Time‐Restricted Eating on Appetite and Inflammation in Adults With Obesity and Prediabetes: A Randomized Feeding Study Authors: Daisy Duan, Susan Carnell, Scott J Pilla, Stephanie T Chung, Jeanne M Clark, Nisa M Maruthur Published: 2026 ABSTRACT ABSTRACTObjectiveTime‐restricted eating (TRE) has gained attention for weight loss and cardiometabolic health improvement. The Time‐Restricted Intake of Meals (TRIM) study was a 12‐week randomized isocaloric and eucaloric feeding trial comparing 10‐h TRE to a 16‐h usual eating pattern (UEP) in adults with obesity and either prediabetes or diet‐controlled type 2 diabetes. This post hoc analysis investigates TRE's effects on eating behavior, appetite hormones, and inflammatory markers.MethodsParticipants completed the Three‐Factor Eating Questionnaire (Restraint, Disinhibition, Hunger) and provided fasting blood samples for leptin, adiponectin, CRP, cortisol, and sRAGE at pre‐ and post‐intervention. Ghrelin was measured during a 75 g OGTT. Wilcoxon signed rank and Mann–Whitney U tests were used for within‐ and between‐group differences.ResultsThirty‐nine participants (19 TRE; 20 UEP) were analyzed, with a mean (SD) age of 60 (7) years and BMI of 35.2 (6.0) kg/m2. Weight loss was similar across groups (2.7% TRE; 2.5% UEP). Both arms increased restraint and decreased disinhibition and hunger. No statistically significant between‐group differences were observed in eating behavior, appetite hormones, and inflammatory markers.ConclusionsBoth 10‐h TRE and 16‐h UEP promoted improvements in eating behavior in the context of modest and equivalent weight loss. Larger prospective studies are needed to confirm whether clinically meaningful differences exist.Trial Registration ClinicalTrials.gov NCT03527368Keywords: appetite, inflammation, prediabetes, time‐restricted eating Introduction Time‐restricted eating (TRE) has gained attention as an intervention targeting weight loss and improvement of cardiometabolic risk factors [1]. TRE involves restricting daily food intake to a consistent eating window, most commonly spanning 8–10 h. Preclinical studies demonstrate that limiting feeding to active hours, even without caloric restriction, can prevent diet‐induced obesity and reverse metabolic diseases [2, 3, 4]. However, evidence for TRE's efficacy in promoting weight loss in humans remains mixed, which likely stems from differences in duration and timing of eating windows, heterogeneity of study populations, and confounding effects of variable calorie restriction and weight loss across studies [5]. While TRE does not intentionally restrict calories, the inherent caloric reduction associated with a restricted eating window could largely explain any weight loss observed in TRE studies. There is a growing interest in understanding the potential cardiometabolic benefits of TRE that may occur independently of, or disproportionately to weight loss, particularly in populations at high risk for cardiometabolic diseases [6, 7]. Key mediators, such as appetite regulation and inflammation, warrant further investigation as they may play crucial roles in the cardiometabolic effects of TRE. TRE may impact appetite regulation given that hunger and related hormones, such as ghrelin and leptin, exhibit a circadian pattern [8, 9, 10]. Mechanistic studies have found that food intake earlier in the day produced greater postprandial ghrelin suppression, lower appetite score, and higher satiety than meals consumed later in the day [10, 11]. Although some controlled isocaloric feeding studies employing early TRE protocols support these findings [6, 12, 13], results remain inconsistent [14]. Similarly, chronic low‐grade inflammation associated with obesity poses significant risks for insulin resistance and metabolic dysfunction [15]. While weight loss interventions have consistently demonstrated reductions in inflammatory markers among individuals with obesity [16, 17, 18], the effects of TRE on inflammation have shown mixed results. Some studies report reductions in inflammatory mediators, such as TNF‐α and C‐reactive protein (CRP) [19, 20], while others fail to observe significant changes [21, 22]. To address these gaps, we conducted the Time‐Restrict Intake of Meals (TRIM) study, a randomized isocaloric feeding trial designed to compare the effects of a 10‐h early TRE schedule versus a 16‐h usual eating pattern (UEP) on weight and glycemic outcomes in 41 adults with obesity and either prediabetes or diet‐controlled type 2 diabetes [23]. Notably, 10‐h TRE, in the context of controlled isocaloric intake, did not lead to differential weight loss compared to UEP. The purpose of this analysis is to investigate the impacts of isocaloric TRE compared to UEP on appetite symptoms and biomarkers of appetite regulation and inflammation. Methods Study Design The TRIM study was a 1:1 randomized, controlled, parallel‐arm, isocaloric feeding trial that compared 12 weeks of 10‐h TRE to 16‐h UEP in 41 adults with obesity and prediabetes or diet‐controlled type 2 diabetes. The primary outcome was change in weight at 12 weeks. Details regarding the study protocol and results of the parent study have been previously published [23, 24]. The current study is a post hoc analysis of the Three‐Factor Eating Questionnaire (TFEQ) [25] and venous blood samples prospectively collected before and after the intervention. This study was approved by the Johns Hopkins University School of Medicine Institutional Review Board, and all participants provided written informed consent. Participants Inclusion and exclusion criteria of the TRIM main study were previously reported [24]. Major inclusion criteria were age 18–69 years, prediabetes (hemoglobin A1c 5.7% to 6.4%) or diet‐controlled type 2 diabetes (hemoglobin A1c 6.5% to 6.9% without any diabetes medications), and body mass index (BMI) 30–50 kg/m2. Notably, we excluded persons with clinically significant sleep or circadian rhythm disorders, psychiatric/psychologic disorders that would interfere with participation, or active substance use disorders or persons who routinely ate within a 10‐h or narrower window within the past year or followed an intermittent fasting protocol. We also excluded persons on any glucose‐lowering or weight loss medications and those with a history of bariatric surgery. Participants were recruited between May 2018 and September 2019 and completed data collection by December 2019. Intervention The TRIM study diet was developed to be healthy and palatable. The diet was adapted from the Dietary Approaches to Stop Hypertension (DASH) diet, with modifications for lower carbohydrates and higher unsaturated fat, following the approach used in the OmniHeart trial [24, 26]. All meals were provided by the study. Both the TRE and UEP groups received the same macro‐ and micronutrient compositions, consisting of approximately 15%–18% protein, 45%–50% carbohydrate, and 32%–37% fat. Individual caloric targets were established at baseline using the Mifflin‐St. Jeor equation [27], incorporating an activity factor derived from the participant's reported activity levels and responses to the International Physical Activity Questionnaire—Short Form [28]. These caloric targets were held constant throughout the study and were not adjusted to account for changes in body weight over time. Participants were instructed to consume all study‐provided meals and followed fixed caloric distributions throughout the day. As previously described [29], participants in the 10‐h TRE arm were prescribed to eat breakfast (40% of total calories) between 08:00 and 10:00, lunch (40% of total calories) between 10:00 and 13:00, dinner (15% of total calories) between 13:00 and 18:00, and snack (5% of total calories) anytime between 08:00 and 18:00. In contrast, participants in the 16‐h UEP arm were prescribed to eat breakfast (20% of total calories) between 08:00 and 10:00, lunch (25% of total calories) between 10:00 and 13:00, dinner (50% of total calories) between 17:00 and 20:00, and snack (5% of total calories) anytime between 16:00 and midnight. All study foods were prepared in the metabolic kitchen at our community‐based Johns Hopkins ProHealth Clinical Research Unit in Baltimore, Maryland, under the supervision of research dietitians. Both the TRE and UEP groups received the same diet, differing only by eating window. To promote adherence, all participants completed a 1‐week run‐in phase prior to randomization, during which they followed the TRE diet for 4 days and UEP diet for 3 days. This phase ensured participants could tolerate and adhere to meal schedules and study foods. During the intervention, participants ate lunch or dinner on‐site 3 days per week, while the remaining meals, including those for the weekend, were freshly prepared and taken home. Adherence data, previously published in [29], were collected through daily food diaries. Participants recorded the start time, duration, and end time of each meal and snack, along with any uneaten study food, non‐study food intake, and beverage consumption. For on‐site meals, study dietitians directly monitored intake and weighed any leftovers. They also reviewed participants' food diaries during on‐site visits, reinforced meal timing, and addressed any issues related to adherence. Adherence to the intervention was high in both arms of the main TRIM study, without significant differences between groups. Perfect adherence to meal timing, defined as consuming the entire meal within the assigned time window, was achieved on 88% of participant‐days (participants multiplied by number of days in the intervention) in the TRE group and 85% in the UEP group. Similarly, adherence to food consumption, defined as consuming all study food and avoiding any non‐study food, was maintained on 93% of participant‐days in TRE and 95% of participant‐days in UEP [29]. Outcome Measures Participants completed the TFEQ [25] at baseline and during the last week of the 12‐week intervention, after an overnight fast of at least 8 h. The 51‐item TFEQ includes three subscales (factors) capturing different dimensions of eating behavior, cognitive restraint (e.g., I deliberately take small helpings as a means of controlling my weight, 21 items), disinhibition (e.g., Sometimes when I start eating I just can't seem to stop, 16 items), and hunger (e.g., I often feel so hungry that I just have to eat something, 14 items). Item response options include true/false and Likert scales, which are each converted to a 0 or 1 score. Scores for items within each subscale are then summed to generate a total score for each factor. Cronbach's alphas calculated for each subscale are all above 0.70, indicating acceptable reliability. A standardized 75 g oral glucose tolerance test (OGTT) was performed at baseline and at week 12. Participants were instructed to fast for at least 8 h prior to the test. An intravenous catheter was inserted into an antecubital vein, and a fasting blood sample was collected, followed by ingestion of a 75 g glucose load. Venous blood samples were collected at 30, 60, 90, and 120 min after glucose ingestion. Total ghrelin, leptin, and adiponectin were measured by radioimmunoassay (EMD Millipore; Billerica, MA). C‐reactive protein (CRP) and total cortisol were measured by ELISA (ALPCO; Salem, NH). Soluble receptors of advanced glycation end products (sRAGE) were measured by ELISA (R&D; Minneapolis, MN). Incremental area under the curve (iAUC) for total ghrelin was calculated using the trapezoidal method. Data Processing Details of data processing are shown in Figure S1. For this study, our primary analysis included all TRIM participants who completed > 80% of the 51 items on the pre‐ and post‐intervention TFEQ. From the 41 participants in the main TRIM trial (N = 20 UEP; N = 21 TRE), 2 participants in the TRE arm were excluded from analysis due to missing ≥ 20% of TFEQ items. From the remaining 39 participants, 4 participants in the UEP arm and 4 participants in the TRE arm had ≤ 3 questions missing per factor at each time point. For these participants, each factor score was rescaled as follows: If a participant answered 20 out of 21 restraint questions and scored 15 out of 20, his or her score was rescaled as 15*(21/20) = 15.75, which essentially gives an average score based on the questions that the participant did answer. The final TFEQ analytical dataset included 20 UEP and 19 TRE participants, with a similar number of rescaled scores across factors, intervention arms, and time points. For biomarker analysis, an additional TRE participant was excluded due to missing blood sample. The final dataset for biomarker analysis included 20 UEP and 18 TRE participants. Statistical Analysis Baseline variables were compared between intervention arms using Wilcoxon rank sum tests for continuous variables and Fisher's exact tests for categorical variables. For each TFEQ factor score and biomarker, we compared pre‐ and post‐intervention scores using Wilcoxon signed rank tests, stratified by intervention arm. Between treatment arms, we compared within‐individual changes in TFEQ factor scores and biomarkers (i.e., pre/post‐intervention) using Wilcoxon rank sum tests. Effect estimates and 95% confidence intervals (CI) were derived using Hodges‐Lehmann estimator. We also conducted an exploratory descriptive analysis of changes in scores for individual questions within each TFEQ factor. Data analyses were performed in Stata/SE version 16 software (StataCorp LLC, College Station, Texas) and R Studio. Results Baseline characteristics of the TRIM cohort with complete TFEQ data are presented in Table 1. Mean ± SD for age was 60 ± 7 years, and 92% of the participants were women and 92% were Black. Mean ± SD for BMI was 35.2 ± 6.0 kg/m2. Similar to the main TRIM study, participants in the TRE group had lower BMI than those in the UEP group, and the mean ± SD for weight loss at 12 weeks was not different between treatment arms (−2.67% ± 3.18% in TRE and −2.53% ± 1.64% in UEP; p = 0.97). TABLE 1Baseline characteristics.CharacteristicsUsual eating pattern, N = 20Time restricted eating, N = 19 p Age, years59 (7)61 (6)0.48Female sex, n (%)19 (95)17 (90)0.61Race, n (%)1.00Black18 (90)18 (95)White2 (10)1 (5)Employment status, n (%) a 0.77Employed full‐time10 (50)10 (63)Employed part‐time3 (15)2 (12)Retired/other5 (25)4 (25)Disabled2 (10)0 (0)Education, n (%) a 0.52High school2 (10)0 (0)College14 (70)11 (70)Postgraduate4 (20)5 (30)Baseline weight, kg103.6 (17.7)93.5 (11.7)0.06BMI, kg/m2, median (IQR)37.3 (7.1)34.0 (4.2)0.04Waist circumference, cm114.8 (13.6)108.2 (10.0)0.16HbA1c, %5.9 (0.26)5.9 (0.26)0.72 Note: Data presented as mean (SD) unless specified otherwise. Categorical variables compared using Fisher's exact tests. Continuous variables compared using Wilcoxon rank sum tests. a N = 3 missing education status in TRE arm. N = 3 missing employment status in TRE arm. Baseline characteristics. Note: Data presented as mean (SD) unless specified otherwise. Categorical variables compared using Fisher's exact tests. Continuous variables compared using Wilcoxon rank sum tests. N = 3 missing education status in TRE arm. N = 3 missing employment status in TRE arm. Three‐Factor Eating Questionnaire Overall, restraint scores increased, disinhibition scores decreased, and hunger scores decreased after intervention compared to baseline (Table 2). By intervention group, the TRE group had a significant increase in restraint scores and a decrease in disinhibition, while the UEP group had a significant decrease in disinhibition. While changes in scores for each factor were not statistically different between arms, individual changes in scores for both intervention groups presented as waterfall plots in Figure 1 demonstrate interindividual variability. Notably, although median increases in restraint scores were comparable between arms (+1.5 in UEP vs. +2.5 in TRE), the TRE arm demonstrated a wider range of restraint score changes (−4 to +10 points) compared to the UEP arm (−7 to +5 points). Median reductions in disinhibition were also similar (−2.0 in UEP vs. −2.6 in TRE), but the TRE group had a wider range of −13 to +8 points, while the UEP group had a nonsignificant median reduction of 2.0 points with a range of −8 to +2.6 points. Lastly, for hunger, the TRE group had a nonsignificant median reduction of 1.5 points (range −10.5 to +7.5 points), and the UEP group had a nonsignificant median reduction of 0.9 points (range −4.5 to +3 points). Our exploratory descriptive analysis of changes in scores for individual questions within each factor did not reveal any consistent pattern for either treatment group (Table S1A–C). TABLE 2Comparison of Three‐Factor Eating Questionnaire Scores at baseline and week 12, stratified by arm.Eating behaviorUsual eating pattern, N = 20Time restricted eating, N = 19Between‐arm differences (Δ UEP vs. Δ TRE)BaselineWeek 12Median (95% CI), p a BaselineWeek 12Median (95% CI), p a Median (95% CI), p b Restraint (score range 0–21) 10.5 (6.4)12.0 (7.0)1.510.0 (4.0)14.0 (7.0)2.5−1.0(−1.0, 3.0)(0.6, 4.3)(−3.0, 1.0)0.310.010.44 Disinhibition (score range 0–16) 7.2 (7.7)5.3 (4.6) −2.0 (−3.8, −0.5) 0.02 7.9 (7.9)5.3 (6.6)−2.61.1(−5.3, −0.7)(−1.3, 3.2)0.020.53 Hunger (score range 0–14) 4.7 (4.4)4.5 (3.8)−0.96.0 (4.5)4.5 (4.5)−1.50.0(−3.0, 0.8)(−3.0, 1.5)(−1.5, 2.7)0.170.170.68 Note: Baseline and week 12 data presented as median (IQR).Abbreviations: TRE, time‐restricted eating; UEP, usual eating pattern. a Wilcoxon signed rank tests were used to compare baseline versus week 12. Effect estimates and 95% CI were derived from the Hodges‐Lehmann estimator. b Wilcoxon rank sum tests were used to compare within‐individual changes in TFEQ factor scores between treatment arms. Effect estimates and 95% CI were derived from the Hodges‐Lehmann estimator. Comparison of Three‐Factor Eating Questionnaire Scores at baseline and week 12, stratified by arm. Restraint (score range 0–21) Disinhibition (score range 0–16) −2.0 (−3.8, −0.5) 0.02 Hunger (score range 0–14) Note: Baseline and week 12 data presented as median (IQR). Abbreviations: TRE, time‐restricted eating; UEP, usual eating pattern. Wilcoxon signed rank tests were used to compare baseline versus week 12. Effect estimates and 95% CI were derived from the Hodges‐Lehmann estimator. Wilcoxon rank sum tests were used to compare within‐individual changes in TFEQ factor scores between treatment arms. Effect estimates and 95% CI were derived from the Hodges‐Lehmann estimator. Waterfall plots for changes in restraint, disinhibition, and hunger scores in individual participants, stratified by treatment arm. Dashed lines indicate median score change. [Color figure can be viewed at wileyonlinelibrary.com] Appetite Hormones, Adipokines, and Inflammatory Markers Compared to baseline, fasting ghrelin and 2‐h iAUC ghrelin after oral glucose ingestion did not change significantly after the intervention in the overall cohort or in either intervention arm (Figures 2 and 3, Table S2). Leptin levels reduced significantly after the intervention in the overall cohort and in both intervention arms. Adiponectin levels reduced significantly in the overall cohort and in the UEP arm but not in the TRE arm. However, there were no statistically significant between‐arm differences in the changes in leptin or adiponectin levels (Table S2). The inflammatory markers (CRP, cortisol, and sRAGE) were not significantly different before and after the intervention in either treatment arm. Appetite hormones, adipokines, and inflammatory markers at baseline versus week 12. (A) Box plots for fasting total ghrelin stratified by arm. (B) Box plots for leptin stratified by arm. (C) Box plots for adiponectin stratified by arm. (D) Box plots for CRP stratified by arm. (E) Box plots for cortisol stratified by arm. (F) Box plots for sRAGE stratified by arm. Pre‐ and post‐intervention levels were compared using Wilcoxon signed rank tests, stratified by intervention arm; within‐individual changes were compared between intervention arms using Wilcoxon rank sum tests. * indicates p < 0.05. Total ghrelin response after 2‐h 75 g oral glucose tolerance test at baseline and week 12, stratified by treatment arm. Data are presented as mean ± SEM. Wilcoxon signed rank tests were performed to compare baseline versus week‐12 iAUC ghrelin levels stratified by treatment arm (p > 0.05). iAUC, incremental area under the curve; SEM, standard error. [Color figure can be viewed at wileyonlinelibrary.com] Discussion In our 12‐week randomized controlled isocaloric feeding study that compared 10‐h TRE to a conventional 16‐h eating pattern in adults with obesity and prediabetes/diet‐controlled diabetes, we found significant within‐individual increases in dietary restraint and decreases in disinhibition and hunger scores across both eating patterns. The favorable reductions in disinhibition and hunger likely reflect improvements in eating self‐regulation and preoccupation with food and are consistent with the modest weight loss (2%–2.5%) observed in both groups. The increase in cognitive restraint may also reflect improved eating self‐regulation although it should be noted that higher conscious dietary restraint has been associated with disordered eating behavior [30]. Notably, in the absence of caloric restriction and differential weight loss, 10‐h TRE did not elicit distinct changes in eating behaviors, appetite‐regulating hormones, adipokine levels, or inflammatory markers when compared to 16‐h UEP. The improvements in eating behaviors seen in our study are comparable or more pronounced than those reported in the existing literature comparing TRE to other dietary interventions for weight loss. Thomas et al. compared 10‐h TRE plus daily caloric restriction to caloric restriction alone in adults with overweight and obesity. Both intervention groups increased dietary restraint by 4–5 points, though neither demonstrated significant changes in disinhibition or hunger scores after 12 and 39 weeks [31]. Compared to our study, participants in Thomas et al. were much younger, with a mean age of 38 years compared to 60 years in TRIM. Additionally, they exhibited higher baseline disinhibition (~10.0 vs. 8.0 in TRIM) and hunger (~7.0 vs. 6.0 in TRIM) scores, suggesting more disordered eating traits. In our study, we observed a modest yet favorable shift from pre‐ to post‐intervention, with dietary restraint increasing by 2.5 points and disinhibition decreasing by 2.6 points in the TRE arm, reflecting positive changes in an older population that had a relatively low prevalence of adverse eating traits at baseline. Notably, the 2.6‐point improvement in disinhibition score observed in our study is comparable to the changes reported 1 year after bariatric surgery [32] and after 6 weeks of treatment with liraglutide [33]. Moreover, while TRE did not yield statistically more favorable changes in eating behaviors compared to UEP, the waterfall plot reveals greater interindividual variability in change, most notably in disinhibition, potentially reflecting phenotypic differences in eating response to TRE that require further study. A larger sample size could facilitate the identification of predictors for favorable responses to TRE, helping to determine which individuals are most likely to benefit from this dietary approach. In contrast to the eating behavior changes we observed across groups, Lowe et al. found no within‐individual or between‐group differences in reward‐based eating drive (RED) comparing 8‐h TRE (12 p.m. to 8 p.m.) to three structured meals per day (6 a.m. to 10 p.m.) over 12 weeks [34]. The RED scale is a 9‐item questionnaire with similar domains as the TFEQ: lack of control, lack of satiation, and preoccupation with food [35]. Notably, Lowe et al. studied a late variation of TRE (i.e., eating window spanned afternoon to late evening), and 70% of study participants were men, in contrast with our female‐predominant study population. Lastly, Steger et al. conducted a 14‐week trial comparing 8‐h TRE in conjunction with caloric restriction to caloric restriction alone in adults with obesity [36]. Consistent with our study, Steger et al. found that both interventions increased dietary restraint and reduced emotional eating (akin to disinhibition on the TFEQ) and external eating (comparable to hunger on the TFEQ), without between‐group differences. This study used the Dutch Eating Behavior Questionnaire so the magnitude of the change could not be directly compared to the TFEQ. Overall, intraindividual changes in TFEQ scores likely reflect in part the extent to which the intervention diet and eating schedule deviated from the participants' baseline eating habits. In our study, the implementation of controlled isocaloric feeding for both interventions represented a substantial shift from the participants' usual habits, resulting in within‐individual changes in TFEQ scores across both arms. Furthermore, the TRIM study meals provided to both arms likely represented a healthier diet compared to participants' baseline intake, which may have contributed to the improvement in eating behaviors and comparable weight loss observed across both groups. While the TFEQ measures tonic traits, ghrelin levels and subjective hunger and satiety visual analog scale (VAS) scores measure episodic signals that drive periodic food consumption. It has been hypothesized that these episodic signals may be particularly sensitive to changes in meal timing. Indeed, isocaloric feeding studies where food intake is restricted to earlier in the day have reported reductions in subjective hunger and overall desire to eat, especially when assessed at various times throughout the day [6, 12, 13]. These findings suggest that earlier food intake may have a more pronounced effect in attenuating appetite throughout the day. In our study, we did not have complete data collection on hunger VAS scores to reliably assess any within‐ or between‐group changes. This is a notable limitation in our ability to thoroughly evaluate episodic hunger and satiety signals in this study. We observed no significant differences in fasting ghrelin levels or ghrelin suppression responses during the OGTT, either within or between groups. While we noted a significant reduction in morning leptin levels in both groups—likely attributable to weight loss—this change did not differ between arms. Findings regarding appetite hormones in prior TRE studies are mixed. For instance, Sutton et al. reported that a 5‐week intervention of early 6‐h TRE (with dinner before 3 p.m.) led to a reduction in morning fasting peptide YY (PYY) levels but did not affect ghrelin, GLP‐1, leptin, or adiponectin [6]. Conversely, Ravussin et al. found that a 4‐day early 6‐h TRE regimen (8 a.m. to 2 p.m.) significantly reduced both morning and evening ghrelin levels, as well as morning leptin and GLP‐1, while PYY levels remained unchanged [13]. Moreover, other studies with minimal or comparable weight changes between TRE and control groups also indicated that appetite hormones were not significantly altered by TRE [12, 37, 38]. Notably, while our analysis and that of other studies focused on total ghrelin, we did not distinguish between active (i.e., acylated) ghrelin and unacylated ghrelin, which may have distinct and even opposing biological effects on metabolism [39]. We found that inflammatory markers remained unchanged with TRE which is consistent with similar studies conducted in participants with obesity or metabolic syndrome. While some research involving healthy volunteers or resistance‐trained males has reported reductions in TNF‐α after an 8‐h TRE regimen, these studies have not observed significant changes in cortisol levels or CRP [19, 20]. Data from healthy populations cannot be directly compared to our study, which focused on adults with obesity and prediabetes. In contrast, TRE studies involving individuals with obesity or metabolic syndrome have consistently shown no significant differences in CRP levels compared to control groups [6, 21, 22, 40]. This lack of change may reflect the modest degree of weight loss, as highlighted in a review that noted a reduction in CRP typically requires weight loss exceeding 6%—a threshold not commonly reached in TRE studies [41]. Lastly, advanced glycation end products (AGE) and their receptors (RAGE) have been strongly linked to inflammation and the progression of diabetes, with soluble RAGE (sRAGE) believed to mitigate the harmful effects of RAGE [42]. In our study, we observed no changes in sRAGE following TRE, mirroring findings from a study that investigated the weight‐neutral effects of an 8‐h TRE regimen (8 a.m. to 4 p.m.) in patients with prediabetes and obesity [22]. Notably, the immune‐related benefits of TRE may require a more prolonged fasting period than the 10‐h eating window employed in our study [43], potentially limiting our ability to detect effects on inflammatory markers. Our study has several strengths, notably the isocaloric controlled feeding design comparing TRE and a control feeding pattern, which is one of only a few in TRE research. This robust design enabled us to compare changes in eating behaviors, appetite hormones, and inflammatory markers between the TRE group and the control group without the confounding variables of differing caloric intake or weight loss. Such an approach enhances our ability to discern any unique effects associated with shifting meal timing to earlier in the day. However, our study also has several limitations. First, the modest sample size limits statistical power to detect between‐group differences. Given the sample size and post hoc design, only large effect sizes would likely achieve statistical significance. Accordingly, nonsignificant findings should not be interpreted as evidence of equivalence. Second, we only collected fasting samples for leptin, which precluded a comprehensive assessment of 24‐h patterns. While we measured both fasting and post‐OGTT ghrelin levels, a more standardized method for measuring appetite would be a mixed meal tolerance test and measuring ghrelin and other satiety hormones such as GLP‐1 and PYY in the fasting and post‐meal time points. However, as appetite was not a prespecified outcome, our initial study protocol did not include a mixed meal tolerance test. Furthermore, specimen collection and processing did not incorporate the stabilization steps required for reliable GLP‐1 and PYY quantification, which precluded measurement of these hormones. Future studies examining TRE effects on appetite should prospectively incorporate a mixed meal tolerance test with appropriate specimen handling protocols to enable a more complete characterization of the gut hormone response. Third, the TFEQ has limitations in distinguishing between various types of dietary restraint, such as rigid restraint which may be potentially pathological versus flexible restraint which may be more beneficial for weight regulation [44]. Additionally, the TFEQ's binary scoring of item responses may diminish its sensitivity in detecting meaningful changes. Lastly, our study population was composed primarily of older, female, and African American participants, which limits the generalizability of our results. However, our findings fill a critical gap, as these populations are often underrepresented in research. Conclusion In conclusion, in the context of equivalent and modest weight loss observed in both 10‐h TRE and 16‐h UEP groups, we found improvements in eating behaviors in both groups, including increased dietary restraint and decreased disinhibition and hunger scores. No statistically significant differences in eating traits, appetite hormones, or inflammatory markers were detected between the two interventions; however, given the modest sample size and post hoc design, clinically meaningful differences cannot be excluded. Our findings suggest that TRE may offer a pragmatic approach for dietary intervention, although its unique advantages over conventional eating patterns in terms of appetite regulation and inflammatory responses may be limited, particularly in populations experiencing modest weight loss. Importantly, the considerable interindividual variability in how TRE affected eating behaviors underscores the importance of identifying potential predictors of favorable response, which could inform individualized medicine approaches. Larger prospective studies are needed to confirm these findings and to better characterize who is most likely to benefit from TRE. A key area for future research will be examining the role of TRE in weight loss maintenance after initial successful weight reduction, such as after or as an adjunct to effective pharmacotherapy. Funding This study was supported by American Heart Association grant number 17SFRN33590069. D.D. is supported by National Institutes of Health (NIH) grant K23DK133690. S.J.P. receives research funding from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (R03DK144349, R01DK126825, and R01DK139324) and Resolve to Save Lives. The contributions of the NIH author (S.T.C) were made as part of official duties as NIH federal employee, are in compliance with agency policy requirements, and are considered Works of the United States Government. However, the findings and conclusions presented in this paper are those of the author and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services. Conflicts of Interest D.D. serves as a consultant for Rhythm Pharmaceuticals and receives honoraria from the American Diabetes Association. S.C. receives grant funding from Eli Lilly and Company. J.M.C. participates on a Data Safety Monitoring Board for Eli Lilly and Company and serves on the advisory board of Novo Nordisk. N.M.M. receives royalties from UpToDate. The other authors declare no conflicts of interest. Supporting information Data S1: CONSORT 2025 checklist of information to include when reporting a randomized trial. Figure S1: Flowchart of Data Processing. Table S1A. Changes in Individual Restraint Factor Questions. Table S1B. Changes in Individual Disinhibition Factor Questions. Table S1C. Changes in Individual Hunger Factor Questions. Table S2. Comparison of Appetite Hormones, Adipokines, and Inflammatory Markers at Baseline and Week 12, Stratified by Arm.