The aim of this R34 Exploratory Research Grant is to begin to obtain much needed controlled data regarding the treatment of anorexia nervosa (AN). AN is a serious disorder and is associated with substantial morbidity and mortality. Treatment has changed significantly in recent years due to economic pressures that have limited hospital-based treatments and there are no rigorous data regarding the impact of these changes. In fact, there are no evidence-based recommendations regarding the essential components of treatment for this challenging disorder. This application proposes a small randomized trial of two treatments for weight restoration in patients with AN. Extended inpatient weight-restoration treatment will be compared to a """"""""real world"""""""" treatment that includes primarily an intensive outpatient approach. The initial six months of the project will be used to develop and manualize the two treatment approaches. The inpatient treatment will be based on the treatment offered by The New York State Psychiatric Institute, a state-supported research hospital. The """"""""real world"""""""" treatment will be based on the well regarded and successful treatment program at Johns Hopkins Medical Institution. The Hopkins program consists of a shorter inpatient stay followed by a period of day program treatment, reimbursed primarily by health insurance plans. The Hopkins model will be adapted for the purposes of manualization and execution of the treatment at NYSPI. Representatives from two managed care companies will participate in the study treatment development in order to assure that the program is consistent with """"""""real world"""""""" practice and reimbursement. It will likely be a time-limited, sequenced treatment that will be less treatment-intensive than will the hospital program. In the final 2.5 years of the project, a small randomized comparison trial of inpatient weight restoration and a less intensive model of """"""""real world"""""""" treatment will be conducted at The New York State Psychiatric Institute. Follow-up information will be collected over an 18-month period following randomization. Outcome measures will include weight, eating behaviors and associated psychopathology, as well as service utilization during the study period. ? ?