The Association of Malnutrition, illness duration, and pre-morbid weight status with anxiety and depression symptoms in adolescents and young adults with restrictive eating disorders: a cross-sectional study

BACKGROUND: Restrictive eating disorders (EDs) are often comorbid with anxiety and depression symptoms, placing patients at risk for more severe disease, worse treatment outcomes, and higher rates of mortality. To identify risks for developing such co-morbidities, we assessed the association of malnutrition, ED illness duration, and pre-morbid weight status with symptoms of anxiety and depression in adolescents/young adults (AYAs) with EDs. METHODS: 145 participants with restrictive EDs (anorexia nervosa [AN], other specified feeding and eating disorders [OSFED], avoidant restrictive food intake disorder [ARFID]) were included from the RECOVERY study, a longitudinal web-based registry of AYAs with EDs. We measured malnutrition as percent of expected body mass index (%eBMI), based on participants’ pre-morbid growth trajectory. Outcomes were anxiety and depression scores from the Generalized Anxiety Disorder 7-item (GAD-7) and Center for Epidemiologic Studies Depression (CES-D) scales. We used multiple linear regression to examine the association of malnutrition, ED duration, and pre-morbid weight status with symptoms of anxiety and depression. RESULTS: Mean (SD) age was 16.4(3.0) years; 87% were female; 89% white; 85% had AN, 6% OSFED, 10% ARFID. Of these, 2/3 had ED symptoms ≥1 year, 1/3 had previous higher level of ED care (HLOC), and half were taking psychiatric medications. Mean %eBMI was 90% (range 57–112%). Mean GAD-7 was 9.4(5.9) and CES-D was 24(13.8), indicating most participants had clinically significant anxiety and/or depression. Degree of malnutrition was not significantly associated with anxiety or depression adjusting for age, sex, sexual orientation, ED diagnosis, and use of psychiatric medication. Those with longer duration of ED symptoms had higher depression scores after adjusting for malnutrition, HLOC, length of ED symptoms, and time in our care (p = 0.038). Patients with pre-morbid BMIs ≥75th percentile had lower depression scores than those with pre-morbid BMIs <75th percentile (p = 0.014). CONCLUSIONS: We find high degree of clinically relevant anxiety and depression symptoms in a population of AYAs with EDs. Our findings suggest that factors beyond malnutrition play a role in the co-morbid mood and anxiety disorders in this population. Overall, rapid ED diagnosis and comprehensive treatment for patients with EDs across the weight spectrum—and especially those with psychiatric co-morbidities—will likely aid in recovery.