How Restriction Varies Across People: The Case for Individualized Approaches to Eating Disorders Research and Treatment

By: Molly Robinson, EAT Lab Undergraduate Lab Research Assistant

Eating disorders (EDs) affect millions of people, with 900,000 Kentuckians—including 29,000 children—estimated to be living with one1-2. These disorders involve a problematic relationship with eating, shape, and/or weight, which disrupts daily life and overall well-being. Common EDs include:

While EDs share similarities, each person’s experience is unique, raising concerns about whether standard treatment models work for everyone.
 
Limitations of Standard Treatment: CBT-EThe most widely used treatment for EDs is Cognitive Behavioral Therapy for Eating Disorders (CBT-E)3. Based on the Transdiagnostic Theory of Eating Disorders4, CBT-E assumes that an excessive focus on weight and shape is the core issue underlying all EDs. This concern is thought to drive dietary restraint, bingeing, purging, and other behaviors that maintain the disorder3,4. CBT-E targets these beliefs and behaviors to break the cycle. While it has the strongest research support of any ED treatment, its effectiveness is limited. Only about 50% of patients fully recover following CBT-E or similar first-line treatments5-7. Even in controlled research settings, roughly 45–50% of individuals do not achieve full remission. Relapse rates are high, especially for anorexia nervosa, where 35–50% of patients relapse after treatment5. These statistics highlight a key issue: many people do not respond to CBT-E, suggesting that a single treatment approach may not be enough for all cases.
 
Why CBT-E May Fall ShortThe main limitation of CBT-E is that it was developed for the “average” ED patient, but real patients are highly variable. Two people with the same ED diagnosis can have completely different symptom patterns and drivers8.
 
For example, one person with anorexia might restrict food due to a need for control, while another restricts due to intense fear of weight gain. Some individuals binge eat when stressed, while others restrict food when feeling sad. Body checking behaviors (e.g., frequent weighing, mirror-checking) are another behavioral symptom that may reinforce restriction for one person but be unrelated for another person. If CBT-E assumes weight concerns are the main problem in all EDs, it may miss the true underlying issues driving an individual’s symptoms.
 
Individual Symptom Patterns: Key Findings from The EAT Labs Pilot Study:Research now suggests that eating disorders are not a single disease with one cause but rather a complex network of symptoms that vary between individuals9. While weight and shape concerns are common, the way symptoms connect can differ significantly from person to person8. Some people experience anxiety that leads to food restriction, whereas for others, restriction stems from concerns about body image. Binge eating may be triggered by loneliness in one person, while in another, it is driven by impulsivity.
Traditional approaches may not account for the individual differences that influence how symptoms develop and maintain disordered eating. In response to this challenge, our researchers at The Eating and Anxiety (EAT) Lab are turning to idiographic (person-specific) analysis and symptom-network modeling to look at how eating disorder symptoms play out for an individual person. In a recent pilot study, we followed 26 participantswith various ED diagnoses and tracked their symptoms five times a day for 90 days, resulting in over 9,000 data points. This study gathered data by tracking real-time symptom fluctuations in individuals’ daily life through EMA (Ecological Momentary Assessments) to uncover the unique ways their disorders are maintained. 
Specifically, we were interested in what factors are related to restriction (e.g. cutting calories, omitting food groups). We first estimated a group-level model that included six symptoms relevant to the entire sample: the urge to restrict food intake, anxiety, desire to be thin, feeling fat, placing too much importance on weight and body shape, and fear of gaining weight. Then, we estimated person-specific models that included the 6 symptoms relevant to each person. The output from the group-level model revealed 7 key symptom pathways.

What we found:

The group-level model revealed a potential cycle where drive to thinness leads to over-evaluation of weight and shape, over-evaluation of weight and shape leads to fear of weight gain, and the subsequent fear of weight gain leads to the urge to restrict, extending back into a drive for thinness. 

​The group model did not fit for any individual.

In this figure, the colored points are participants. This figure shows the relationships among different symptoms with the urge to restrict. Points toward the left of the figure are weaker symptom associations, and points on the right are stronger symptom associations. For the urge to body check, we see a big range in the strength of the relationship between this symptom and restriction! Only one person had significant relationships between restriction and feeling overwhelmed by emotions and self-criticism.

•  No two individuals had the same symptom network.
  
• Some participants’ EDs were driven by mood and anxiety symptoms (i.e. anxiety, worry), while others were more influenced by body image concerns (e.g., feeling fat, drive for thinness).
  
• We also found that the urge to restrict was strongly associated with behavioral restriction at the next time point.

Our study revealed that each person’s eating disorder symptoms formed a distinct web of interactions.Within this study, person-specific networks showed that different symptoms were associated with restriction. That means, if treatment was based on an average, it might not be helpful for everyone. Such an approach could lead to poor outcomes, such as prolonged suffering from an eating disorder. For instance, some studies have shown that dietary restriction is closely linked to suicidal thoughts in certain individuals, while in others, it has no such connection. Similarly, body checking behaviors may reinforce food restriction for some, whereas for others, the primary factor maintaining their disorder is related to mood regulation. Pending future clinical trials, it is possible that personalized treatments that target specific symptoms related to the urge to restrict may be effective for preventing the behavior restriction. 

The Future of Eating Disorder TreatmentTo enhance the effectiveness of treatments for EDs, it may be beneficial to adopt a tailored approach that targets the unique network of symptoms each individual experiences. Rather than broadly focusing on concerns about weight and shape for all cases, clinicians could improve outcomes by addressing specific underlying factors. For instance, if anxiety drives a person’s restrictive eating behaviors, interventions aimed at anxiety management might be appropriate. Similarly, treatments could focus on challenging perfectionistic thinking if it sustains the disorder, or on fostering self-compassion if shame plays a central role. It's also valuable for clinicians to understand the varied ways symptoms can interrelate. For example, while anxiety might lead to avoidance in one individual, it could cause another to confront challenging situations. While clinicians may already apply personalized methods in their treatment, this approach offers a data-driven way to make these tough, clinical decisions! This individualized, network-based approach is still being researched, but early findings suggest it is practical and could lead to better treatment success10.

The next steps in ED treatment involve:
 

•  Identifying which symptoms drive each individual’s disorder.
  
• Developing treatment plans that target those key symptoms.
  
• Refining symptom-network-based interventions for broader clinical use.

In summary, EDs are complex and highly individualized conditions. While CBT-E remains an important tool, it is not effective for everyone. New research from our team on individual symptom networks is paving the way for more precise, effective treatments. By understanding what maintains the disorder in each person, clinicians may improve recovery rates and offer a more effective, personalized path to healing. In short, the future of ED treatment may lie in “one-size-fits-one,” rather than “one-size-fits-all.”

You can read a preprint of the paper, which is currently under peer review, here: https://osf.io/preprints/psyarxiv/km3bu_v1
 
References

1. Deloitte Access Economics. (2020). The Social and Economic Cost of Eating Disorders in the United States of America: A Report for the Strategic Training Initiative for the Prevention of Eating Disorders and the Academy for Eating Disorders. https://www.hsph.harvard.edu/striped/report-economic-costs-of-eating-disorders/
2. Ladd, S. (2025, February 25). As diabetes drives up Kentuckians’ use of weight-loss drugs, worries arise about unintended effects. Retrieved March 2, 2025, from https://kentuckylantern.com/2025/02/25/as-diabetes-drives-up-kentuckians-use-of-weight-loss-drugs-worries-arise-about-unintended-effects/
3. Fairburn, C. G. (2008). Cognitive Behavior Therapy and Eating Disorders. Guilford Press.
4. Fairburn, C. G., Cooper, Z., & Shafran, R. (2003). Cognitive behaviour therapy for eating disorders: A “transdiagnostic” theory and treatment. Behaviour Research and Therapy, 41(5), 509–528. https://doi.org/10.1016/s0005-7967(02)00088-8
5. Castellini, G., Lo Sauro, C., Mannucci, E., Ravaldi, C., Rotella, C. M., Faravelli, C., & Ricca, V. (2011). Diagnostic crossover and outcome predictors in eating disorders according to DSM-IV and DSM-V proposed criteria: A 6-year follow-up study. Psychosomatic Medicine, 73(3), 270–279. https://doi.org/10.1097/PSY.0b013e31820a1838
6. Fairburn, C. G., Bailey-Straebler, S., Basden, S., Doll, H. A., Jones, R., Murphy, R., O’Connor, M. E., & Cooper, Z. (2015). A transdiagnostic comparison of enhanced cognitive behaviour therapy (CBT-E) and interpersonal psychotherapy in the treatment of eating disorders. Behaviour Research and Therapy, 70, 64–71. https://doi.org/10.1016/j.brat.2015.04.010
7. Agüera, Z., Riesco, N., Jiménez-Murcia, S., Islam, M. A., Granero, R., Vicente, E., Peñas-Lledó, E., Arcelus, J., Sánchez, I., Menchon, J. M., & Fernández-Aranda, F. (2013). Cognitive behaviour therapy response and dropout rate across purging and nonpurging bulimia nervosa and binge eating disorder: DSM-5 implications. BMC Psychiatry, 13(1), 285. https://doi.org/10.1186/1471-244X-13-285
8. Levinson, C. A., Hunt, R. A., Christian, C., Williams, B. M., Keshishian, A. C., Vanzhula, I. A., & Ralph-Nearman, C. (2022). Longitudinal group and individual networks of eating disorder symptoms in individuals diagnosed with an eating disorder. Journal of Psychopathology and Clinical Science, 131(1), 58–72. https://doi.org/10.1037/abn0000727
9. Levinson, C. A., Cusack, C., Brown, M. L., & Smith, A. R. (2022). A network approach can improve eating disorder conceptualization and treatment. Nature Reviews Psychology, 1–12. https://doi.org/10.1038/s44159-022-00062-y
10. Levinson, C. A., Williams, B. M., Christian, C., Hunt, R. A., Keshishian, A. C., Brosof, L. C., Vanzhula, I. A., Davis, G. G., Brown, M. L., Bridges-Curry, Z., Sandoval-Araujo, L. E., & Ralph-Nearman, C. (2023). Personalizing eating disorder treatment using idiographic models: An open series trial. Journal of Consulting and Clinical Psychology, 91, 14–28. https://doi.org/10.1037/ccp0000785