely center of the pandemic. His study found that in late 1916 the camp was struck by a new illness with high mortality and flu-like symptoms. A similar outbreak followed in March 1917 at army barracks in Aldershot, England, and military pathologists later recognized both as the same disease. The crowded camp suited a respiratory virus. Its hospital treated thousands of victims of poison gas and other war casualties, and the camp kept a piggery and regularly received poultry for food. Oxford and his team proposed that a bird virus mutated and then moved to pigs kept near the front.

  A 2016 report in the Journal of the Chinese Medical Association found that the virus had circulated among European armies for months and possibly years. Political scientist Andrew Price-Smith published data from Austrian archives suggesting the flu began in Austria in early 1917. Yet a 2009 study in Influenza and Other Respiratory Viruses found that mortality peaked in the same two months, October and November 1918, in all fourteen European countries it analyzed. A virus spreading outward from a European source would not be expected to behave that way.

  In 1993, Claude Hannoun, the leading expert on the Spanish flu at the Pasteur Institute, argued that the precursor virus came from China. He suggested it mutated near Boston, then reached Brest in France, Europe's battlefields and the rest of the world, with Allied forces as the main carriers. In 2014, historian Mark Humphries of the Memorial University of Newfoundland pointed to the mobilization of 96,000 Chinese laborers who worked behind the British and French lines. He found archival evidence of a respiratory illness in northern China, where the laborers came from, in November 1917. A year later Chinese health officials identified it as identical to the Spanish flu. Some reports described respiratory illness along the laborers' route, which also crossed North America.

  China did appear to be spared in 1918, with several studies documenting a comparatively mild flu season, though that is disputed because of the lack of data in the Warlord Period. Some speculated that earlier exposure had given the Chinese population immunity. In Guangdong Province, early outbreaks in 1918 struck the young. The June outbreak infected children and adolescents aged 11 to 20, and the October outbreak was most common at ages 11 to 15. The 2016 report found no evidence that the virus reached Europe with Chinese and Southeast Asian workers. Mortality among those workers was an estimated one in a thousand, no different from other Allied units in France. Many entered Europe by other routes that produced no detectable spread. No tissue samples from northern China survive for modern comparison, so the dispute rests on archives and statistics.

  7574787984

• 4. Four Waves

  ›

  The conventional start date is the 4th of March 1918, when an army cook named Albert Gitchell reported sick at Camp Funston in Kansas, although earlier cases existed. Within days, 522 men at the camp had fallen ill. By the 11th of March the virus had reached Queens, New York. Camp Funston was a major training ground for the American Expeditionary Forces, so soldiers carried the illness to other army camps and then to Europe. Critics later faulted the failure to take preventive measures in March and April. By April it was an epidemic in the Midwest, on the East Coast and in French ports, and it reached the Western Front by mid-April. France, Great Britain, Italy and Spain followed, and in May it reached Wrocław and Odessa. Germany released Russian prisoners of war after the Treaty of Brest-Litovsk in March, and they carried the disease home. North Africa, India and Japan saw it in May, and China reported an outbreak in June.

  The first wave looked ordinary in the numbers. About 75,000 flu-related deaths were reported in the United States in the first six months of 1918, against about 63,000 in the same months of 1915. In Madrid, fewer than 1,000 people died of influenza in May and June. No quarantines were reported. Yet the war felt the blow. Three-quarters of French troops, half of the British forces and over 900,000 German soldiers fell sick.

  In the second half of August 1918, ships from Brest probably carried a far deadlier second wave to Boston, Massachusetts, and to Freetown, Sierra Leone. It had likely reached Brest with American troops or French recruits for naval training. From the Boston Navy Yard and Camp Devens, about 30 miles west of Boston, it spread to other U.S. military sites and to troops bound for Europe. Over the next two months it covered North America, then Central and South America, and reached Brazil and the Caribbean by ship. From Freetown it moved along the West African coast, rivers and railways. South Africa received it in September on ships returning members of the South African Native Labour Corps from France, and it reached Ethiopia in November. The Ottoman Empire had already recorded its first cases, among soldiers, in July.

  On the 15th of September, New York City recorded its first influenza death. Thirteen days later, on the 28th of September 1918, Philadelphia held the Liberty Loans Parade to promote government war bonds. The parade led to an outbreak that caused 12,000 deaths.

  In Russia, the second wave advanced in a southwest-to-northeast diagonal, and the North Russia intervention carried it to Arkhangelsk. Through the Russian Civil War and along the Trans-Siberian railway it spread across Asia. It reached Iran through Mashhad, India in September, and China and Japan in October. The Armistice celebrations of the 11th of November 1918 set off outbreaks in Lima and Nairobi, and by December the wave was largely spent.

  October 1918 was the deadliest month of the entire pandemic. In the United States about 292,000 deaths were reported from September through December, compared with about 26,000 in the same months of 1915. The Netherlands counted over 40,000 deaths from influenza and acute respiratory disease. Bombay reported about 15,000 deaths among 1.1 million residents.

  Michigan saw a swift resurgence that peaked in December 1918, possibly because it had lifted its ban on public gatherings. In early December the Public Health Service reported a 'recrudescence of the disease' in 'widely scattered localities'. Many places reimposed bans and school closures. Los Angeles, New York City, Memphis, Nashville, San Francisco and St. Louis had significant outbreaks in January and February 1919. Nearly 160,000 deaths were attributed to pneumonia and influenza in the first six months of 1919. In Europe, England and Wales saw a third wave that peaked in early March and lasted until May. France peaked in February, and Portugal's resurgence ran from March to September 1919.

  Australia, shielded through 1918 by a strict maritime quarantine, saw influenza for the first time in January 1919. It became an epidemic first in Melbourne, peaking in mid-February. Land quarantines slowed its spread. Queensland was not infected until late April, Western Australia until early June, and Tasmania until mid-August. Madagascar recorded its first cases in April 1919, and by June the outbreak had reached nearly the whole island. Chile, first hit in October 1918, suffered a severe second wave from July 1919 to February 1920, with mortality peaking in August 1919.

  In September 1919, U.S. Surgeon General Rupert Blue said a return of the flu was 'probably, but by no means certainly' coming. Japan's outbreak began in December 1919 and peaked in January. Between October 1919 and the 23rd of January 1920, the country reported 780,000 cases and at least 20,000 deaths, and health authorities recommended masks. Chicago suffered one of the first major outbreaks from mid-January 1920. Within a week its new cases surpassed the peak of the 1919 wave, though fewer patients died. New York City reported 6,374 deaths between December 1919 and April 1920, almost twice the toll of the first wave in spring 1918. Hawaii recorded 1,489 deaths at its peak in early 1920, against 615 in 1918 and 796 in 1919. Warsaw reached 158 deaths in a single week, compared with 92 in December 1918, but the epidemic there passed within weeks. Lima endured a severe wave from January to March 1920, with excess mortality about four times that of 1918-1919. Overall in the United States, the 1920 epidemic caused about a third as many deaths as the 1918-1919 experience, and western Europe called it benign.

  By mid-1920, the public and governments regarded the pandemic as over. Chile, however, suffered a milder third wave from November 1920 to March 1921. A fourth wave then struck 7 of its 24 provinces between June and December 1921.

  316386697

• 5. War and Weather

  ›

  The virus had a basic reproduction number between 2 and 3, and the conditions of World War I shaped how deadly it became. Close quarters and massive troop movements probably raised both transmission and mutation. In civilian life, natural selection favors mild strains, because the severely ill stay home while the mildly ill keep spreading the virus. In the trenches this was reversed. Soldiers with mild cases stayed where they were, while the severely ill were sent by crowded train to crowded field hospitals, spreading the deadlier virus. Some speculate that malnourishment, combat stress and chemical attacks had weakened soldiers' immune systems. Modern transportation helped the virus travel, and lies and denial by governments left populations unprepared.

  Copenhagen shows how much immunity from the first wave mattered. Its combined mortality rate was 0.29%, split between 0.02% in the first wave and 0.27% in the second. Those who recovered from first-wave infections were largely immune, which showed both waves were the same strain.

  After the second wave, cases collapsed. In Philadelphia, 4,597 people died in the week ending the 16th of October 1918, yet by the 11th of November the flu had almost disappeared from the city. One explanation credits doctors with better prevention and treatment of pneumonia. John Barry noted in his 2004 book The Great Influenza that researchers have found no evidence for this. Another theory holds that the virus mutated rapidly into a less lethal strain, since dangerous strains tend to die out along with their hosts. Fatal cases did continue into 1919. Ice hockey player Joe Hall died of the flu in April, after an outbreak that cancelled the 1919 Stanley Cup Final.

  Researchers at Harvard University and the Climate Change Institute at the University of Maine identified a severe climate anomaly over Europe from 1914 to 1919. From September to December 1918, precipitation rose across all of Europe, and mortality tracked the rainfall and falling temperatures closely. Cold, wet weather may have weakened immune systems, especially among troops exposed to incessant rain. It may also have given the virus ideal conditions to replicate and spread. Pneumococcal infections affected about one fifth of victims, with a 36% mortality rate. The anomaly likely altered the migration of avian carriers of H1N1, whose droppings contaminate water, and which reached 60% infection rates in autumn. The anomaly has also been linked to dust thrown up by the incessant bombardment, since dust particles help form the clouds that bring rain.

  Modelers later weighed the causes of the three waves in the United States. A 2013 study by He and colleagues combined school openings and closings, temperature changes and human behavior in an epidemic model. All three mattered, but changes in human behavior in response to the outbreak had the strongest effect.

  69358

• 6. Inside the Body

  ›

  Two mahogany spots over the cheekbones were the first warning of the worst cases in the second wave. Within hours the whole face turned blue, a condition called heliotrope cyanosis, which earned the illness the name 'purple death'. Black coloring then spread through the extremities, limbs and torso. Death followed within hours or days, as the lungs filled with fluid. Most infected people had only a sore throat, headache and fever, especially during the first wave. Others suffered spontaneous nosebleeds, miscarriages, a peculiar smell, falling teeth and hair, delirium, dizziness, insomnia, loss of hearing or smell, and impaired vision. One observer wrote that 'One of the most striking of the complications was hemorrhage from mucous membranes, especially from the nose, stomach, and intestine.'

  Bacterial pneumonia caused most of the deaths. Common bacteria from the upper respiratory tract entered the lungs through damaged bronchial tubes. The virus also killed directly, through massive hemorrhages and edema in the lungs.

  In 1918, the virus was too small to be seen under a microscope. Doctors blamed the bacterium Haemophilus influenzae instead, because it was large enough to see and present in many patients. A vaccine aimed at that bacillus did not make infection rarer, but it did lower the death rate. During the second wave, some feared the disease was plague, dengue fever or cholera. Others guessed typhus, which was common amid social upheaval and was striking Russia after the October Revolution. In Chile, the elite believed the nation was in severe decline, so doctors assumed typhus caused by poor hygiene. Their response was mismanaged, and mass gatherings were not banned.

  In 2005, a collaboration of the Armed Forces Institute of Pathology, the USDA ARS Southeast Poultry Research Laboratory and the Mount Sinai School of Medicine in New York City announced the virus's genetic sequence. They used tissue recovered by pathologist Johan Hultin from an Inuit woman buried in the Alaskan permafrost, along with samples preserved from American soldiers. A team led by Dr Terrence Tumpey at the Centers for Disease Control and Prevention and Mount Sinai synthesized RNA segments and rebuilt infective virus particles. They then infected mice, ferrets and macaques. In 2007, Kobasa and colleagues reported that infected monkeys showed the classic symptoms of 1918 and died of an overreaction of the immune system, known as a cytokine storm. In December 2008, Yoshihiro Kawaoka of the University of Wisconsin linked three genes, PA, PB1 and PB2, and a nucleoprotein to the virus's ability to invade the lungs and cause pneumonia. Inserted into a modern H1N1 strain, they triggered similar symptoms in animal testing.

  In the United States in 1918-1919-99% of pandemic influenza deaths occurred in people under 65. Nearly half were in young adults aged 20 to 40. Influenza typically kills infants under two, adults over 70 and the immunocompromised. The immune overreaction may explain the difference, since a stronger immune system could mean a stronger overreaction, though the claim of high young-adult mortality has been contested. Older adults may also have had partial protection from the 1889-1890 pandemic. In 1920, the death rate among people under 65 fell sixfold to half that of people over 65, yet 92% of deaths still occurred under 65.

  According to historian John M. Barry, pregnant women were the most vulnerable of all. In thirteen studies of hospitalized pregnant women, the death rate ranged from 23% to 71%. Of those who survived childbirth, over a quarter, 26%, lost the child. Another oddity was that the pandemic raged through summer and autumn in the Northern Hemisphere, although influenza is usually worse in winter.

  The reconstructed virus resembled avian viruses, and the human pandemic came before the first reports of influenza in swine. Researchers concluded that the virus jumped from birds to humans and that pigs caught it from people. More recent research suggests the strain may have arisen in a nonhuman mammal between 1882 and 1913. It diverged about 1913-1915 into the classical swine and human H1N1 lineages. The last common ancestor of the human strains dates to between February 1917 and April 1918.

  389540

• 7. Counting the Dead

  ›

  In 1927, an early estimate put global deaths at 21.6 million. A 1991 estimate gave 25 to 39 million, and a 2005 estimate gave 50 million, possibly as high as 100 million. A 2018 reassessment in the American Journal of Epidemiology estimated about 17 million, though that figure has been contested. In 2021, John M. Barry put the toll well above 100 million. Against a world population of 1.8 to 1.9 billion, these estimates correspond to between 1 and 6 percent of humanity.

  A 2009 study of fourteen European countries estimated 2.64 million excess deaths in 1918-1919, about 1.1% of Europe's roughly 250 million people. The authors suspect the severe effects of the war explain why this exceeds the American rate. The excess mortality rate in the U.K. was estimated at 0.28%-0.4%, far below that average, and as many as 250,000 Britons died. France lost more than 400,000. In Finland, 20,000 died out of 210,000 infected, and Sweden lost 34,000. In Ireland, during the worst 12 months, the flu accounted for one-third of all deaths. Russia's toll has been estimated at 450,000, a figure its own authors called a 'shot in the dark'. Another study argues that civil war and collapsing infrastructure make that unlikely, and suggests nearer 2%, or 2.7 million people.

  India lost an estimated 12 to 17 million people, about 5% of its population. The toll in British-ruled districts was 13.88 million, and 1911-1921 was the only census decade in which India's population fell. One study argues that other factors help explain the very high figures, citing unusually high mortality in 1917 and regional variation from 0.47% to 6.66%. A 2006 study in The Lancet noted provincial excess mortality from 2.1% to 7.8%, which commentators at the time attributed to nutrition and daily temperature swings. In Japan the flu killed nearly 500,000 people over two waves. In the Dutch East Indies, now Indonesia, 1.5 million were assumed dead among 30 million inhabitants. Iran, in the midst of the Persian famine of 1917-1919, lost an estimated 902,400 to 2,431,000 people, or 8% to 22% of its population. Istanbul counted between 6,403 and 10,000 deaths.

  China's toll is the most contested. The first estimate, by Patterson and Pyle in 1991, was 5 to 9 million, but later studies criticized its method. In 1998, Iijima estimated 1 to 1.28 million, based on Chinese port cities. Hong Kong and Canton reported mortality of 0.25% and 0.32%, far below Calcutta or Bombay. Shanghai, with over two million people, recorded only 266 influenza deaths among its Chinese residents in 1918. Yet these figures were gathered largely by the Chinese Maritime Customs Service, which was staffed mostly by foreigners. Hong Kong and Canton counted only deaths in colony hospitals, and the Shanghai figures covered only the International Settlement. Contemporary newspaper, post office and missionary doctor reports suggest the flu penetrated the interior. A survey in Houlu County, Hebei Province, found a case fatality rate of 9.77%, and 0.79% of the county's population died in October and November 1918.

  In the United States, about 20 million of 105 million people were infected in 1918-1919, and an estimated 500,000 to 850,000 died. Native American tribes were hit particularly hard, with 3,293 registered deaths in the Four Corners area alone. Entire Inuit and Alaskan Native villages died. Canada lost 50,000 people. Brazil lost 300,000, including president Rodrigues Alves. In New Zealand, an estimated 6,400 Pākehā and 2,500 Māori died in six weeks, and Māori died at eight times the rate of Pākehā. Australia lost around 12,000 to 20,000 people, a rate of 2.7 per 1,000 that was one of the lowest recorded. As much as 40 percent of Australians were infected, and some Aboriginal communities recorded a mortality rate of 50 percent. In Ghana, at least 100,000 died. Tafari Makonnen, the future Emperor of Ethiopia, was among the first Ethiopians infected and survived.

  Bristol Bay, Alaska, suffered a death rate of 40 percent, and some villages disappeared entirely. In Nenana, Alaska, the flu arrived in spring 1920. Most of the town was infected in the first two weeks of May, and an estimated 10% of the population died, mostly Alaska Natives. The Pacific islands suffered terribly after ships left New Zealand carrying the flu, because New Zealand was too slow to stop them. Tonga lost 8% of its population, Nauru 16% and Fiji 5%, about 9,000 people. Tahiti lost 13% in a single month. Western Samoa, occupied by New Zealand since 1914, had 90% of its 38,000 people infected. Thirty percent of adult men, 22% of adult women and 10% of children died. Custom required gathering oral tradition from chiefs on their deathbeds, which infected many elders.

  South Africa lost about 300,000 people, 6% of its population, within six weeks. Government actions after the flu arrived in September 1918 are believed to have accelerated its spread. In Kimberley, almost a quarter of the working population, diamond miners, died. In British Somaliland, one official estimated that 7% of the native population died. Even where mortality was low, so many adults fell ill that daily life stalled. Some communities closed all stores or asked customers to leave orders outside. Healthcare workers could not tend the sick, and gravediggers could not bury the dead, because they were ill too. In many places steam shovels dug mass graves and bodies were buried without coffins.

  A few places escaped. American Samoa and the French colony of New Caledonia used quarantines to avoid any deaths during the pandemic. The flu arrived in 1926 for American Samoa and in 1921 for New Caledonia, once quarantine ended. Iceland shielded a third of its population by blocking the island's main road. Marajó, an island in Brazil's Amazon River Delta, never reported an outbreak, and Saint Helena reported no deaths.

  In Oslo, death rates were inversely correlated with apartment size, so the poor in small apartments died at higher rates. Italian Americans, a recently arrived group, were nearly twice as likely to die as the average American. African Americans, paradoxically, were relatively spared. More men than women died, since men went out more and were likelier to have tuberculosis. In India the opposite held, possibly because women ate poorly and were expected to nurse the sick. Some parts of Asia had death rates 30 times higher than some parts of Europe. Mortality in Hungary and Spain was three times that of Denmark, and cities generally fared worse than rural areas.

  459210084

• 8. Response and Memory

  ›

  In 1918, systems for alerting public health authorities to infectious spread existed, but they generally excluded influenza, and the response came late. Officials closed schools, theatres and places of worship, limited public transportation and banned mass gatherings. Face masks became common in places such as Japan, though people debated whether they worked. In San Francisco, the Anti-Mask League resisted them. In New York City, the health commissioner ordered businesses to open and close on staggered shifts so the subways would not overcrowd. A later study found that banning gatherings and requiring masks could cut the death rate by up to 50 percent, but only if imposed early and not lifted prematurely. The vaccines of the day were based on bacteria rather than the virus, so they could only help with secondary infections.

  With no antiviral drugs and no antibiotics, doctors reached for aspirin, quinine, arsenics, digitalis, strychnine, Epsom salts, castor oil and iodine. Bloodletting, ayurveda and kampo were also applied, with varying degrees of effect.

  The Italian newspaper Corriere della Sera was prohibited from reporting daily death tolls. Newspapers of the time were also paternalistic and worried about mass panic. Rumors filled the gap. In Ireland, some believed noxious gases were rising from the mass graves of Flanders Fields and being 'blown all over the world by winds'. Others said Germans were behind the disease, whether by poisoning the aspirin made by Bayer or by releasing poison gas from U-boats.

  Political scientist Andrew Price-Smith argues that the virus helped tip the war toward the Allies. The waves hit the Central Powers first, and a 2006 Lancet study found excess mortality of 0.76% in Germany and 1.61% in Austria, against 0.34% in Britain and 0.75% in France. Historian Nancy Bristow argues the pandemic, combined with more women attending college, helped women succeed in nursing. Doctors, mostly men, had failed to contain the illness, while nurses, mostly women, celebrated their success in patient care. A 2006 study in the Journal of Political Economy found that people who were in the womb during the pandemic had lower educational attainment and income and more physical disability. The flu has also been linked to the outbreak of encephalitis lethargica in the 1920s.

  The winter of 1921-1922 brought the first major return of seasonal influenza in the Northern Hemisphere. All-cause mortality in the Netherlands roughly doubled in January 1922 alone. The virus, H1N1, stayed endemic, and every flu pandemic since has been caused by its own descendants. In 1957, the novel H2N2, a reassortment of human H1N1 and an avian virus, displaced it. H3N2 then displaced H2N2 in a pandemic in 1968. In 1977, a virus closely resembling the seasonal H1N1 of the 1950s appeared in the USSR and caused a 'technical' pandemic that mainly affected people aged 26 and under. Explanations range from a virus frozen for 20 years to an accidental lab leak. In 2009, a novel H1N1 emerged, caused a pandemic, and replaced the seasonal H1N1 to circulate alongside H3N2.

  Historians have called the Spanish flu a 'forgotten pandemic'. Historian Guy Beiner has challenged that label, showing that it was overshadowed by commemoration of the First World War in mainstream historiography but remembered in private and local traditions across the globe. The pandemic killed most of its American victims in under nine months, which limited media coverage. Typhoid, yellow fever, diphtheria and cholera had also made epidemics familiar. In some areas the only mention of the flu was in advertisements for medicines claiming to cure it. Most deaths, from both the war and the flu, were among young adults, so war casualties may have overshadowed the flu. The last known survivor, José Ameal Peña, was alive as of 2020.

  The pandemic never became a major theme in American literature, and Alfred Crosby suspects the war overshadowed it. Katherine Anne Porter's 1939 novella Pale Horse, Pale Rider remains one of the best-known fictional accounts. The 2006 novel The Last Town on Earth follows a town that tries to stop the flu by barring entry and exit. Emma Donoghue's 2020 novel The Pull of the Stars is set in Dublin during the pandemic. Publishers fast-tracked it because the COVID-19 pandemic was then under way.

  4797603043