Why We Need to Stop Treating Anorexia as a Weight Disorder

BMI is not a measure of brain recovery.

Anorexia nervosa is one of the only psychiatric illnesses in which discharge decisions are often tied to a number on a scale. Once a minimum BMI is reached, stabilised bloods, and a restored heart rate are achieved, treatment intensity is frequently reduced.

But if anorexia is a disorder of the brain, why are we measuring recovery primarily by body weight?

Weight restoration is essential. Without it, there is no recovery. But stopping at the lower boundary of “healthy weight” risks mistaking medical stabilisation for neurological recovery.

BMI was never designed to measure brain function.

It was developed as a population-level epidemiological tool. It does not measure cellular energy sufficiency. It does not measure cognitive flexibility. It does not measure relapse risk. It does not measure the degree to which starvation-driven neurobiological changes have reversed.

Reaching a minimum “healthy” BMI does not automatically mean the brain has recovered.

Anorexia Is a Disorder of Energy Availability

At its core, anorexia nervosa involves sustained low energy availability. When energy intake does not meet physiological demand, the brain adapts.

The Minnesota Starvation Study demonstrated that prolonged caloric restriction in healthy men produced obsessional thinking about food, irritability, depression, anxiety, and cognitive changes, all in individuals without prior psychiatric illness (Keys et al., 1950). Starvation alone can generate many of the psychological features seen in anorexia.

Neuroimaging studies in anorexia nervosa show alterations in reward processing, executive function, and cognitive control networks (Kaye et al., 2013). Cognitive rigidity and set-shifting difficulties are well documented (Tchanturia et al., 2012). These deficits are thought to involve highly energy-dependent prefrontal cortical networks.

When glucose availability is inconsistent or insufficient, the brain prioritises survival. Threat sensitivity increases. Flexibility decreases. Behaviour narrows.

Under-fuelling impairs the very cognitive skills required for therapy.

Expecting meaningful cognitive restructuring, exposure work, or emotional processing from a brain operating in sustained deficit may be unrealistic. Nutritional rehabilitation is not an adjunct to therapy; it is foundational.

Weight Restoration vs Functional Recovery

Weight restoration and functional recovery are not synonymous.

Weight restoration typically refers to:

• Reaching a BMI threshold

• Stabilising vital signs

• Normalising key laboratory values

  
  

Functional recovery includes:

• Reduced preoccupation with food

• Improved emotional range

• Spontaneous flexibility around meals

• Decreased compulsive movement

• Restored menstrual function where applicable

• Tolerance of uncertainty

• Reduced rigidity in daily routines

  
  

An individual may technically meet a minimum BMI while still experiencing intense rigidity, high anxiety around food variation, and persistent obsessive thinking.

If someone is “weight restored” but cannot tolerate small deviations in routine, a different spoon, an unplanned snack, or someone else portioning their meal, it raises an important question: has the brain fully recovered, or has the body simply reached a lower boundary?

Relapse rates following weight restoration remain high, often cited between 30–50% within the first year (Berends et al., 2018). One contributing factor may be premature reduction of support when neurological recovery is incomplete.

Cognitive Flexibility as a Functional Marker

Cognitive inflexibility is a well-established feature of anorexia nervosa (Tchanturia et al., 2012). Laboratory tasks such as the Wisconsin Card Sorting Test consistently show set-shifting difficulties.

Importantly, some studies demonstrate partial improvement in cognitive flexibility with weight restoration (Roberts et al., 2010), suggesting that starvation contributes to rigidity beyond trait vulnerability.

In clinical practice, flexibility can be observed behaviourally. Structured, low-stakes deviations from routine can serve as both assessment and intervention. For example:

• Changing cereal brands

• Using a different plate or utensil

• Adding fats without measuring

• Eating at a slightly different time

• Allowing someone else to plate the meal

• Taking a different route home

These are not arbitrary exposures. They probe cognitive flexibility.

If distress is extreme and persistent despite weight gain, further nutritional rehabilitation may be required. If flexibility improves alongside consistent fuelling, this supports the hypothesis that energy restoration is enabling prefrontal recovery.

Cognitive flexibility should not replace weight targets. It complements them. Weight restoration is necessary but not sufficient. Functional brain recovery requires sustained energy availability beyond the minimum threshold.

Starvation Rigidity and Autism-Related Rigidity

Rigidity is not unique to anorexia. Neurodevelopmental conditions, including autism, often involve strong preferences for routine and predictability.

It is important to distinguish between trait-level rigidity and starvation-amplified rigidity.

Trait rigidity tends to remain relatively stable across contexts and energy states. Starvation-induced rigidity intensifies during low energy availability and may soften with sustained refeeding.

Cognitive flexibility exists on a continuum. Autism influences baseline position on that continuum. Starvation shifts anyone further toward inflexibility.

The goal of nutritional rehabilitation is not to erase neurodivergent traits. It is to determine whether under-fuelling is compounding rigidity beyond an individual’s baseline.

Careful observation over time, as energy availability improves, can help differentiate state effects from trait characteristics.

Why Under-Targeting Weight Fuels Relapse

Discharging individuals at the lowest acceptable BMI may prioritise short-term medical stability over long-term neurological recovery.

Energy restoration often requires overshoot beyond pre-illness weight before metabolic, hormonal, and psychological stability is consolidated (El Ghoch et al., 2014). In adolescents, weight suppression and insufficient restoration are associated with poorer outcomes (Leclerc et al., 2013).

The body may require a period of surplus to re-establish metabolic safety. From a neurobiological perspective, this is not excess. It is insurance.

If the brain remains in a semi-starved state, rigidity, anxiety, and food preoccupation may persist, increasing vulnerability to relapse.

What Should We Be Measuring?

A more comprehensive recovery framework might include:

• Sustained weight restoration beyond minimum BMI

• Return of menses where applicable

• Stable energy throughout the day

• Reduced food obsession

• Improved sleep

• Decreased compulsive exercise

• Observable cognitive flexibility

• Ability to tolerate uncertainty and imperfection

  
  

This does not discard BMI. It places BMI in context.

Anorexia nervosa is not primarily a weight disorder. It is a disorder of energy availability affecting the brain.

Weight restoration is essential. But it is not the finish line.

Until we measure recovery by how the brain functions, not just what the scale shows, we risk discharging individuals who are medically stable yet neurologically vulnerable.

If we are serious about improving recovery outcomes in anorexia nervosa, we must move beyond minimum BMI targets and begin measuring what truly matters: brain function, flexibility, and sustained energy availability. If you are a clinician, parent, or someone navigating recovery yourself, consider whether the current weight target reflects neurological recovery or simply medical stabilisation.

To explore this framework in greater depth, including the nutritional neuroscience underpinning the “starved brain” model and practical applications in treatment, visit the Food Mad page here: FOOD MAD . The conversation needs to evolve, and it starts with how we define recovery.

References

Berends, T., Boonstra, N., & van Elburg, A. (2018). Relapse in anorexia nervosa: A systematic review and meta-analysis. Current Opinion in Psychiatry, 31(6), 445–455.

El Ghoch, M., Milanese, C., Calugi, S., et al. (2014). Body composition, eating disorder psychopathology, and psychological status in anorexia nervosa: A longitudinal study. European Eating Disorders Review, 22(3), 188–193.

Kaye, W. H., Wierenga, C. E., Bailer, U. F., Simmons, A. N., & Bischoff-Grethe, A. (2013). Nothing tastes as good as skinny feels: The neurobiology of anorexia nervosa. Trends in Neurosciences, 36(2), 110–120.

Keys, A., Brožek, J., Henschel, A., Mickelsen, O., & Taylor, H. L. (1950). The Biology of Human Starvation. University of Minnesota Press.

Leclerc, A., Turrini, T., Sherwood, K., & Katzman, D. K. (2013). Evaluation of a nutrition rehabilitation protocol in hospitalized adolescents with restrictive eating disorders. Journal of Adolescent Health, 53(5), 585–589.

Roberts, M. E., Tchanturia, K., & Treasure, J. L. (2010). Exploring the neurocognitive signature of poor set-shifting in anorexia and bulimia nervosa. Journal of Psychiatric Research, 44(14), 964–970.

Tchanturia, K., Davies, H., Roberts, M., et al. (2012). Poor cognitive flexibility in eating disorders: Examining the evidence using the Wisconsin Card Sorting Task. PLoS One, 7(1), e28331.