Outcomes of multimodal inpatient treatment for severe eating disorders using clinical and statistical significance

Most practice guidelines suggest that eating disorders (EDs) treatment should be delivered on a continuum of frequency and intensity depending on the patient’s severity of symptoms, medical stability, and treatment history (APA, 2006). Despite residential, multimodal, and multidisciplinary programs are recommended as the first line of treatment for more severe EDs, empirical data on their effectiveness are still limited and evaluated primarily the statistical significance of treatment outcomes at group level (Schlegl et al., 2016). The main aim of this study was to evaluate the long-term effectiveness of an inpatient treatment program for individuals with EDs at both group and individual level applying the Jacobson and Truax’s (1991) criteria for clinical significance. A sample of 69 patients were assessed at intake, at discharge and at 6-month follow-up on the following measures: the Eating Disorder Inventory-3 (Garner, 2004), the Beck Depression Inventory-II (Beck et al., 2006), and the Symptom Checklist-90-Revised (Derogatis, 1994). ED diagnoses were established at intake using the Structured Clinical Interview for DSM-5 (SCID-5; First et al., 2013). Findings showed a statistically significant symptom reduction at both discharge and follow-up, with moderate to large effect sizes, especially on overall eating disorder risk and depressive symptoms severity. 31-42% of patients showed clinically significant changes and 19-29% reliable changes in the various outcome measures at discharge, whereas about 35% remained unchanged. Noteworthy, patients showed higher clinically significant (35-54%) and reliable (35-38%) symptom changes at 6-month follow-up. Patients who showed symptomatic improvement were less impaired at intake. These findings suggest the beneficial effects of intensive and multidisciplinary inpatient treatment setting for EDs, along with the importance of combining clinical significance approach with traditional significance testing (Diedrich et al., 2018).