GBALLGIGGSPHOTO / Shutterstock

In 2007, Carlo Carandang, then an attending physician at a hospital in Halifax, Nova Scotia, saw a most unusual patient: an 8-year-old boy who had recently adopted some strange beliefs, all while losing 18 pounds. The boy thought that nurses were “evil,” and that he could inject other people with his fat cells simply by walking past them.

The boy’s symptoms had begun a few months prior. After his school held a lesson on healthy eating, he started to scrutinize food labels and avoid fat and carbs, according to Carandang, who now works as a data scientist. The boy worried that he was too fat, and he would examine his stomach in the mirror throughout the day. He grew suspicious of what his mother might be putting in his food and began preparing all of his own meals. Before long, he was eating just 200 calories a day.

The boy also developed odd tics, flapping his arms and tapping his mouth to undo what he perceived to be “contamination.” Carandang knew that the boy had always been an anxious child and that he had a history of recurring strep throat. But the food-related symptoms far exceeded what would normally accompany anxiety. The boy was admitted to the hospital, where it took months for Carandang and his team to successfully treat him. Ultimately, the boy had to get his tonsils taken out to stop the strep infections. Around the same time, his eating disorder stopped.

In a report about the patient, Carandang wrote that the boy appeared to have PANDAS, or pediatric autoimmune neuropsychiatric disorders associated with streptococcal infection, a type of obsessive-compulsive disorder that sometimes comes on in children after a bout of strep throat. While PANDAS is a fairly well-established condition, it was unusual that the infection-induced psychological symptoms had brought about an eating disorder.

Other researchers have reported isolated cases of children developing eating disorders after coming down with infections. In the late 1990s, Mae Sokol, a psychiatry professor at Creighton University, described treating several patients whose eating disorders had begun after strep infections. One 12-year-old treated by Sokol lost 30 pounds after he suddenly became afraid to eat fats and liquids. He had experienced an untreated upper-respiratory-tract infection just a month before the symptoms began. A 16-year-old had a series of upper-respiratory-tract infections, then suddenly became concerned about weight gain and “dead animals on plates,” according to Sokol’s report.

These cases hinted at a relationship between the infections and the subsequent disordered eating, but childhood infections are so common, and eating disorders so multifaceted, that scientifically connecting the two conditions has been hard. It seems so counterintuitive: Why would a sore throat lead to a state in which a person feels irrationally preoccupied with thinness? This year, though, a large study found that the boys Carandang and Sokol treated weren’t isolated incidents. Infections might, in fact, spark eating disorders in some people.

For the study, Lauren Breithaupt, a clinical psychologist at Massachusetts General Hospital, and several of her colleagues from Denmark and North Carolina looked at the health histories of 525,643 Danish teen girls born from 1989 to 2006. (The rate of eating disorders among Danish boys was too low for them to be included in the analysis.) The researchers examined the girls’ medical records to see if they had ever been hospitalized for a variety of infections, including rheumatic fever, strep throat, viral meningitis, mycoplasma pneumonia, coccidioidomycosis, or influenza, and also whether they had ever been diagnosed with an eating disorder.

A connection between the two ailments immediately became clear. The overall number of girls diagnosed with eating disorders was relatively small—as it is in the United States. But the teens hospitalized with a severe infection were 22 percent more likely to be diagnosed with anorexia, 35 percent more likely to be diagnosed with bulimia, and 39 percent more likely to have an eating disorder that doesn’t quite meet the criteria for an anorexia or bulimia diagnosis. The diagnosis of an eating disorder tended to happen soon after the infection took place, such that the girls were at their greatest risk of developing one within the first three months after being hospitalized for an infection.

The study seemed to crystallize the connections among infections, obsessive behavior, and eating disorders that Breithaupt and other researchers had been seeing. In her work as a psychologist, Breithaupt says, she has seen patients who, after an infection, “have really rigid thoughts and impressions about either food or weight or its shape, or they might have lots of concerns about fat in foods and fat in their body.” Carandang’s PANDAS patient, too, seemed to first grow obsessed with food, then fixate on avoiding it.

No one knows exactly why, precisely, infections might spark eating disorders. Breithaupt suggests that either the infection itself or the antibiotic used to treat it might be disrupting the patient’s gut microbiome, the collection of microorganisms in the intestine that plays a role in health and disease. This disruption might change the amount of chemicals called neuropeptides circulating in the gut. Because the gut communicates with the brain, the quantities of neuropeptides circulating in the brain might then change, as well. That could, in essence, make people think differently about food or their body.

Perhaps other mechanisms are at play. One competing theory is that the body’s own immune response to an infection might end up invading the brain. When the body senses a dangerous bug, it produces proteins that destroy the invader. But some of those proteins can also attack our own cells. In possible cases of anorexia or bulimia induced by bacteria, some scientists suspect that these proteins get into parts of the brain that control impulses such as disgust and hunger. There, they might attack the brain tissues or switch on the “I’m not hungry anymore” impulse, or even the “I’m disgusted by my own body” impulse.

There’s no direct evidence for these theories; for now they’re merely speculation. And even if one of them proved correct, researchers would still have to contend with the mystery of why people get infections all the time but relatively few develop eating disorders. Or, for that matter, why not everyone with an eating disorder recently dealt with an infection.

It might be that underlying factors about people predispose them to developing an eating disorder after an infection. “Maybe you have more of a genetic risk for obsessive-compulsive disorder or anorexia, and the infection then unmasks that vulnerability. That’s one possibility,” says Kyle Williams, the director of the pediatric-neuropsychiatry-and-immunology program at Massachusetts General Hospital for Children.

If confirmed, these findings could eventually affect how eating disorders are treated, leading doctors to check if their eating-disorder patients have any lingering infections, Breithaupt says. The results also have the potential to radically change our notion of the many ways eating disorders might originate. While most professionals acknowledge that anorexia and bulimia are deeply psychologically rooted, some eating-disorder patients still face stigma for supposedly being so “vain” as to starve themselves. It’s less likely that people would accuse a person of getting meningitis on purpose. Similarly, people who are attacked for compulsively dieting out of vanity might simply be under the spell of antibodies gone awry.

Jim Morris, a professor at Lancaster University in the United Kingdom, says there are still too many unanswered questions to begin treating patients with eating disorders any differently. Instead, he says, this research should prompt a consideration of just how closely intertwined our brains are with our bodies. Just as some problems that seem physical might have psychological aspects, some problems that seem psychological might have physical instigators.

“We say disease is due to biological factors, social factors, and psychological factors all interacting together,” Morris says. “Well, it works with psychiatric disease as well.”

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.