By 1919, the Spanish Flu pandemic had spread influenza to a third of the world's population, or around 500 million people. Psychiatrist Karl Menninger was treating people at Boston Psychopathic Hospital who'd recently been infected. But his patients had symptoms far beyond what's usually associated with the flu. In a paper on 100 cases he saw over three months, he described seeing extreme mental disturbances—over half of his patients had some sort of psychosis, and almost two-thirds had hallucinations.
Then, in a follow-up study, Menninger reported that nearly all of those patients got better. Their madness disappeared when they recovered from the flu. He suggested that, in these patients, there might be some connection between the flu and their psychiatric illnesses.
It wasn't the first time this association has been made. While today, we consider viral infections to be diseases of the body—they infect the lungs, give us fevers, stuffy noses, or a cough—throughout history there’s also been a strange link between influenza and psychotic disorders similar to schizophrenia, a severe mental disorder that can affect how people think. More recently, this link re-emerged from the observation that babies born in the winter or early spring, when mothers may have been exposed to the flu, are more likely to develop schizophrenia as adults.
Infection and inflammation are now recognized as risk factors for mental disorders of all kinds. Yet, the underlying reason for the flu’s sticky relationship with schizophrenia is still a mystery. Some experts propose that influenza could interfere with fetal development through a mother’s immune system, or that influenza could bring on some kind of autoimmune disorder that interacts with the brain.
Figuring out a potential connection between widespread infection and mental illness has a larger purpose: it could lead to new treatment options for select individuals whose disorders have an immunological origin. And, with a new viral infection sweeping the globe, one that’s not the flu but nonetheless shares some things in common with past pandemics, it’s especially relevant to lend a critical eye to how viruses and infections could influence our minds.
The idea that psychosis has an infectious origin has deep historical roots. There are multiple examples from hundreds of years ago in which people got infected with what we now recognize as the flu, and had periods of mental instability during or immediately after.
After outbreaks of influenza in St. Petersburg, Russia, in 1889, people experienced insomnia, depression, suicidal thoughts, and homicidal urges. One neurologist wrote that it wasn't uncommon for people with the flu to feel “dark forebodings of…impending disaster” or to think they had committed a “fearful crime” and were about to be punished.
“I have seen men, within twenty-four hours of taking the influenza, sob like children for hours together as though their hearts would break,” England's Prime Minister Lord Rosebery wrote of influenza in 1895. “The sacredness of life is seen only through a flimsy veil of conflicting emotions of doubt, of dread, and of determination to burst these bonds by one fell swoop of severance from the mortal to immortality.”
Watch more from VICE:
During historical influenza epidemics, it was common to refer to the “psychoses of influenza.” In the mid-1890s, British medical journals were full of tales like these, according to Mark Honigsbaum, a historian of medicine at City University of London.
“These kinds of stories have been recounted again and again,” said Tom Pollak, a psychiatrist and researcher at King's College London. He and his colleagues recently published a review paper on the associations between influenza and schizophrenia. It came out for the 100-year anniversary of the 1918-1919 Spanish influenza pandemic, but the timing turned out to be "slightly unnerving," as a viral infection, COVID-19, caused by a new coronavirus, spreads throughout the world. The outbreaks were officially declared a pandemic by the World Health Organization.
More recent studies have continued examining the connection between influenza and schizophrenia—but through the flu’s potential effects on a developing fetus. More than 200 papers have found that winter- and spring-month births are associated with greater risk for schizophrenia, raising the possibility that if mothers get the flu, it could somehow affect their children. In 1988, psychologist Sarnoff A. Mednick looked into the 1957 influenza pandemic and showed that there were higher rates of schizophrenia in people born during the pandemic compared to a group of people born shortly afterwards.
In all, over 25 epidemiologic studies have now looked specifically at the flu-schizophrenia connection in pregnant women. Yet, the results have been mixed. Half have replicated the finding, and half have not. The studies may have inconsistent results because it’s hard to know exactly who did and didn’t have the flu when you look into the past and don't have data to prove who was sick.
Alan Brown, a professor of psychiatry and epidemiology at Columbia University, tried to fix that in a study from 2004: He examined medical records of more than 12,000 women in California who were pregnant between 1959 and 1966 who had blood samples available—making it possible to test if mothers had antibodies present for the flu.
Brown and his colleagues followed up on the children who were eventually diagnosed with schizophrenia or similar psychotic disorder, and compared their mother's blood samples to matched controls. They found that children of women who had been exposed to the flu during the first half of pregnancy were three times more likely to have schizophrenia. (About 11 percent of pregnant women get the flu during pregnancy.)
While not all the women who were exposed to the flu had children with schizophrenia—meaning it wasn't a sure thing—they concluded that their data suggested up to 14 percent of schizophrenia cases would not have occurred if those mothers hadn’t been exposed to influenza during early to mid-pregnancy.
In Denmark, large studies have found that infections and autoimmune diseases throughout life are associated with the increased risk of many other mental disorders too. A study from 2016 looked at every person born in Denmark from 1983 to 2002 and found that those with infections treated with anti‐viral medications and those requiring hospitalizations were more likely to have schizophrenia and affective disorders.
These ties between infection and mental illness are varied, and we don't yet understand if they are knotted together, or completely separate. There’s evidence that children who get a lot of infections when they’re young have increased rates of getting schizophrenia, and that postmortem brains of people with schizophrenia have abnormal immune cells. In 1918, Menninger observed that psychiatric symptoms can arise right after influenza, while research from the 1957 pandemic suggested that the effects waited until children grew up, years later.
It's unknown if all these risks happen by the same or different mechanisms. But taken together, it's evidence that shouldn't be ignored, said Pollak. "I think the idea that infection is a risk factor for the development of psychiatric illness is pretty much indisputable now," he said. “It gives us a mandate to start looking at the immune processes that might underlie these psychiatric disorders.”
One fringe theory is that the body might make antibodies that trigger an immune response that attacks the brain. One example where something similar to this happens is autoimmune encephalitis. Like in Susannah Cahalan's memoir, Brain On Fire, autoimmune encephalitis can severely compromise mental health and look a lot like psychosis. But there's no hard evidence yet of any such response for the flu.
How might a viral infection affect the brain in a growing fetus? Brad Pearce, a neuroscientist and epidemiologist at Emory University and author of Can a Virus Cause Schizophrenia?, said that other infections, like rubella, can cross the placenta. In one study, 20 percent of people exposed to rubella in the first trimester developed schizophrenia as an adult, compared to the 1 percent of the population that gets it normally. “The organism itself, the virus, gets into the [fetus’s] brain and it alters the growth of brain cells,” Pearce said.
The story for influenza must be different, because it doesn’t appear to attack the fetus’s brain directly. But when pregnant mothers get sick, Pearce said, their bodies are doing a lot to fight the virus, like making inflammatory molecules.
“You feel like crap,” Pearce said. “You have zero energy, you can’t get up. A lot of that is from pro-inflammatory molecules that are floating around.” Those symptoms and molecules may be subtly altering how a baby’s brain is forming, even though the baby doesn’t get infected with the virus itself.
This concept has been observed in animal models, like mice given immune stimulants at different phases of pregnancy. Mice don’t actually have to be infected by something for their babies to have behavioral differences—if researchers trigger a severe immune reaction (the same kind that occurs after an infection), it’s enough. Rhesus monkeys that were infected with the flu during pregnancy had babies with smaller brains and other abnormalities similar to those seen in schizophrenia, Spectrum reported. (There is no evidence, though, of an association between the immune response brought on by the flu vaccine and schizophrenia or other psychotic disorders.)
Pollak said it could be that these viral infections and immune responses don’t cause schizophrenia on their own, but contribute to an overall risk. Someone might then need to have an additional event happen later in life, like a second severe infection, psychological stress, or trauma—all overlaid over genetics.
“The infection is a sort of a primer and it puts you on this trajectory, which is abnormal, but you probably need another hit later on in order to develop the full illness,” Pollak said.
The next step is to better understand how the immune system affects the brain more generally, said Emily Severance, an assistant professor of pediatrics at Johns Hopkins University. “This disorder is complex and multifactorial and also has a genetic component that we haven’t yet addressed,” Severance said.
When geneticists started to look for genes that are responsible for schizophrenia, large genome-wide association studies (GWAS) have often picked up genes that are associated with the immune system.
“Neuroscientists can think of the brain first and everything else, the body, as second,” Pearce said. “But kind of to their surprise, immune areas of the genes that were expressed in immune cells kept coming up. So now I think we've accepted the fact that the strongest signals for schizophrenia and GWAS studies are still around immune areas.”
And what about the new coronavirus? There are few answers about how pregnant women should operate during this pandemic—though one very small study in China found that three babies born to mothers with COVID-19 did not test positive for the virus when they were born.
Only one previous study has looked at coronaviruses and psychotic symptoms later in life. In 2011, Severance and her colleagues did a study on four kinds of coronaviruses (coronaviruses are a large family of viruses, and they chose four previously known ones). They found adults with schizophrenia had higher exposure rates—they found more antibodies to the coronaviruses in that group than in people without schizophrenia. What that means for the current COVID-19 outbreak is hard to say, but Severance noted that “the association between infection and psychosis would be relevant for any emerging pathogen, especially for women who are pregnant."
“We honestly don’t yet know what's going to happen, but if you had to extrapolate from the flu, you could say women that are pregnant with really severe symptoms, their children could be at higher risk for neurodevelopmental consequences and that would conceivably include schizophrenia," Pearce said. In the current coronavirus outbreaks, for people with mild symptoms, he said the risk is low.
Pollak wouldn’t be surprised if there were a few cases of psychosis associated with the new coronavirus—it happened during the SARS epidemic in Hong Kong: Hallucinations, mania, and depression were also observed after the SARS epidemic. In 2015, researchers reported two cases of children who had H1N1 influenza and psychosis that eventually went away.
“Certainly when something affects as many people as this will, it would be odd not to see at least a few cases reported," he said. About 8,000 people worldwide got sick with SARS; as of publication time, there are more than 170,000 confirmed cases of COVID-19.
And there is one more important factor that focusing too much on antibodies or inflammation can miss: the psychological effects of pandemics. “They essentially cast a shadow of fear around the world, paranoia," Pollak said. “It’s sometimes very difficult to disassociate these effects from the acute biological effects, and possibly what we saw with these earlier sorts of psychosis of influenza [in 1919] might've actually reflected some more of those kinds of psychosocial factors as well.”
Remember that it's not likely that infections cause most or even many cases of schizophrenia. But, the unsettling possibility exists that they may contribute to a small number of them. This link between infection and mental illness can, at the very least, serve as a reminder that mental illness isn’t just about the brain. “The body and the brain are integrated,” Pearce said. “The immune system influences the brain and vice versa.”
It also suggests that there are probably different avenues to developing schizophrenia, and different subtypes of schizophrenia. In some people, the immune system might be a heavily contributing factor, but maybe in others it isn’t. If we can figure out whose schizophrenia is caused by these immunological factors, the treatment options for those people might be different. Some treatments could, for example, target inflammation in the brain, rather than giving people drugs that alter other brain chemicals.
Pollak finds a sense of optimism in this emerging field of immunopsychiatry—which is that there could be a subset of patients who have an inflammatory basis to their symptoms, and might be able to respond to a different kind of treatment.
He's pursuing this with his colleagues now: they’re looking at groups of 300 to 400 young people at high risk for developing psychosis, or who have just had a first episode. They measure their exposures to different kinds of infections, like influenza, herpes, toxoplasma, Epstein–Barr virus, or Cytomegalovirus.
“We then ask, ‘Do the folks who have been exposed to these pathogens look any different in some way?'" Pollak said. "Do they have more severe symptoms? Do they have abnormal brain scans or are particular brain areas smaller or larger? Do they have evidence of an autoimmune attack on the brain?' But most importantly: 'Does the fact they have been exposed to these pathogens tell us anything about their outcome?'"
The hope is that someday researchers could use infections as a kind of roadmap, or biomarker, to decide what treatment a person would receive—maybe a person with high levels of past infections would get anti-infective agents, or other immunologically-targeted therapies better suited for them.
"We haven't had a new development in treatment of schizophrenia, a new pharmacological development, for a very long time," Pollak said. "If there's even a small subset of patients who would benefit from an immunologically-targeted therapy, then that's potentially going to be transformative for those patients."
Sign up for our newsletter to get the best of VICE delivered to your inbox daily.
Follow Shayla Love on Twitter.