Trauma Keeps Surfacing in Food Addiction. A 722-Patient Study Puts Numbers on It

Most addiction intake paperwork is built around substances that arrive in a bottle, a bag or a foil square. It asks how many days in the past month a patient drank, how many times they used opioids, whether they ever injected. It rarely asks how often they ate an entire package of something engineered in a factory to be impossible to stop eating.

A study published October 19 in the Journal of Eating Disorders suggests that omission leaves a distinctly sicker group of patients unidentified. Among 722 people admitted to residential treatment, partial hospital programs and intensive outpatient programs, the 151 who screened positive for ultra-processed food addiction carried trauma-related diagnoses at rates far above the rest of the same clinical population — including a complex PTSD rate nearly double, and a dissociative-subtype rate more than one and a half times higher.

The gap held after the researchers adjusted for age, body mass index at admission, sex at birth, gender and eating disorder diagnosis. It was consistent enough across measures that the authors, led by Timothy D. Brewerton of the Medical University of South Carolina with colleagues at SunCloud Health and the University of Illinois Chicago College of Medicine, describe the finding as evidence that food addiction is not a matter of willpower or poor dietary habits but a condition entangled with severe, frequently overlooked psychological trauma.

What the Study Actually Measured

The setting matters more than the headline number. These were not community survey respondents or college volunteers. Every participant had been admitted to a higher level of psychiatric care for an eating disorder, a substance use disorder, PTSD or a mood disorder, and every one completed the assessment battery within 72 hours of arrival — before treatment had a chance to change what they reported.

Food addiction was defined with the modified Yale Food Addiction Scale 2.0, which takes the eleven DSM-5 substance use disorder criteria and rewrites them for eating: continued consumption despite distress, tolerance, withdrawal, giving up activities, repeated failed attempts to cut back. Trauma exposure came from the Life Events Checklist for DSM-5. PTSD symptoms came from the PTSD Checklist for DSM-5. Complex PTSD came from the International Trauma Questionnaire, which captures the ICD-11 construct. Dissociation came from the Dissociative Subtype of PTSD Scale, built around depersonalization and derealization.

The analysis then compared the two groups with multivariate analyses of variance, holding the baseline characteristics constant.

Two Diagnostic Systems, One Direction

The results split along a fault line that will be familiar to anyone who works across the DSM and the ICD. Under DSM-5 criteria, provisional PTSD was present in 72 percent of patients with ultra-processed food addiction against 51 percent of those without. Under the ICD-11 framework, the same comparison read 40 percent against 25 percent.

Measure	Food addiction (n=151)	No food addiction (n=571)
Provisional PTSD, DSM-5	72%	51%
PTSD, ICD-11	40%	25%
Complex PTSD, ICD-11	37%	20%
Dissociative subtype of PTSD	53%	32%
Major depression	86%	68%
Any eating disorder	78%	24%
BMI above 30	46%	30%

All of those differences were significant at p less than 0.001.

The DSM-5 figure is the looser of the two by construction: it asks about symptom clusters, not functional impairment. The ICD-11 construct is stricter, and complex PTSD is stricter still — it requires the core PTSD picture plus what the ICD calls disturbances in self-organization, meaning chronic difficulty regulating emotion, a persistently negative self-concept and relationships that keep breaking in the same place. That 37 percent of a food-addiction group clears a bar that demanding is the part of the paper worth sitting with.

Inside the Food-Addiction Group

The study did not stop at group averages. Within the 151 patients who screened positive, those who also met criteria for complex PTSD, the dissociative subtype, or both scored significantly higher on every trauma measure the researchers collected: overall PTSD symptoms, depersonalization, derealization, loss of awareness, psychogenic amnesia and the self-organization disturbances. They were also more likely to carry a diagnosed eating disorder.

Twenty-six percent of the food-addiction group met criteria for complex PTSD and the dissociative subtype at the same time. That overlap is not a technicality. A patient who is depersonalizing during a meal-based exposure exercise is not processing the exercise; the treatment can move faster than the patient can stay present for it.

Why the Design Cannot Answer the Question Everyone Will Ask

The finding is an association, and the authors say so plainly. A cross-sectional design cannot establish which came first — whether trauma and dissociation raised the risk of compulsive eating, whether years of addictive eating compounded shame and isolation into a trauma picture, or whether a third factor drives both. Diagnoses were derived from self-report instruments and described as provisional, not from structured clinical interviews. The participants were treatment-seeking and severely ill, which is exactly the population a treatment center wants evidence about and exactly the population that cannot tell you what this looks like in the general population.

What the design does support is the magnitude and the consistency: a dose of difference large enough that it is unlikely to be an artifact of the adjustment variables.

The Comorbidity Stack

Read the rest of the table and a pattern emerges that is bigger than trauma. Major depression hit 86 percent of the food-addiction group against 68 percent of the others. Eating disorders were present in 78 percent against 24 percent. Nearly half — 46 percent — had a body mass index above 30, against 30 percent.

This is the profile of a patient who arrives with several problems at once, and the food addiction flag is tracking severity rather than marking out a separate clinical silo. That has a practical consequence for how programs are organized. Eating disorder units, addiction units and trauma units are typically staffed, licensed and paid for separately, and a patient who belongs in all three tends to get whichever door they walked through first.

For anyone navigating that maze, the practical implication is that dual diagnosis care — treatment built for co-occurring conditions rather than sequential referrals — is the model this data argues for.

What Routine Screening Misses

The authors' central claim is about assessment, not about food. Intake instruments in eating disorder and addiction settings probe substance use, mood and anxiety with reasonable reliability, but rarely ask systematically about complex PTSD or dissociative symptoms. The consequence is structural: the patients with the most layered trauma histories are the ones most likely to be treated for the surface behavior alone.

That matters because dissociation interferes with treatment in a specific way. Trauma-focused work and eating disorder protocols both depend on the patient being present for the session. Depersonalization and derealization reduce exactly that presence, and a program that has not measured them will read the resulting non-response as resistance or lack of motivation. The study's authors argue for assessment across the full spectrum of trauma-related symptoms and for integrated, individualized treatment that addresses trauma sequelae directly rather than focusing narrowly on eating behavior.

The Construct That Is Still Being Argued

Ultra-processed food addiction is not a DSM-5 diagnosis, and the researchers are careful not to present it as one. The modified Yale scale operationalizes substance use criteria for eating, and the field remains divided over whether the resulting pattern reflects a genuine addictive process, a behavioral phenotype of binge eating, or a marker of dietary environment. Food is not a substance anyone can abstain from, which is why the abstinence framing that works for alcohol does not transfer cleanly.

The paper's contribution is not that it settles that argument. It is that it shows what the argument looks like inside a real treatment population: a subgroup with substantially more trauma, more dissociation, more depression, more disordered eating and higher body weight, whose care is currently organized around the food and not around the trauma underneath it. For families trying to understand why a residential stay for eating disorder treatment so often ends in relapse, that is a useful lens — and a reason to ask, at intake, whether anyone has screened for what happened before the eating started.