The impact of weight suppression and weight loss speed on baseline clinical characteristics and response to treatment.

Objective: Growing evidence suggests an impact of weight suppression (WS) on severity and course of symptoms in patients with eating disorders (ED), but no study explored also the role of the weight loss speed (WLS) together with WS on the same clinical variables, which is the aim of the present work.

Method: A mixed cross-sectional and longitudinal cohort study was employed. Four hundred and fourteen patients with anorexia nervosa (AN = 208) or bulimia nervosa (BN = 206) according to DSM-5 criteria were recruited and assessed at referral by means of clinical interviews and self-reported questionnaires. Body mass index and diagnostic status were re-evaluated at the end of treatment.

Results: WS was positively correlated with body dissatisfaction in patients with AN (p = .005), but negatively correlated in BN (p = .022). In contrast, WLS was significantly inversely correlated with age and duration of illness in all ED (p < .001), and positively correlated with drive for thinness in BN (p = .007). After treatment, WS at intake predicted higher BMI increase in both AN and BN (p < .03), while higher WLS was significantly associated with a lower drop-out rate in patients with BN (p = .02), and predicted BMI increase only in restricting AN patients (p = .02). In the whole group, WLS significantly predicted remission status (p = .039).

Discussion: In our study, both WS and WLS were associated with baseline "core" clinical variables and provided complementary abilities to predict weight gain and remission at the end of treatment. If replicated, our data suggest the importance of considering both WS and WLS as useful clinical variables in the baseline assessment of ED.