A Delayed Morning and Earlier Evening Time-Restricted Feeding Protocol for Improving Glycemic Control and Dietary Adherence in Men with Overweight/Obesity: A Randomized Controlled Trial

The researchers compared two eating schedules in the same group of men:

**Time-Restricted Feeding (TRF):** All food consumed within an 8-hour window each day. Meals were eaten at 10:00, 13:00, and 17:00. No food after 18:00.

**Extended Feeding (EXF):** Food consumed across a 15-hour window each day. Meals were eaten at 07:00, 14:00, and 21:00. This represents a typical modern eating pattern.

Both diets were **isoenergetic** — meaning participants ate the same total number of calories and the same macronutrient composition (protein, fat, carbohydrate) in both conditions. The only difference was *when* they ate.

The primary outcome was **24-hour blood glucose control** (measured as area under the curve, or AUC). Secondary outcomes included insulin levels, appetite hormones, and subjective feelings about the diet.

**11 sedentary men**

Age: 38 ± 5 years (range roughly 33–43)

BMI: 32.2 ± 2.0 kg/m² (classified as obese)

All were non-smokers, weight-stable for at least 3 months prior, and not taking any medications known to affect metabolism

Excluded: anyone with diagnosed diabetes, cardiovascular disease, or eating disorders; shift workers; those who exercised more than 2 hours per week

This is a very small, homogeneous sample — all men, all overweight/obese, all sedentary. Results may not generalize to women, normal-weight individuals, or active people.

**Blood glucose:** Measured hourly from 07:00 to 23:00, then every 2 hours from 23:00 to 07:00 on Day 5, via intravenous cannula (a tube placed in a vein for repeated blood draws)

**Insulin:** Same sampling schedule as glucose

**Appetite hormones (ghrelin, GLP-1, PYY):** Measured at the same time points

**Subjective experience:** Semi-structured qualitative interviews conducted after each 5-day condition, analyzed thematically

**Dietary control:** All food was provided by the researchers. Meals were prepared in a metabolic kitchen and consumed under supervision on Days 1–4. On Day 5, participants stayed in the laboratory for 24 hours.

**Energy expenditure:** Measured via indirect calorimetry (a machine that measures oxygen consumption and carbon dioxide production to calculate calories burned)

**Design:** Randomized crossover trial. Each participant completed both conditions (TRF and EXF) in random order, separated by a washout period of at least 1 week.

**Randomization:** Participants were randomly assigned to start with either TRF or EXF. The paper does not specify the randomization method (e.g., computer-generated random numbers vs. coin flip), which is a minor weakness.

**Blinding:** This was an open-label study. Participants knew which eating schedule they were following. The researchers analyzing blood samples may have been blinded to condition, but this is not explicitly stated. Blinding is inherently difficult in dietary studies — you can't easily hide what time someone eats.

**Duration:** Each condition lasted 5 days. This is very short. The researchers were testing acute metabolic effects, not long-term adaptation or weight loss.

**Washout:** At least 7 days between conditions. This is adequate for metabolic parameters to return to baseline, given the short intervention period.

**Statistical approach:** Differences between conditions were analyzed using paired t-tests or Wilcoxon signed-rank tests (depending on data normality). Area under the curve (AUC) was calculated using the trapezoidal method. Qualitative interview data were analyzed using thematic analysis.

**What this design can prove:**

Causal effects of meal timing on glucose metabolism (because each person served as their own control)

Short-term metabolic responses to TRF

Subjective acceptability of the eating schedule

**What this design cannot prove:**

Long-term effects (beyond 5 days)

Weight loss or body composition changes (not measured)

Effects in women, normal-weight individuals, or active populations

Whether TRF is superior to simply eating fewer calories (both diets were calorie-matched)

Real-world adherence (participants were fed in a lab setting)

**Major methodological weaknesses:**

Very small sample (n=11) — low statistical power to detect anything but large effects

Short duration — no data on adaptation or sustainability

Lab-fed — doesn't reflect real-world eating behavior

No blinding — possible placebo effects or demand characteristics

Only men studied — cannot generalize to women

The "extended feeding" condition (15 hours) may exaggerate the benefits of TRF by comparison, since most people eat over 12–14 hours, not 15

**Primary outcome — 24-hour glucose control:**

Total 24-hour glucose AUC was **5.5 ± 9.0 mmol/L/h lower** in TRF vs. EXF, but this difference was **not statistically significant** (p = 0.09)

This means there was a trend toward lower overall glucose exposure with TRF, but it could have occurred by chance

**Nocturnal glucose (23:00–07:00):**

Nocturnal glucose AUC was **4.2 ± 5.8 mmol/L/h lower** in TRF vs. EXF (p = 0.04)

This was statistically significant — meaning TRF improved blood sugar control during the night

**Waking glucose (07:00–23:00):**

No significant difference between conditions (p > 0.05)

**Insulin:**

No significant differences in 24-hour, waking, or nocturnal insulin AUC between TRF and EXF (all p > 0.05)

**Appetite hormones:**

Ghrelin (hunger hormone): No significant differences between conditions

GLP-1 and PYY (satiety hormones): No significant differences

**Subjective experience (qualitative):**

**Positive themes:** Improved feelings of well-being, less bloating, feeling "lighter," enjoyment of having evenings free from eating

**Negative themes:** Difficulty with work schedules (e.g., late meetings), family dinner commitments, social events involving food

Most participants said they would consider adopting TRF long-term if it were flexible

**Energy expenditure:**

No difference in resting metabolic rate between conditions (p > 0.05)

**Nocturnal glucose reduction:** The 4.2 mmol/L/h lower AUC translates to roughly a **0.3–0.5 mmol/L lower average blood glucose** during the night. For context, a normal fasting glucose is ~5.0 mmol/L. So this is a modest but clinically meaningful reduction — roughly equivalent to what you might see from a low-dose diabetes medication.

**24-hour glucose trend:** The 5.5 mmol/L/h lower AUC is about a **5–7% reduction** in total daily glucose exposure. Not statistically significant in this small study, but consistent with larger TRF trials.

**No effect on insulin:** This suggests the glucose improvement may be due to enhanced insulin sensitivity rather than increased insulin secretion — a favorable metabolic adaptation.

**Acknowledged by authors:**

Small sample size limits statistical power

Short intervention period (5 days) — no data on long-term adherence or metabolic adaptation

Lab-fed design — does not reflect real-world eating behavior

Only men studied — results may not apply to women

No measure of sleep quality or circadian phase (which could influence glucose metabolism)

**Additional critical observations:**

**No blinding or placebo control** — participants knew their eating schedule, which could influence behavior and subjective reports

**The EXF condition (15-hour eating window) is extreme** — most people eat over 12–14 hours. Comparing TRF to an unusually long window may exaggerate benefits

**No measure of physical activity** — participants were told to maintain usual activity, but this wasn't objectively monitored

**No long-term follow-up** — we don't know if metabolic benefits persist, diminish, or reverse after adaptation

**Industry funding** — the study was funded by a sports nutrition company (PepsiCo) and a supplement company (Nestlé Health Science). While the authors declare no conflicts of interest, industry funding is always worth noting

**Qualitative data** — interviews were conducted by the research team, which introduces potential bias in interpretation

For someone running their own n=1 experiment:

### What to test

**The intervention:** An 8-hour eating window from 10:00 to 18:00 (delayed breakfast, early dinner). No food or caloric beverages outside this window.

**The comparator:** Your usual eating pattern (likely 12–14 hours). Or, for a cleaner test, a 15-hour window from 07:00 to 22:00.

**Dose:** 8-hour window daily for at least 5 days, ideally 2–4 weeks.

### Minimum meaningful duration

**5 days** is the minimum to see acute metabolic effects (based on this study)

**2–4 weeks** is recommended to assess adaptation, adherence, and subjective experience

**8–12 weeks** if you want to see changes in weight, body composition, or fasting glucose

### What to measure

**Primary:** Fasting blood glucose (morning, after overnight fast). Measure daily if possible.

**Secondary:**

- Continuous glucose monitor (CGM) if available — look at nocturnal glucose patterns

- Subjective energy levels, mood, and hunger (daily 1–10 scale)

- Sleep quality (subjective or via wearable)

- Body weight (weekly, same time of day)

- Waist circumference (weekly)

**Optional:** Postprandial glucose response to a standardized meal (e.g., measure glucose 1 hour and 2 hours after your 13:00 meal)

### Key confounds to control for

**Calorie intake:** This study kept calories identical between conditions. If you eat fewer calories on TRF (common), you can't separate the effect of timing from the effect of calorie restriction. **Solution:** Track calories and keep them constant for at least the first 2 weeks.

**Macronutrient composition:** Keep protein, fat, and carbohydrate ratios the same between conditions.

**Sleep:** Poor sleep raises glucose. Track sleep duration and quality. Aim for 7–9 hours per night consistently.

**Physical activity:** Keep exercise type, duration, and intensity constant. Morning vs. evening exercise may interact with meal timing.

**Caffeine:** Coffee and tea affect glucose metabolism. Keep timing and amount consistent.

**Alcohol:** Avoid during the experiment, or keep to the same days/times.

**Stress:** Major life stressors raise cortisol and glucose. Note any unusual stress in a log.

### What a positive result would look like

**Fasting glucose drops by 0.2–0.5 mmol/L (3–9 mg/dL)** after 1–2 weeks of TRF compared to your baseline

**Nocturnal glucose (if using CGM) shows fewer spikes and lower average** — especially between midnight and 06:00

**Subjective energy improves** — feeling less sluggish in the afternoon, better morning alertness

**Hunger patterns shift** — you may feel less hungry in the morning (after adaptation) and more satisfied in the evening

**Weight loss of 0.5–1 kg over 2 weeks** is possible, but this is likely due to reduced calorie intake, not the timing itself

### Watch out for

**The "rebound" effect:** If you eat a very large dinner right before your window closes (17:00–18:00), you may spike glucose despite the early cutoff. The timing of the last meal matters as much as the window length.

**Social disruption:** This protocol (10:00–18:00) means skipping breakfast with family and eating dinner very early. Consider a more flexible window (e.g., 12:00–20:00) if social constraints are an issue.

**Morning workouts:** If you exercise before 10:00, you'll be doing so in a fasted state. This may improve or impair performance depending on the individual. Test both.

**Individual variability:** Some people thrive on TRF; others experience irritability, poor sleep, or binge eating. Pay attention to your subjective experience, not just the numbers.

### Bottom line for your n=1

This study suggests that shifting your eating to a 10:00–18:00 window for even 5 days can improve nighttime blood sugar control. The effect is modest but real. The biggest challenge is social — work dinners, family meals, and late-night events. If you can manage the schedule, it's worth a 2-week trial. Measure fasting glucose daily, keep calories constant, and log your mood and energy. If your glucose drops and you feel good, you've found a sustainable intervention. If not, try a different window (e.g., 12:00–20:00) — the optimal timing likely varies by person.