TL;DR: In the 16-week ADAPT trial in adults with overweight or obesity and prediabetes, calorie-restricted high-carbohydrate and reduced-carbohydrate diets produced similar weight loss, but the high-carbohydrate counseling lowered fasting glucose more.

What ADAPT asked about diet sequence

The central question was which combination and order of eating and activity strategies best supports weight loss and cardiometabolic health in adults with overweight or obesity and prediabetes, according to the PMC article titled “Adaptive dietary and exercise strategies for weight loss in Adults with Prediabetes Trial (ADAPT): a sequential multiple assignment randomized trial.”

The investigators focused on three widely used approaches: restricting dietary carbohydrate, limiting the daily eating window, and adding exercise. The background section notes that each approach is commonly used for weight loss and glycemic control, yet the optimal combination and sequence remained unclear.

The trial team enrolled adults with a body mass index of 27 kg/m² or higher who had prediabetes. The study was conducted by researchers in the Department of Family and Community Medicine and the Department of Nutrition Sciences at the University of Alabama at Birmingham, with a collaborator in the School of Medicine at the University of Missouri–Kansas City, according to the PubMed record for the ADAPT trial.

The full report appeared in The American Journal of Clinical Nutrition as volume 122, issue 4, pages 1093–1102 in October 2025, according to that PubMed record. The PMC version of the same report was posted on August 7, 2025. The trial was registered at ClinicalTrials.gov as NCT04745572.

Why researchers used a SMART design

ADAPT used a Sequential Multiple Assignment Randomized Trial, often called a SMART, delivered as a 16-week group-based weight-loss program.

The rationale described in the PMC article is that conventional behavioral programs deliver many diet, exercise, and behavior-change components at once. That bundled approach makes it difficult to identify the contribution of each component and does not efficiently guide when to modify treatment based on early response.

A SMART is designed to develop and evaluate dynamic treatment regimens and to allow evaluation of individual treatment components. The article explains that this structure can help identify who is likely to succeed with fewer or less costly components and reserve more intensive or costly components for people who do not meet early goals.

ADAPT applied that logic in two stages. All participants first received calorie-restricted dietary counseling. Only participants who did not meet an early weight-loss threshold were randomly assigned a second time to an added strategy. That second randomization is the defining feature of the adaptive design.

How the first diet stage worked

In total, 83 adults were initially randomized to either a calorie-restricted reduced-carbohydrate diet, abbreviated RC, or a calorie-restricted high-carbohydrate diet, abbreviated HC, according to the PubMed abstract.

The initial diet assignment continued through the 16-week program. The trial did not test ad libitum eating; both first-stage prescriptions were calorie-restricted weight-loss diets that differed in carbohydrate content.

Response was judged at week 4. Participants who had lost less than 2.5% of starting weight at that assessment were classified as nonresponders. Those participants were then rerandomized to augment the initial diet with either time-restricted eating counseling or exercise counseling.

Of the 82 participants who completed week 4 assessments, 46 were nonresponders, representing 55.4% of that assessment sample. Among those rerandomized, 22 were assigned to add time-restricted eating and 24 were assigned to add exercise counseling.

The assessment sample that reached week 4 averaged 53.8 years of age with a standard deviation of 11.7 years, was 84.3% female, and had a mean body mass index of 38.3 with a standard deviation of 7.2, according to both the PMC article and the PubMed abstract.

What happened to body weight

Weight loss at week 16 was similar between the two initial diet assignments.

The estimated difference between HC and RC was 0.15 kg, with a 95% confidence interval from −1.86 to 2.16 kg and a P value of 0.88, according to the PubMed abstract. In practical terms, the analysis did not detect a weight-loss advantage for either carbohydrate level when calories were restricted.

The conclusions describe the overall group-based program as producing clinically significant weight loss that was similar between the calorie-restricted HC and RC diets. That wording indicates that participants lost a meaningful amount of weight on average, but the carbohydrate prescription did not separate the groups.

Early response mattered more than diet label. The results state that less weight loss was achieved among early nonresponders compared with responders, despite the addition of time-restricted eating or exercise counseling after week 4. In other words, participants who lost at least 2.5% by week 4 finished with larger losses than participants who missed that milestone, even though the latter group received an additional intervention.

What changed for blood sugar and insulin

Although weight change was similar, fasting glucose changed differently between diet groups.

The HC group showed greater improvement in fasting glucose than the RC group, with an estimated difference of −8.0 mg/dL, a 95% confidence interval from −15.29 to −0.67 mg/dL, according to the PubMed abstract. The conclusion restates this finding as counseling to follow an HC diet reduced fasting glucose compared with RC counseling.

Other glycemic and insulin-related measures did not differ significantly between HC and RC. The results list glycated hemoglobin, commonly called A1c, fasting insulin, and quantitative insulin sensitivity check index, known as QUICKI, as outcomes without a difference between the two diet assignments.

Among nonresponders, the choice of second-stage augmentation did not differentially affect any study outcomes. Assignment to time-restricted eating versus exercise counseling after week 4 did not produce different changes in the measured weight or cardiometabolic outcomes.

The trial therefore separated two ideas that are often linked. Carbohydrate level did not alter 16-week weight loss under calorie restriction in this sample, but it was associated with a different fasting glucose response. Adding a second counseling component did not overcome the gap between early responders and nonresponders.

Context from diabetes prevention research

The PMC article places ADAPT in the context of the Diabetes Prevention Program, known as DPP. The DPP lifestyle program is described as a highly effective but complex behavioral package.

That program used 16 individually delivered behavioral counseling sessions over 24 weeks, supervised exercise, individualized maintenance strategies, and a toolbox of incentive strategies. Its activity target included engaging in 150 or more weekly minutes of physical activity. Its dietary advice was characterized by low fat intake at 25% of energy, moderate protein intake at 15% to 20% of energy, and relatively higher carbohydrate intake at 55% to 60% of energy, according to the PMC article.

The article notes that some evidence has suggested carbohydrate reduction may be more effective than a low-fat, high-carbohydrate pattern for weight loss, particularly in people with prediabetes or diabetes. It also notes more recent interest in changing the timing rather than the content of eating.

Time-restricted eating is described as constricting daily food intake to a shortened window, typically 4 to 10 hours per day. The background cites prior research showing modest weight loss and improved glycemic control with that approach, including in people with prediabetes. ADAPT tested whether adding that timing strategy after an initial diet would help people who were not losing weight early.

What nonresponse may require next

The authors conclude that patients with obesity and prediabetes who are unable to achieve early weight loss may require more intensive and costly intervention strategies to improve obesity treatment outcomes.

The examples named are meal provisions and supervised exercise. That statement does not report that those intensive options were tested in ADAPT; it presents them as the type of higher-intensity approach that may be needed when counseling alone does not produce early loss.

The finding points toward stepped care rather than simply adding another counseling message. In ADAPT, nonresponders received additional counseling on either eating timing or exercise, yet their 16-week weight loss remained lower than that of early responders. The authors interpret that pattern as a signal that a different intensity level, rather than another low-intensity addition, may be necessary.

For consumers and clinicians, the practical distinction is between starting strategy and rescue strategy. The starting strategy in this trial could be either calorie-restricted HC or RC counseling for weight loss, with HC showing an added fasting-glucose benefit. The rescue strategy of adding time-restricted eating or exercise counseling did not differentially change outcomes for nonresponders in this 16-week group program.

The registration NCT04745572 provides the formal trial record at ClinicalTrials.gov, with the University of Alabama at Birmingham listed as sponsor and Holly Wyatt listed as the responsible party. The published report lists R. Drew Sayer as corresponding author and lists Katie M. Ellison as first author, with coauthors Aseel El Zein, Navneet Kaur Baidwan, Christine C. Ferguson, Lauren Adele Fowler, David R. Bryan, Chelsi Reynolds, Dalton Hermanson, Kelly J. Berg, Tapan Mehta, James O. Hill, and Holly R. Wyatt.

Sources referenced