Heterogeneous Pathways of Readiness to Change in Anorexia Nervosa: A Longitudinal Trajectory Analysis

Purpose: Readiness to change constitutes a key determinant of treatment outcome in anorexia nervosa (AN), yet its developmental patterns during inpatient care remain poorly understood. This study aimed to characterize longitudinal readiness trajectories in AN inpatients and explore associations with psychiatric comorbidity. Methods: We conducted a prospective longitudinal study including 97 female inpatients with AN (mean age 26.7 ± 9.4 years) and 29 age-matched healthy controls. We assessed readiness to change at three optimal timepoints: baseline (1st admission), 8 weeks into treatment, and readmission baseline (2nd admission). We applied K-means cluster analysis with silhouette validation to identify distinct readiness trajectories. Complete case analysis (n=24) and multiple imputation sensitivity analyses were performed. Comprehensive statistical analyses included confidence intervals and effect sizes for all major comparisons. Results: K-means cluster analysis with optimal silhouette score (0.427) identified two distinct readiness trajectories. Cluster 1 (Stable-High, n=17, 70.8%) maintained consistently moderate-to-high readiness, while Cluster 2 (Declining, n=7, 29.2%) showed substantial decline. Between-cluster differences were significant at 8 weeks (Cohen's d = 3.34, p < 0.001) and readmission (Cohen's d = 1.67, p = 0.001). Patients with AN demonstrated significantly lower baseline readiness than healthy controls (37.2 ± 13.6 vs 13.4 ± 7.8, Cohen's d = 1.86, p < 0.001). Psychiatric comorbidities showed no significant associations with baseline readiness or trajectory membership. Conclusions: Readiness to change in AN inpatients follows two distinct trajectories, with approximately 30% experiencing severe motivational decline during treatment. The substantial difference from healthy controls confirms readiness as a core treatment target. These findings support trajectory-specific interventions to optimize treatment engagement. Level of evidence: Level III: Prospective observational cohort study with statistical modeling