Time-restricted eating, what the controlled trials actually show for weight, metabolism, and adherence

Key takeaways

• TREAT (Lowe et al. JAMA Intern Med 2020) randomized 116 adults with overweight to 16:8 TRE or three-meal control for 12 weeks. Weight loss was modest in both arms (about 0.94 kg vs. 0.68 kg) and the difference was not statistically significant. Fat-free mass loss was higher in the TRE arm.
• Liu et al. NEJM 2022 randomized 139 adults with obesity to TRE plus calorie restriction or calorie restriction alone for 12 months. Weight loss was equivalent (8.0 kg vs. 6.3 kg, not statistically different). The trial concluded that TRE did not provide additional benefit beyond calorie restriction.
• Sutton et al. 2018 (early-TRE, 8 AM to 2 PM) showed metabolic improvements (insulin sensitivity, blood pressure, oxidative stress) independent of weight change, suggesting that the timing of the eating window may matter.
• Jamshed et al. JAMA Intern Med 2022 compared early-TRE (eating before 1 PM) with a control eating window in adults with obesity and reported additional weight loss with early-TRE at matched caloric prescriptions.
• TRE adherence is a meaningful concern. The TREAT trial reported 80 percent adherence over 12 weeks, while longer trials show adherence drift. TRE may also produce inadvertent caloric restriction primarily in adults who do not consciously track intake.

What TRE is, and isn't

Time-restricted eating limits the daily eating window without explicitly limiting calories. Common patterns include 16:8 (eat for 8 hours, fast for 16), 18:6, and the more restrictive 20:4. TRE is one form of intermittent fasting; the other major form is alternate-day fasting (ADF) and the 5:2 pattern (5 days normal eating, 2 non-consecutive days at 500 to 600 kcal). The trial literature is largest for TRE because the protocol is the most behaviorally tractable.

TRE is sometimes presented as a metabolically privileged intervention that produces benefits beyond what caloric restriction alone would explain. The published controlled trials are more equivocal. The strongest signal is in adults who experience inadvertent caloric restriction during TRE because compressing the window leads to fewer eating occasions; the second is in early-window TRE patterns (eating earlier in the day) where circadian alignment may matter.

TREAT (Lowe 2020) and Liu 2022

TREAT (Time Restricted Eating on Weight Loss) randomized 116 adults with overweight or obesity to 16:8 TRE (eating window noon to 8 PM) or a three-meal control for 12 weeks. Weight loss was modest in both arms with no statistically significant difference (TRE 0.94 kg, control 0.68 kg). The TRE arm showed numerically more fat-free mass loss .

Liu et al. NEJM 2022 randomized 139 adults with obesity to either calorie restriction (1500 to 1800 kcal/day for men, 1200 to 1500 for women) plus 16:8 TRE, or to calorie restriction alone, for 12 months. Both arms lost weight: 8.0 kg in the TRE arm vs. 6.3 kg in the calorie-restriction arm, not statistically different. Body composition, lipids, and glycemic markers improved in both arms with similar magnitude. The trial concluded that TRE did not provide additional weight or metabolic benefit beyond calorie restriction .

Early-window TRE and the circadian question

Sutton et al. Cell Metabolism 2018 conducted a 5-week crossover trial in 8 men with prediabetes, with eating windows of 8 AM to 2 PM (early TRE) vs. 8 AM to 8 PM (control). Weight change was matched between arms (the trial fed all subjects to weight maintenance). Insulin sensitivity, beta cell responsiveness, blood pressure, and oxidative stress all improved in the early-TRE arm independent of weight change .

The Sutton trial is small, short, and male, but it is the cleanest demonstration that a TRE pattern can produce metabolic improvements independent of caloric or weight change. Subsequent trials (Jamshed et al. 2022 in adults with obesity) reported similar advantages of early-TRE over late-TRE for body weight and metabolic markers when caloric intake was matched .

Lean mass and adherence

TRE without an explicit lean-mass-protective stimulus tends to produce greater fat-free-mass loss than caloric restriction alone, particularly when the protein-distribution literature on per-meal protein doses (0.3 to 0.4 g/kg per meal, three to four meals per day) is applied to a compressed window with fewer eating occasions. The Lowe TREAT data support this concern; subsequent trials with explicit protein and resistance-training arms have moderated the effect .

Adherence to TRE patterns drops over months. Cienfuegos et al. 2020 reported that adults randomized to 4-hour and 6-hour eating windows lost weight at similar rates over 8 weeks, but with progressive dropouts in the 4-hour arm. Long-term sustained TRE outside trial settings is less common than the short-term study literature would suggest .

TRE in the GLP-1 or structured-training context

TRE compresses eating; GLP-1 agonists suppress eating. The combination has not been studied head-to-head in controlled trials of meaningful size. Mechanistically, layering TRE on a GLP-1 deficit risks compressing protein and resistance-training-relevant nutrient timing into a window that may be too narrow to support the per-meal protein doses associated with lean-mass preservation. The published evidence does not yet describe this interaction quantitatively.

TRE in active resistance-training contexts is more characterized. Tinsley et al. 2017 randomized resistance-trained men to 8-hour TRE (1 PM to 8 PM) or no time restriction during 8 weeks of training. Strength and lean-mass outcomes were similar between arms; the TRE arm showed slightly more fat loss. The implication is that TRE is compatible with resistance training when total daily protein and caloric intake are matched .

What the evidence does not show

• TRE has not been shown to produce weight loss or metabolic improvement beyond what caloric restriction alone produces in controlled trials lasting 12 months.
• TRE does not consistently improve insulin sensitivity or cardiometabolic markers independent of weight loss when the eating window is mid-day or late.
• TRE has not been demonstrated to extend lifespan or healthspan in any human trial. The lifespan-extension claims trace to model organisms (rodents, fruit flies) under feed-restriction protocols, not to controlled human longevity trials.
• Extreme TRE (4-hour windows or shorter) is not better than standard 16:8 in published controlled trials and may impair protein distribution and resistance-training adaptation.