At the physical, Maxwell stared at her doctor in disbelief. She always thought that eating disorders were for skinny people. “I laughed,” she says. “I don’t use language like this any longer, but I told her she was crazy. I told her, ‘No, I have a self-control problem.’”
For centuries, the eating disorder that would become known as anorexia nervosa mystified the medical community, which struggled to understand, or even define, an illness that caused people to deliberately deprive themselves of food. As cases rose over the course of the 19th and 20th centuries, anorexia was considered a purely psychological disorder akin to hysteria. Sir William Withey Gull, an English physician who coined the term “anorexia nervosa” in the late 1800s, called it a perversion of the ego. In 1919, after an autopsy revealed an atrophied pituitary gland, anorexia was thought to be an endocrinological disease. That theory was later debunked, and in the mid-20th century, psychoanalytic explanations arose, pointing to sexual and developmental dysfunction and, later, unhealthy family dynamics. More recently, the medical field has come to believe that anorexia can be the product of a constellation of psychological, social, genetic, neurological and biological factors.
Since anorexia nervosa became the first eating-related disorder listed in the Diagnostic and Statistical Manual of Mental Disorders in 1952, its criteria have shifted as well. Initially, anorexia had no weight criteria and was classified as a psychophysiological disorder. In a 1972 paper, a team led by the prominent psychiatrist John Feighner suggested using a weight loss of at least 25 percent as a standard for research purposes, and in 1980, the D.S.M. introduced that figure in its definition (along with a criterion that patients weigh well below “normal” for their age and height, although normal was not defined). Doctors who relied on that number soon found that patients who had lost at least 25 percent of their body weight were already severely sick, so in 1987, the diagnosis was revised to include those who weighed less than 85 percent of their “normal” body weight (what qualified as normal was left to physicians to decide). In the 2013 D.S.M., the criteria shifted again, characterizing those who suffer from anorexia as having a “significantly low weight,” a description that would also appear in the 2022 edition.
In that 2013 edition, a new diagnosis appeared — atypical anorexia nervosa — after health care providers noticed more patients showing up for treatment with all the symptoms of anorexia nervosa except one: a significantly low weight. Those with atypical anorexia, doctors observed, suffer the same mental and physical symptoms as people with anorexia nervosa, even life-threatening heart issues and electrolyte imbalances. They restrict calories intensively; obsess about food, eating and body image; and view their weight as inextricably linked to their value. They often skip meals, eat in secret, adhere to intricate rules about what foods they allow themselves to consume and create unusual habits like chewing and spitting out food. Others exercise to the point of exhaustion, abuse laxatives or purge their meals. But unlike those diagnosed with anorexia, people with atypical anorexia can lose significant amounts of weight but still have a medium or large body size. Others, because of their body’s metabolism, hardly lose any weight at all. To the outside world, they appear “overweight.”
Starting in the mid-2000s, the number of people seeking treatment for the disorder rose sharply. Whether more people are developing atypical anorexia or seeking treatment — or more doctors are recognizing it — is unknown, but this group now comprises up to half of all patients hospitalized in eating-disorder programs. Studies suggest that the same number of people, even as many as three times as many, will develop atypical anorexia as traditional anorexia in their lifetimes. One high estimate suggests that as much as 4.9 percent of the female population will have the disorder. For boys, the number is lower — one estimate was 1.2 percent. For men, it is likely even lower, though little research exists. For nonbinary people, the number jumps to as high as 7.5 percent.
