# Using a SMART Design to Examine Pharmacological and Behavioral Treatments to Treat Loss-of-Control Eating and Improve Weight Outcomes after Metabolic and Bariatric Surgery

> **NIH NIH R01** · YALE UNIVERSITY · 2021 · $708,218

## Abstract

Although metabolic and bariatric surgery (MBS) is the most effective treatment for severe obesity and results in
robust average improvements in weight and medical comorbidities, weight outcomes are markedly variable
after MBS and weight regain with reoccurrence of medical comorbidities is common, suggesting that MBS
alone is often not enough. To date, only one reliable modifiable predictor of post-MBS weight loss has
emerged - “loss-of-control eating” (LOC-eating), the core feature of binge eating. LOC eating, which occurs in
30% of patients following MBS, is associated prospectively with poorer weight-loss and functioning outcomes.
 There exists little guidance as to what treatments should be provided to enhance MBS outcomes in
general or to enhance outcomes among the high-risk subgroup with LOC eating. Emerging research suggests
that behavioral treatments for LOC-eating adapted from obesity and binge-eating disorder (BED) literatures
might have utility. No randomized controlled trials (RCTs) have examined effectiveness of any pharmacologic
agents for either LOC-eating or to enhance weight loss following MBS. This study aims to perform a two-stage
RCT to test the effectiveness of treatments for LOC-eating and improving weight outcomes following MBS.
 In Stage 1 RCT, N=160 patients with obesity and LOC-eating following MBS will be randomly assigned
(double-blind) in a balanced factorial (2 X 2) design, to one of four 16-week conditions to test BWL and
pharmacotherapy with naltrexone/bupropion (NB; FDA-approved for weight-loss): BWL+NB, BWL+Placebo,
NB, or Placebo. In Stage 2 RCT, “Responders” to Stage 1 treatments (defined as less than once weekly LOC
eating) will be randomized (double-blind) in equal proportions (stratified blocked randomization with first
treatment as stratifying variable) to NB or placebo for 12 weeks. “Non-responders” to Stage 1 treatments will
be randomized (double-blind) to an alternative (distinct) medication, lisdexamfetamine (LDX; recently FDA-
approved for BED) or to placebo for 12 weeks. Independent comprehensive (blinded) assessments of LOC
eating and obesity and their associated outcomes will occur monthly during treatments and then at 6- and 12-
month follow-ups after completing all treatments (i.e., 19 months after randomization).
 The first stage RCT will provide new findings regarding effectiveness of BWL and of NB weight-loss
medication and whether this specific combination of behavioral and pharmacologic treatments is effective for
patients who have LOC-eating and obesity following MBS. The second stage RCT will provide new and novel
findings from a controlled test, amongst Responders to Stage 1 treatments, whether NB medication results in
superior maintenance and longer-term outcomes than placebo. The second stage RCT will also explore, in
double-blind fashion, amongst Non-responders to Stage 1 treatments, whether switching to an alternative
(distinct) LDX medication enhances outcomes.

## Key facts

- **NIH application ID:** 10267187
- **Project number:** 5R01DK126637-02
- **Recipient organization:** YALE UNIVERSITY
- **Principal Investigator:** CARLOS M GRILO
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2021
- **Award amount:** $708,218
- **Award type:** 5
- **Project period:** 2020-09-22 → 2025-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10267187

## Citation

> US National Institutes of Health, RePORTER application 10267187, Using a SMART Design to Examine Pharmacological and Behavioral Treatments to Treat Loss-of-Control Eating and Improve Weight Outcomes after Metabolic and Bariatric Surgery (5R01DK126637-02). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/10267187. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*