Anorexia Nervosa: What It Is, and What Actually Helps

Anorexia is a mental illness in which someone eats far too little and is terrified of gaining weight.

The third part of the diagnosis is where people get it wrong. It is usually described as being unable to see how much danger you are in, and that does happen. But the criterion is broader than that: it is met by a distorted sense of your own body, or by weight dominating how you judge yourself, or by not recognising the seriousness of the situation. Any one of the three.

So the popular picture, someone who genuinely cannot tell they are ill, fits some people and not others. Plenty know exactly what is happening to them. They can recite their own blood results and tell you what their heart is doing and carry on anyway, because the illness does not need them to be ignorant of the risk. It only needs the risk to matter less than the alternative.

That is worth holding onto, because it changes what you are dealing with. Telling someone the facts works when the problem is that they do not know. Here it often is not.

It is not about vanity. It is not a diet that went too far. People with anorexia often describe it as feeling like control, or safety, or being good at something. That is why it is so hard to give up, and why telling someone to just eat does nothing.

This is the part most people get wrong.

Someone can have every symptom of anorexia and still be a normal weight, or a large one, if they have lost a lot of weight to get there. Doctors call this atypical anorexia. It is not a milder version.

In one study that followed 496 teenage girls for eight years, 2.8% developed atypical anorexia and 0.8% developed the classic kind, which is about three and a half times as many.

That is one group of girls in one country, not a universal ratio. But every study that has looked has found atypical anorexia to be at least as common as the classic kind, and often more so.

Men get it. Roughly one man for every thirteen or fourteen women, and they are diagnosed late because nobody looks for it. Children get it. Adults in their thirties and forties get it. It is not confined to any race or income level, and being poor or food-insecure does not protect you.

Two other illnesses get confused with it. ARFID is when someone barely eats because of texture, taste, or a fear of choking, with no worry about weight at all. Bulimia is binge eating followed by purging, usually at a normal weight. Both are real, both are different.

Anorexia has one of the highest death rates of any mental illness.

You will often see it called the deadliest. That is the popular version and it is not quite right: drug and alcohol addictions kill at higher rates, with opioid addiction running at roughly fourteen times the expected death rate against anorexia’s five to six. Among mental illnesses that are not addictions, anorexia is at or near the top.

Someone with anorexia is roughly five to six times more likely to die than someone their age without it. That figure comes from several separate reviews covering tens of thousands of patients, and they all land in the same place.

About one in five of those deaths is suicide. About one in five is the heart giving out. The rest are the other things starvation does.

A Swedish study followed six thousand women who had been hospitalised. Their risk of dying was still raised more than twenty years later.

Starvation damages nearly everything. The heart slows and can develop dangerous rhythms. Blood pressure drops. Bones thin, and break more easily. Periods stop. Growth stalls in children. The gut slows down. Blood counts fall. Salt levels in the blood go wrong, which can cause seizures. The brain physically shrinks, though that largely reverses when someone eats again.

Feeding someone back up is itself risky if done too fast, which is why it is supervised.

You will see wildly different recovery rates quoted. There is a reason, and it is worth knowing.

Roughly half of the people who had not recovered at nine years recovered later.

So when a study says only a third of patients get better, check how long it watched them. A short study does not find a worse illness. It finds a shorter window.

Recovery is slow and it is common. Something like one in five people stays ill long-term, though that figure is a rough average across studies rather than a finding from any single one. Depression alongside the anorexia is the strongest warning sign for a long course.

No, not in the way people mean when they ask.

It runs in families. Studies of twins put the genetic contribution at roughly half to two thirds. That does not mean a gene for anorexia exists. It means some people are far more vulnerable than others before anything happens to them.

The genetics look strange in a telling way. The largest genetic study, nearly 17,000 patients, found anorexia shares genetic ground with other mental illnesses like OCD, and also with how the body handles blood sugar and fat. That second part was unexpected. It does not prove metabolism causes anorexia, but it makes a purely cultural explanation very hard to sustain.

Culture matters, but not the way people think. When television arrived in Fiji in 1995, researchers tracked teenage girls before and after and found real increases in disordered eating and body dissatisfaction. Media clearly shapes what the illness fixes on. It does not create the illness, or everyone watching would have it.

There is a serious argument that some choice is involved, and it is not made by people being cruel. It comes from clinical ethicists who interviewed patients at length.

Their point is that people with anorexia are not unable to think. They can reason clearly about most things. What changes is what they value — weight comes to matter more than health, more than relationships, sometimes more than staying alive. And the illness stops feeling like an illness and starts feeling like who they are.

So they do make choices. They attend appointments, argue about meals, decide things. Treating them as having no say is both wrong and unhelpful.

The honest answer is that their ability to choose is damaged and narrowed, not gone. Which is nothing like choosing a haircut, and also nothing like a seizure.

It is called family-based treatment. Parents take charge of their child’s eating for a while, then hand control back as they recover. Guidelines are blunt that families are not to blame for causing this.

Notice that the difference shows up after treatment finishes, not during it. Notice also that even the best treatment leaves half of patients not fully recovered at a year.

And treat “best evidenced” as a comparative rather than a verdict. The trials are few and small, Cochrane’s review of family therapy for anorexia found only limited, low-quality evidence, and recovery is defined so differently across studies that one reanalysis of the same patients produced figures ranging from 22% to 88% depending on the cutoff applied. This is the strongest evidence in the field and it is still that thin.

The largest trial that pitted the main adult therapies against each other, 120 patients, found no meaningful difference between them. All helped somewhat. None won.

That sounds like bad news and it has a useful implication: since no therapy beats the others, the one the patient will actually turn up for is the right one.

This is not a gap waiting to be filled. It is the finding.

Antipsychotics help slightly with weight gain and do nothing for the psychological symptoms. Antidepressants were properly tested for preventing relapse in patients who had already recovered weight, and performed no better than placebo.

Medication can treat depression or anxiety alongside anorexia. Guidelines say it should never be the only treatment.

Relapse is common, and the first year after treatment is the riskiest. The strongest predictor is how much weight was restored before treatment ended.

Three things, if you only read this far.

Know the emergency signs. Fainting, chest pain, palpitations, confusion, a seizure, or being unable to stand from squatting without pushing with the arms. Those mean an ambulance. And someone can be in real danger while looking well and having normal blood tests, so normal results are not reassurance.

Do not comment on how they look. Including kindly. Patients report hearing “you look healthier” as “you have gained weight.”

Get an assessment early, and if it is a teenager, ask specifically about family-based treatment.

In 2022 three clinicians proposed that a small number of patients with very long-standing anorexia should be considered terminal, opening the door to palliative care and, in some places, assisted dying.

The backlash was severe. Specialists pointed out that people recover from anorexia after decades, and that calling it terminal writes off patients who might still get better. Patients themselves argued the criteria ignored how often good treatment is simply unavailable.

This one is unresolved. Anorexia can leave someone's reasoning intact while changing what they care about, so a patient can pass a standard test of decision-making capacity while the decision itself is being made by the illness. There is no agreed way to handle that, and practice varies by country.

1. Anorexia is eating too little and fearing weight gain, alongside a distorted sense of the body, or weight dominating self-worth, or not recognising the danger. Any one of those three. Plenty of people know exactly how ill they are and carry on regardless.

2. You do not have to be underweight to have it, and that version is about three and a half times more common. It is not milder.

3. It has one of the highest death rates of any mental illness, roughly five to six times the expected rate, with a fifth of those deaths being suicides. Addictions kill at higher rates; among mental illnesses that are not addictions, anorexia is at or near the top.

4. People recover, often slowly. One study found 31% recovered at nine years and 63% at twenty-two. Short studies make recovery look rarer than it is.

5. It is not a choice. It is around half to two thirds genetic, and a starvation experiment on healthy men reproduced most of the behaviour.

6. For teenagers, family-based treatment works best. For adults, no therapy clearly beats another. No drug treats it.

7. "Terminal anorexia" is not a real category, and the person who proposed the term retracted it.