| Paper authors: | Maren C.G. Kopland, KariAnne Vrabel, Margarita Slof-Op ‘t Landt, Asle Hoffart, Sverre U. Johnson, Erik J. Giltay |
| Year of paper publication: | 2023 |
| Post authors: | Caleb Pozdnikoff, Sarah Kesler, Fidel Vila-Rodriguez |
| Check out the research article: | Kopland et al. (2023) Breaking the cycle: Identifying key symptom pathways of eating disorders and the influence of childhood maltreatment |
Introduction
Eating disorders are complex conditions, often with numerous interacting and reinforcing factors. To help understand these disorders, researchers can use network analysis.
Network analysis can describe what symptoms interact with other symptoms. Most network analysis only includes symptoms at one discrete point in time. The researchers of this study used a new approach, where they administered questionnaires at multiple points in a treatment course and used all of this data in the network analysis. These types of networks are called directed, temporal networks. These networks enable researchers to understand if certain symptoms come before and can predict changes in other symptoms. Using this information can help researchers investigate the mechanisms of change that occur during therapy, allowing them to identify the most central and influential symptoms in the network.
Previous research has found that people with eating disorders who have a history of childhood maltreatment or trauma may react and change differently during treatment when compared to those without. To investigate this, the authors compared the networks of people with and without childhood maltreatment. The authors did this in hopes to understand why people with childhood maltreatment may relapse or drop out of treatment at higher rates than those without. They also wanted to understand if there are different processes ongoing for people with eating disorders depending on if they experienced childhood maltreatment or not.
The researcher’s goal was to investigate what symptoms change and which symptoms drive change during therapy. They did this in the hopes of illuminating why people with childhood maltreatment have higher dropout and relapse during or after treatment.
Methods
The researchers included 122 people with eating disorders that were inpatients (admitted in the hospital for treatment) in their analysis. All 122 people had histories of eating disorders and of non-response to treatment. Everyone was administered the Eating Disorder Examination Questionnaire weekly for 13 weeks to evaluate symptoms. Researchers used network analysis to determine which symptoms came before other symptoms.
To determine if a person had a history of childhood maltreatment, researchers used the Childhood Trauma Questionnaire. If a person reached a score above a specific threshold, they would be considered to have a history of childhood maltreatment. The researchers then compared the networks of participants with childhood maltreatment to networks of participants without.
Results
- The network analysis revealed three clusters of symptoms that changed over the course of therapy:
- Eating disorder behavior (binging, vomiting, etc.).
- Inhibition (which included symptoms like food avoidance, guilt, and exercise among others).
- Thoughts and feelings about body and weight.
- The symptom that tended to come before and predict changes in other symptoms was overvaluation of body shape.
- Other influential symptoms included the wish for an empty stomach, preoccupation with body and weight, and dissatisfaction with one’s figure.
- The most influential symptom differed depending on whether the patient had a history of childhood maltreatment.
- In the network for patients with a history of childhood maltreatment, dissatisfaction with weight was possibly the most influential symptom driving change.
- In the network for patients without childhood maltreatment, overvaluation of shape and the desire for an empty stomach were the most significant.
Conclusion
Targeting feelings and cognitions about shape, specifically the overvaluation of shape, appears crucial for achieving lasting symptom improvement in the overall group. The researchers came to this conclusion because this symptom comes before and predicts changes in other eating disorder symptoms. The study found that during therapy, eating disorder symptoms change dynamically and cluster into three robust groups. Crucially, the specific symptom driving change differed for patients with a history of childhood maltreatment (where dissatisfaction with weight was most influential). This suggests that treatment processes may need to be tailored based on the individual’s trauma history.
For the purposes of the NINET Lab, we hope to examine depression using network analysis in the future.