How Many People Recover From Anorexia?

Roughly two out of three people with anorexia nervosa eventually recover, but that number comes with a crucial caveat: it takes a long time. A landmark longitudinal study found that about 63% of participants had recovered by the 22-year mark, up from just 31% at the 9-year follow-up. Recovery from anorexia is real and common, but it rarely looks like a clean arc from sick to well. The timeline, what “recovered” even means, and the factors that tilt the odds are all more complicated than a single percentage can capture.

What “Recovery” Means Depends on Who You Ask

One of the frustrating realities about recovery statistics is that researchers have never agreed on a single definition. As one review put it, there are nearly as many definitions of recovery in eating disorders as there are studies on the topic.¹ Some studies define recovery purely by weight restoration, others by the absence of binge-purge behaviors, and still others require normalized scores on psychological questionnaires measuring body image distress and food-related anxiety. That range of definitions means a study reporting 70% recovery might be counting people who are weight-restored but still deeply preoccupied with food, while a stricter study reporting 40% recovery might be filtering out everyone with residual body dissatisfaction.

The field has gradually moved toward a more comprehensive framework. Research suggests that genuine recovery should include physical health (a stable, adequate body weight), behavioral change (no restriction, bingeing, or purging), and psychological normalization (thinking about food and body shape in ways that resemble someone who never had an eating disorder).² When researchers applied that three-part definition in one study, the “fully recovered” group was indistinguishable from people who had never had an eating disorder on every measure used. The “partially recovered” group, people who met some criteria but not all, still showed lingering body image difficulties and some interference with daily functioning. This distinction matters because partial recovery is common and represents a gray zone that gets counted differently depending on the study.

Recovery Happens Slowly, and Later Than You Might Expect

If there is one finding that reshapes how people should think about anorexia recovery, it is this: the process often takes years or even decades, not months. In one of the most thorough long-term studies available, only about 31% of participants with anorexia had recovered by the 9-year follow-up. By 22 years, that number had roughly doubled to 63%. About half of those who were still unwell at the 9-year point went on to recover over the following 13 years.³

That finding carries an important practical message: not recovering within the first few years of treatment does not mean someone will never recover. Many people who appear “stuck” at the 5- or 10-year point do eventually reach full recovery. The timeline is just much longer than the timelines associated with most other mental health conditions. This is partly because anorexia involves deeply entrenched patterns of thought, behavior, and physiological adaptation that resist quick resolution, and partly because access to effective treatment is uneven and often interrupted.

At the same time, earlier treatment clearly helps. Treatment response tends to be strongest in the first three years after onset, and the longer the illness persists untreated, the more difficult recovery becomes.⁴ A systematic review found that longer duration of untreated illness was linked to greater symptom severity, lower motivation to change, poorer treatment response, and worse long-term outcomes.⁵ This creates a painful paradox: the illness itself often makes people reluctant to seek help, and each year of delay chips away at the likelihood of a smooth recovery.

Relapse Is Common, Especially in the First Year

Recovery from anorexia is rarely a straight line. Relapse rates across studies range from about 9% to 52%, with higher rates appearing in studies with longer follow-up periods.⁶ The first year after treatment ends is consistently identified as the highest-risk window. One study using a relapse-prevention program found that no full relapses occurred in the first four months, but the peak danger zone for relapse fell between months 4 and 16.⁷ In that particular cohort, about 11% experienced a full relapse and another 19% a partial one, while 70% maintained their gains.

The wide range in reported relapse rates is partly a measurement problem (studies define relapse differently, just as they define recovery differently) and partly a reflection of how much circumstances vary. Someone discharged from an inpatient program into a supportive home environment with follow-up therapy faces very different odds than someone returning to the same stressors that contributed to the illness with no ongoing support.

When Anorexia Shifts Shape Instead of Disappearing

Something that complicates recovery statistics is diagnostic crossover, where a person’s eating disorder changes form rather than resolving. Over a 7-year study, more than half of women with anorexia moved between the restricting and binge-purge subtypes. About a third crossed over to bulimia nervosa, though many of them later relapsed back into anorexia.⁸ A separate study found that out of 88 people initially diagnosed with anorexia, 32 eventually developed bulimia.⁹

This matters because someone who shifts from anorexia to bulimia might no longer meet criteria for anorexia and could be counted as “recovered” in a study that tracks only anorexia diagnoses. They are not well; their eating disorder has just changed clothes. Good recovery research tries to account for this by tracking all eating-disorder diagnoses over time, not just the original one. But older and simpler studies sometimes miss it, potentially inflating recovery figures.

What Makes Recovery More or Less Likely

Not everyone with anorexia faces the same odds. Several factors consistently appear as predictors in the research, and they paint a picture that makes intuitive sense once you see it.

Depression before the eating disorder develops is one of the strongest negative predictors. In a population-based study, premorbid depressive symptoms were the single most important factor associated with a lower likelihood of recovery after controlling for how long the illness had lasted. Women who had depression before their eating disorder began were far less likely to recover than those without that history.¹⁰ Similarly, in a 22-year study, having major depression at the start predicted still carrying an anorexia diagnosis decades later.¹¹

Self-induced vomiting and high trait anxiety also predicted worse outcomes in longitudinal data. Researchers theorize that for people with high baseline anxiety, the rigid routines and food restriction of anorexia serve an anxiety-reducing function, making the disorder harder to give up because it is doing something the person feels they need.¹² Higher body weight at the start of treatment was linked to better long-term outcomes for the binge-purge subtype, and the same study found that greater impulsivity was actually a positive predictor of recovery, at least early on, possibly because impulsive individuals are less locked into the rigid cognitive patterns that maintain the illness.

Social support adds another dimension. People with eating disorders who perceived higher levels of social support were more likely to be in remission than those with lower perceived support.¹³ Interestingly, it was subjective support, feeling supported, that mattered, not the objective number of people in someone’s life.

The Mortality Risk That Makes Recovery Urgent

Anorexia carries the highest mortality rate of any psychiatric disorder, and that backdrop gives the recovery question real weight. A large meta-analysis of 36 studies covering over 12,000 patients found that people with anorexia were nearly six times more likely to die than the general population during an average follow-up of about 14 years.¹⁴ An updated meta-analysis covering studies through 2024 found a similar figure, with the mortality risk for anorexia roughly five times the expected rate.¹⁵

The risk is not evenly distributed over time. One study found that mortality risk peaked within the first 10 years of follow-up, where the rate was nearly eight times the expected level.¹⁶ Deaths come from medical complications of starvation (cardiac arrest, organ failure, electrolyte imbalances) and from suicide, which accounts for a substantial proportion of deaths in anorexia. These numbers are not meant to be frightening for their own sake; they are meant to explain why the medical community treats anorexia as a life-threatening illness and why early, aggressive treatment matters so much for long-term outcomes.

The Brain Can Heal Faster Than People Think

One of the more hopeful findings in recent years involves what happens to the brain during recovery. Prolonged starvation causes measurable thinning of the brain’s cortex and shrinkage of structures involved in memory, emotion, and reward processing. These changes are real and visible on imaging scans, and they likely contribute to the cognitive rigidity, emotional blunting, and difficulty concentrating that people with anorexia often experience. But the encouraging news is that these changes reverse with weight restoration, often faster than expected.

In one longitudinal study, cortical thickness increased at a rate of about 0.06 millimeters per month during weight restoration therapy. After roughly three months of treatment, cortical thickness had already normalized compared to healthy controls. The same pattern was seen in deeper brain structures. Weight gain itself was the best predictor of this brain recovery, not duration of illness or symptom improvement on questionnaires.¹⁷ Another study confirmed that globally thinned cortex in acutely ill patients normalized following long-term weight restoration.¹⁸

However, age and chronicity complicate the picture. Research tracking patients through treatment found that brain restoration was fastest during the initial phase of weight gain, slowing in later phases. Some regions, particularly parts of the frontal cortex, showed residual thinning even after significant recovery.¹⁹ The implication is that while the brain has remarkable capacity to bounce back from starvation, intervening before the disorder becomes chronic gives it the best chance of full structural restoration.

Do Men Recover at Different Rates?

Anorexia is far more commonly diagnosed in women, and the vast majority of research has focused on female participants. This makes it hard to say anything definitive about whether men recover at different rates. What limited evidence exists suggests the answer is: probably not, at least in terms of the overall likelihood of reaching remission.

A sex-matched prospective study found that remission rates at long-term follow-up were essentially identical between men and women with anorexia, at about 40% for each group.²⁰ Men who recovered tended to score lower on measures of drive for thinness and body dissatisfaction at follow-up, suggesting their residual psychological burden was lighter. A separate matched study found no difference in diagnostic outcome between sexes, though females at follow-up had substantially higher drive for thinness and body dissatisfaction.²¹ A review of the broader literature concluded that there was no firm evidence for gender differences in outcome or mortality, noting that most studies did not even statistically compare genders and those that did generally found nonsignificant differences.²²

Men with anorexia do face unique barriers, though. They are often diagnosed later, partly because the illness is stereotyped as female and partly because the presentation can look different (more focus on muscularity, less on thinness per se). Later diagnosis typically means longer illness duration before treatment, which as discussed earlier is itself a risk factor for worse outcomes. So while the biology of recovery may not differ much by sex, the pathway to treatment often does.

Life After Recovery

A natural question for anyone who has had anorexia, or who loves someone who has, is whether recovery truly means getting your life back. The evidence here is cautiously encouraging. A recent study comparing quality of life across people currently ill with anorexia, people who had restored their weight, and healthy controls found that the weight-restored group scored equivalently to controls on measures of independent living, mental health, happiness, relationships, and self-worth. The currently ill group, by contrast, scored significantly lower on all of those dimensions.²³

Recovery also tracks with lower rates of other psychiatric problems. At 22-year follow-up in one large cohort, people who had recovered from their eating disorder were more than twice as likely to be free of major depression and more than five times as likely to be free of substance use disorders compared to those who had not recovered.²⁴ Whether this means recovery directly protects against depression and substance problems, or whether the same factors that enable recovery also protect against those conditions, is hard to untangle. Either way, the practical implication is the same: people who reach full recovery tend to have meaningfully better mental health overall.

Residual Traces Even After Full Recovery

Recovery does not always mean a perfectly clean slate psychologically. Even in people classified as long-term recovered from adolescent-onset anorexia, researchers have found higher levels of depressive, anxious, and obsessive-compulsive traits compared to matched controls who never had an eating disorder. These differences did not rise to the level of clinical disorders, and the recovered group did not differ from controls on measures of psychiatric diagnoses or overall psychosocial functioning.²⁵ The researchers interpreted these as possible personality traits that predated the eating disorder rather than scars left by it, though the distinction is difficult to prove either way.

This finding actually aligns with the recovery definition research: the “fully recovered” group in one study was indistinguishable from controls on eating disorder-specific measures, but the framework was not designed to capture subtle temperamental differences.² In practical terms, what this means is that someone who is fully recovered from anorexia may still be a person who runs a bit more anxious or perfectionistic than average. That is not the same as being sick. Plenty of people who never had an eating disorder carry those same traits. The difference is that for someone with a history of anorexia, those traits are worth being aware of, because they can become the soil in which a relapse takes root if stress or life disruption hits at the wrong time.