Latent Trajectories of Body Mass Index and Depressive Symptoms in Inpatients With Anorexia Nervosa: Predictors and Outcomes.
Inpatient treatment aimed at weight restoration and psychiatric stabilization is often required for individuals with anorexia nervosa (AN). This study aimed to identify distinct trajectories of change in body mass index (BMI) and depressive symptoms during inpatient treatment, examine clinical predictors and outcomes, and test reciprocal associations between BMI and depressive symptom changes.
Weekly BMI and depressive symptom data were collected from 156 inpatients with AN (mean treatment duration = 11.6 weeks). Growth mixture modeling identified trajectory classes. Baseline clinical variables were examined as predictors, and discharge outcomes were compared across classes. A joint Bayesian growth model was used to examine longitudinal associations between BMI and depressive symptoms.
Three BMI trajectories were identified: steady increase (58.3%; higher baseline BMI with consistent gains), gradual weight gain (28.9%; low baseline BMI with steady gains), and rapid response (12.8%; low baseline BMI with early rapid gain followed by slowing or decline). Two depressive symptom trajectories emerged: severe-stable and moderate-improving. The severe-stable group was predicted by higher trauma-related comorbidity (OR 11.11, p = 0.009) and eating psychopathology (OR 3.23, p < 0.001). Depressive symptom class was associated with initial BMI but not BMI change, while BMI classes predicted the curvature of depressive symptom trajectories. No credible week-to-week associations emerged between BMI and depressive symptom changes.
Findings highlight substantial heterogeneity in treatment response. Rapid early weight gain in patients with extremely low BMI and high severity may not indicate sustained improvement. Persistent depressive symptoms, especially in those with trauma histories, underscore the need for trauma-informed care alongside weight restoration.
Weekly BMI and depressive symptom data were collected from 156 inpatients with AN (mean treatment duration = 11.6 weeks). Growth mixture modeling identified trajectory classes. Baseline clinical variables were examined as predictors, and discharge outcomes were compared across classes. A joint Bayesian growth model was used to examine longitudinal associations between BMI and depressive symptoms.
Three BMI trajectories were identified: steady increase (58.3%; higher baseline BMI with consistent gains), gradual weight gain (28.9%; low baseline BMI with steady gains), and rapid response (12.8%; low baseline BMI with early rapid gain followed by slowing or decline). Two depressive symptom trajectories emerged: severe-stable and moderate-improving. The severe-stable group was predicted by higher trauma-related comorbidity (OR 11.11, p = 0.009) and eating psychopathology (OR 3.23, p < 0.001). Depressive symptom class was associated with initial BMI but not BMI change, while BMI classes predicted the curvature of depressive symptom trajectories. No credible week-to-week associations emerged between BMI and depressive symptom changes.
Findings highlight substantial heterogeneity in treatment response. Rapid early weight gain in patients with extremely low BMI and high severity may not indicate sustained improvement. Persistent depressive symptoms, especially in those with trauma histories, underscore the need for trauma-informed care alongside weight restoration.
Authors
Martini Martini, Jørgensen Jørgensen, Dani Dani, Christensen Christensen, Sjögren Sjögren, Micali Micali
View on Pubmed