OBJECTIVE To investigate whether, at study entry, (a) weight suppression (WS), the difference between highest past adult weight and current weight, prospectively predicts time to first full remission from bulimia nervosa (BN) over a follow-up period of 8 years, and (b) weight change over time mediates the relationship between WS and time to first full remission. METHODS A well-characterized sample of women with BN (N = 110; M age = 25.58 years, SD = 6.48) from the Massachusetts General Hospital Longitudinal Study of Eating Disorders was interviewed at 6-12 month intervals over 8 years. The main outcome measure, a "time to first full remission" variable, was based on psychiatric status ratings generated from the Eating Disorders Longitudinal Interval Follow-up Evaluation. RESULTS WS was significantly associated with time to first full remission (p = .01; hazard ratio = .89; 95% confidence interval [0.82, 0.97]), indicating that women who were more weight suppressed at study entry took longer to recover. Weight change did not mediate the relationship between WS and time to remission. CONCLUSIONS Results add to a growing body of evidence that WS predicts maintenance of BN symptoms and extend previous short-term findings by demonstrating, over a period of approximately 8 years, that WS predicts longer time to first full remission. Beyond absolute weight status, WS level may significantly inform the treatment of BN.