Effect of time restricted eating on body weight and fasting glucose in participants with obesity: results of a randomized, controlled, virtual clinical trial

Time-restricted eating (TRE) is a dietary strategy that consolidates daily food intake into a limited window (typically 6–12 hours), extending the overnight fast. Extending fasting promotes a metabolic switch from glucose to ketone utilization, reinforcing circadian rhythms and potentially reducing oxidative stress and inflammation. Prior human and animal studies suggest TRE can improve cardiometabolic health markers, but many trials are small and vary in design. Missing meals with longer fasts (≥16 h) can reduce adherence. Foods designed to mimic fasting’s metabolic effects (high fat, very low carbohydrate and protein) may help individuals sustain longer fasts while maintaining ketosis. This study examined whether an evening-initiated 14:10 TRE schedule, combined with a structured commercial weight loss program, produces greater reductions in body weight and fasting blood glucose (FBG) than a 12:12 TRE schedule. It also evaluated whether a low-calorie, high-fat, low-carbohydrate, low-protein fasting snack at hour 12 during the 14:10 fast affects blood glucose.

Prior work in TRE and animal time-restricted feeding demonstrates benefits across adiposity, glucose tolerance, insulin sensitivity, lipids, blood pressure, inflammation, appetite regulation, and gut microbiome. While many human TRE trials show weight loss, results are mixed, influenced by differences in populations, fasting windows, meal timing (early vs late), and control of caloric/macronutrient intake. Few, if any, trials have evaluated TRE combined with a comprehensive commercial weight loss program controlling calories/macros with structured coaching and physical activity guidance. Evidence suggests earlier eating windows align better with circadian biology and may enhance weight and glycemic outcomes compared with later windows. Fasting-mimicking or ketogenic approaches can help maintain ketosis and may blunt hunger via effects on appetite hormones (e.g., ghrelin, GLP-1, CCK).

Design and ethics: A randomized, comparator-controlled, virtual clinical trial conducted June–October 2020; approved by Argus IRB and registered at ClinicalTrials.gov (NCT04492930). Written informed consent obtained. All procedures were remote due to COVID-19. Participants: Adults 18–65 years with BMI ≥30 kg/m², body weight <192.8 kg, with a camera-enabled device and internet. Exclusions included weight loss/diabetes medications, serious food allergies, current eating or severe psychiatric disorders or psychiatric medications, special dietary requirements, pregnancy or breastfeeding. Participants enrolled in the Jenny Craig Rapid Results program and purchased 8 weeks of food at 50% discount. Randomization and blinding: Participants (n=78) were randomized 1:1 (by screening day; balanced for sex) to 14:10 TRE (intervention) or 12:12 TRE (active control). Participants were blinded to the nature of intervention vs control. Interventions: Both groups followed the Jenny Craig program with energy intake reduced ~500–1000 kcal/day from expenditure; macronutrient targets ~25–35% fat, 45–55% carbohydrate, 20–30% protein. Participants received three prepackaged meals plus one fruit snack daily, with food curbside pickup or home delivery. Physical activity guidance emphasized walking, progressing to 7000–10,000 steps/day. Weekly one-on-one virtual coaching provided weight loss guidance, troubleshooting, and support. TRE schedules: Fasting began immediately after dinner (target ~7 pm; acceptable 5–8 pm). 14:10 group: 14-h fast ending with breakfast 14 h later; instructed to consume a 200 kcal fasting snack at hour 12 on 5 days/week (mixed nuts ~18 g fat, 5 g protein, 4 g carbohydrate) to maintain ketosis. 12:12 group: 12-h fast ending with breakfast 12 h later; no fasting snack. Procedures and measurements: Participants kept daily logs of first/last meal times. Body weight measured weekly, fasting upon awakening, using a BodyTrace cellular-enabled smart scale with automatic uploads. FBG measured via fingerstick using Abbott FreeStyle Freedom Lite; participants emailed photos showing date/time. For both groups, on two days/week FBG was measured at 12, 14, and 15 h after fast initiation. In 12:12, breakfast followed the 12-h measurement; in 14:10, breakfast followed the 14-h measurement. In 14:10, on one measurement day participants consumed the fasting snack after the 12-h measurement; on the other day they did not. The 14:10 group was queried weekly about the snack’s effects on hunger/satiety and ability to complete 14 h fast. Outcomes: Primary: change from baseline in body weight for 14:10 group. Secondary: change from baseline in FBG for 14:10 group. Additional: changes in weight and FBG in 12:12 group; changes in FBG among participants with baseline FBG ≥100 mg/dl; between-group differences at Week 8. Changes at Week 4 and breath acetone were prespecified but to be reported elsewhere. Sample size: No formal calculation; exploratory pilot to inform future research. Statistical analysis: Analyses focused on completers (provided Week 8 weight and FBG). ITT sensitivity analyses used last observation carried forward. ANOVA models estimated LS mean changes, with fixed effect of TRE schedule and gender as covariate; generalized Satterthwaite approximation (SAS PROC MIXED) accounted for unequal residual variances; REML estimation used. Repeated measures analyses across post-baseline timepoints employed an unstructured covariance selected by AIC. Significance threshold P<0.05. Safety and compliance: Weekly phone/video contacts by study doctor/coaches for support, adherence monitoring, and adverse event assessment. The study doctor was blinded to efficacy measurements.

Participants: 78 randomized (39 per group ITT); 60 completed (30 per group). Dropouts included scheduling conflicts (n=6), medical reasons unrelated to study (n=2), protocol non-compliance (n=4 total), preference for more food options (n=2), and lost to follow-up (n=4 total), distributed across groups as reported. Body weight (completers): - 14:10: LS mean change -10.7 kg (-8.5%); 95% CI for % change -9.6 to -7.4; P<0.001. - 12:12: LS mean change -8.9 kg (-7.1%); 95% CI for % change -8.3 to -5.8; P<0.001. - Between-group difference: 1.9 kg (favoring 14:10) or 1.4% (95% CI 0.2 to 2.7); P<0.05. Repeated measures yielded similar significant differences. Body weight (ITT): Both groups lost significant weight; between-group differences were not statistically significant (kg difference 1.6; P=0.063; % difference 1.2; P=0.094). Fasting blood glucose (FBG) (completers): - 14:10: LS mean change -7.6 mg/dl (95% CI -15.1 to -0.1); P=0.047. - 12:12: LS mean change -3.1 mg/dl (95% CI -10.0 to 3.7); P=0.36. - Between-group difference in change: 4.5 mg/dl (95% CI -3.3 to 12.3); not significant. Repeated measures showed significant overall decreases in each group without significant between-group difference. FBG (ITT): Both groups showed significant within-group reductions only for 14:10 (14:10: -8.0 mg/dl, P=0.037; 12:12: -3.4 mg/dl, P=0.33); between-group difference not significant. Participants with baseline FBG ≥100 mg/dl (completers; n=12 per group): - 14:10: -17.6 mg/dl (95% CI -29.8 to -5.5); P=0.008. - 12:12: -11.2 mg/dl (95% CI -21.0 to -1.4); P=0.029. Fasting snack effects: In 14:10, consuming the 200 kcal high-fat, low-carb, low-protein snack at hour 12 did not raise blood glucose at hour 14, indicating no interruption of fasting glycemia; breakfast after hour 14 increased glucose at hour 15. All 14:10 participants reported decreased hunger and increased satiety at hour 12 with the snack, and most felt it improved their ability to complete the 14-h fast. Safety: No adverse events or safety/tolerability issues were reported.

The study addressed whether extending a daily fasting window from 12 to 14 hours, within a structured commercial weight loss program, enhances weight loss and fasting glycemia in adults with obesity. Both TRE schedules produced statistically significant and clinically meaningful weight loss over 8 weeks, with 14:10 yielding greater reductions in body weight among completers. FBG decreased significantly in 14:10 and numerically in 12:12, with larger improvements among participants with elevated baseline FBG, suggesting TRE may be particularly beneficial for those at higher glycemic risk. The fasting snack maintained fasting glycemia while improving hunger and satiety, potentially facilitating adherence to longer fasting windows without disrupting metabolic fasting. Compared with prior TRE studies that often lacked control over caloric and macronutrient intake or emphasized later-day eating windows, the present trial integrated caloric restriction, macronutrient control, physical activity goals, and intensive coaching, which likely contributed to larger weight losses (7–9% of baseline) than typically observed (1–3%). Aligning eating earlier in the day may better reinforce circadian metabolic rhythms, supporting the observed benefits. While between-group differences in FBG were not significant, consistent within-group improvements and pronounced effects among those with higher baseline FBG underscore TRE’s potential in metabolic risk reduction within comprehensive weight loss programs.

Combining TRE with a structured commercial weight loss program produced significant, clinically meaningful weight loss and improved fasting glycemia within 8 weeks. Extending the fasting window from 12 to 14 hours yielded greater weight reduction and a numerically larger decline in FBG, with particularly notable glycemic improvements among participants with elevated baseline FBG. A low-calorie, high-fat, low-carbohydrate, low-protein fasting snack at hour 12 did not raise blood glucose and appeared to improve hunger and satiety, potentially supporting adherence to longer fasts. Future research should comprise larger, longer-duration trials powered to detect between-group differences, include additional cardiometabolic measures (e.g., A1C, lipids, blood pressure, body composition), and evaluate long-term adherence and durability of effects.

- Virtual study design required self-conducted measurements, potentially increasing variability; some participants missed measurements due to logistical constraints. - No blood draws, blood pressure, or body composition assessments; thus, effects on lipids, A1C, blood pressure, and fat/lean mass were not assessed. - Small, short-duration (8 weeks) pilot not powered to detect between-group differences; sample size calculation was not performed. - Minor baseline differences (e.g., slightly higher baseline weight in 14:10) could have influenced outcomes. - COVID-19-related schedule changes contributed to attrition and may affect generalizability. - Long-term adherence and durability of weight and glycemic effects were not evaluated.