This randomized controlled trial (RCT) tested effectiveness of adaptive SMART stepped-care treatment to "standard" behavioral weight loss (BWL [standard]) for patients with binge-eating disorder (BED) and obesity. One hundred ninety-one patients were randomly assigned to 6 months of BWL (standard; n = 39) or stepped care (n = 152). Within stepped care, patients started with BWL for 1 month; treatment responders continued BWL, whereas nonresponders switched to cognitive-behavioral therapy (CBT), and patients receiving stepped care were additionally randomized to weight-loss medication or placebo (double-blind) for the remaining 5 months. Independent assessments were performed reliably at baseline, throughout treatment, and posttreatment. Intent-to-treat (ITT) analyses of remission rates (zero binges/month) revealed that BWL (standard) and stepped care did not differ (74.4% vs. 66.5%); within stepped care, remission rates ranged 40.0% to 83.3%, with medication significantly superior to placebo (overall) and among nonresponders switched to CBT. Mixed-models analyses of binge-eating frequency revealed significant time effects, but BWL (standard) and stepped care did not differ; within stepped care, medication was significantly superior to placebo and among nonresponders switched to CBT. Mixed models revealed significant weight loss, but BWL (standard; 5.1% weight-loss) and stepped care (5.8% weight-loss) did not differ; within stepped care (range = 0.4% to 8.8% weight-loss), medication was significantly superior to placebo and among both responders continued on BWL and nonresponders switched to CBT. In summary, BWL (standard) and adaptive stepped-care treatments produced robust improvements in binge eating and weight loss in patients with BED/obesity. Within adaptive stepped care, weight-loss medication enhanced outcomes for BED/obesity. Implications for clinical practice and future adaptive designs are offered. (PsycINFO Database Record (c) 2020 APA, all rights reserved).