But observable symptoms might be misleading.
Treatment guidelines for AN and BN routinely state that evaluation of a person’s characteristics should be used to predict responses to different treatment approaches. However, the best predictions remain unclear. For example, ED-specific characteristics, such as those given in the DSM-5, have not performed very well as outcome predictors.
Psychologists Dr. Laura Muzi and colleagues at Sapienza University of Rome, Italy, recently reported on the role of personality features in a group of 84 female patients with AN and BN treated in a residential program. They reported that personality factors had the potential to target relevant individual factors and to indicate effective treatment strategies (Eat Weight Disord. 2021. 26:1195). The work was based in part on the idea of pathoplasticity. That is, factors such as disordered eating or personality features (in this case) can influence the course of each other over time.
The participants and the program
Of the 84 participants, 38 were diagnosed with AN, restricting subtype, with an average body mass index, or BMI, of 15.16 kg/m2, while 14 others diagnosed with AN, purging subtype, had an average baseline BMI of 16.82 kg/m2. The remaining 32 had diagnoses of BN; this group had an average baseline BMI of 22.85 kg/m2. The mean age of participants was 16 years.
Once admitted to the program, participants became involved in a full-time, non-hospital-based and multidisciplinary residential treatment program that used a predominantly psychodynamic approach. The average length of treatment was long relative to that provided in many settings, 5.5 months. All medical professionals, including psychiatrists, psychologists, social workers, nutritionists, and nurses, met weekly to discuss individual cases. Patients were offered 24-hr supervision so the therapists might interrupt any repetitive and pervasive ED behaviors. In addition, each patient had individual psychotherapy once or twice a week. Other activities included nutritional counseling and rehabilitation, meal support, interventions focused on affective and emotional experiences, skills training, recreational and art therapy, and social cooking classes.
The researchers used a number of questionnaires to test the patients, including the Clinical Diagnostic Interview (CDI), the Shedler-Westen Assessment Procedure-200 (SWAP-200), the Eating Disorder Inventory-3 (EDI-3), and the Outcome Questionnaire-45.2 (OQ-45.2). The OQ-45.2 measures function and symptomatic impairment, such as interpersonal problems and social roles. The SWAP-200 is a clinician rating of observed personality features.
Higher EDI-3 symptoms could be linked to dietary restrictions
Higher EDI-3 symptom scores at discharge were positively linked to the number of restriction episodes per week and are negatively related to length of treatment. When Dr. Muzi and her colleagues analyzed the SWAP-200 Personality Disorder (PD) scales, they found positive associations between higher ED symptomatic impairment and avoidant PD scales. Associations with higher PD impairment were also noted for paranoid, schizoid, histrionic, avoidant, and other symptoms. Higher healthy personality function was related to lower ED symptom scores, as well as dysphoric and obsessive: depressive high-functioning Q-factors.
Nearly 40% of ED patients showed clinically significant improvement after the treatment, and an additional 8.3% showed reliable symptomatic change even though they remained in the dysfunctional population. On the other hand, it is also worth noting that 46.4% of the women showed no significant improvement, and 6% had deteriorating symptoms.
To the best of the authors’ knowledge, their study was the first to explore the relevance of a broad spectrum of personality traits and styles to determine therapeutic outcomes in people with AN and BN. Overall, the findings did support the authors’ hypotheses. Elements of a person’s personality disorder significantly helped predict therapy outcomes at both the group and individual levels.
The promise of future studies
Some clinical variables emerged as significant predictors. The authors suggest that future studies analyze the relationship between such variables and personality disorders in greater depth. For example, a larger number of dietary restriction episodes per week predicted a worse therapeutic outcome, as did higher rates of compensatory behaviors per week.
The authors also pointed out limitations of the study. First, the moderate sample size of exclusively white/Caucasian women limited the ability to generalize findings to males, minority populations, and other populations with EDs other than AN or BN. Data also came from a single treatment center and only included patients who were discharged after completing treatment.
Nonetheless, as the researchers write, “First and foremost our findings suggest that if clinicians want to understand and treat ED symptoms effectively, they have to know something about the person who hosts them.” Concentrating on observable symptoms may lead clinicians to neglect less overt and less easily measured aspects of patients’ other subjective experiences, such as loneliness and feelings of shame. Thus, applying the principal of pathoplasticity may help point the way to more effective treatment guidelines and strategies.