The causal role of breakfast in energy balance and health: a randomized controlled trial in lean adults

This study investigated the direct, causal effects of eating breakfast versus skipping it on various aspects of energy balance and health markers in healthy, lean adults living their normal lives.

The intervention was a daily breakfast, defined as consuming at least 700 calories before 11:00 AM. This was a substantial breakfast, designed to represent a typical, hearty morning meal.

The comparator was extended fasting, meaning participants consumed zero calories until 12:00 PM (noon) each day. This effectively meant skipping breakfast and delaying the first meal of the day.

The outcome measures included:

• Components of energy balance:
  • Resting Metabolic Rate (RMR): The number of calories the body burns at rest.
  • Physical Activity Thermogenesis (PAT): The calories burned through physical activity.
  • Energy Intake: The total calories consumed from food and drink.
• Body composition: Body mass (weight) and adiposity (body fat percentage).
• Metabolic health markers:
  • 24-hour glycemic responses: How blood sugar levels change throughout the day.
  • Adipose tissue glucose uptake: How well fat cells absorb glucose from the blood.
  • Systemic indexes of cardiovascular health: General markers related to heart and blood vessel health (e.g., from blood tests).

The study aimed to determine if the widely held belief that "breakfast is the most important meal of the day" has a causal basis, particularly regarding energy expenditure, calorie consumption, and metabolic health.

The study included a total of 33 lean adults from southwest England.

• 21 women with a dual-energy X-ray absorptiometry (DEXA)-derived fat mass index of ≤11 kg/m².
• 12 men with a DEXA-derived fat mass index of ≤7.5 kg/m².
• Participants were aged between 21 and 60 years.
• The key characteristic was that all participants were lean, meaning they had a relatively low body fat percentage, as indicated by their fat mass index. This is an important detail, as the results might differ in individuals with higher body fat or different metabolic profiles.
• The study was conducted under free-living conditions, meaning participants continued their normal daily routines outside of the specific dietary intervention.

The researchers used a combination of advanced laboratory techniques and real-world monitoring to capture a comprehensive picture of energy balance and health.

• Body Composition (Fat Mass Index): Dual-energy X-ray absorptiometry (DEXA) was used to precisely measure body fat mass and calculate the fat mass index (FMI). DEXA scans are highly accurate for assessing body composition, providing detailed information on bone mineral density, lean mass, and fat mass. This was crucial for ensuring participants met the "lean" criteria.
• Resting Metabolic Rate (RMR): This was likely measured using indirect calorimetry, a method where participants breathe into a device that measures oxygen consumption and carbon dioxide production. These gases are used to calculate the number of calories burned at rest. This is a gold standard for RMR measurement.
• Physical Activity Thermogenesis (PAT): While the abstract doesn't specify the exact instrument, in free-living studies, PAT is often estimated using accelerometers (wearable devices that track movement) or doubly labeled water (a highly accurate but expensive method that tracks hydrogen and oxygen isotopes in body water to calculate total energy expenditure, from which RMR is subtracted to get PAT). Given the "free-living" context, accelerometers or a similar method for estimating daily activity energy expenditure would be plausible.
• Energy Intake: This would have been assessed through dietary records or recalls, where participants log everything they eat and drink. While prone to self-report bias, this is a common method for estimating energy intake in free-living studies.
• 24-hour Glycemic Responses: This was measured using Continuous Glucose Monitoring (CGM) devices. These small sensors are worn on the body (e.g., on the arm) and continuously measure glucose levels in the interstitial fluid, providing a detailed picture of blood sugar fluctuations throughout the day and night. This is a highly objective and precise way to track glycemic responses.
• Health Markers (Hematology/Biopsies): The abstract mentions "hematology/biopsies" and "systemic indexes of cardiovascular health."
  • Hematology refers to blood tests, which would have been used to measure various blood markers related to metabolic health (e.g., cholesterol, triglycerides, insulin, inflammatory markers) and general health.
  • Biopsies (likely adipose tissue biopsies, as "adipose tissue glucose uptake" is mentioned) involve taking a small sample of fat tissue, typically from the abdomen or thigh. This tissue can then be analyzed in the lab to assess cellular processes, such as how fat cells respond to insulin and take up glucose. This provides a direct measure of tissue-specific metabolic function.

The Bath Breakfast Project employed a randomized controlled trial (RCT) with repeated measures, a robust study design for establishing causal links.

• Study Design:

  • Randomized Controlled Trial (RCT): Participants were randomly assigned to either the "breakfast" group or the "extended fasting" group. Randomization is crucial because it helps ensure that, on average, the two groups are similar at the start of the study in all characteristics (known and unknown) except for the intervention itself. This minimizes the risk that any observed differences at the end of the study are due to pre-existing differences between the groups rather than the intervention. This design is considered the gold standard for determining cause-and-effect relationships.
  • Repeated Measures: Key outcomes were measured at both baseline (before the intervention started) and at follow-up (after the 6-week intervention period). This allows researchers to compare changes within each individual over time, as well as compare the changes between the two groups. This increases the statistical power of the study and helps account for individual variability.
  • Free-living conditions: The study was conducted while participants maintained their normal daily routines, rather than being confined to a metabolic ward. This increases the ecological validity of the findings, meaning the results are more likely to apply to real-world situations. However, it also introduces more potential for uncontrolled variables.

• Randomisation: Participants were randomly allocated to one of two groups:

  1. Breakfast Group: Consumed at least 700 kcal before 11:00 AM daily.
  2. Extended Fasting Group: Consumed 0 kcal until 12:00 PM daily. The random assignment ensures that any observed differences between the groups at the end of the study are more likely due to the breakfast intervention rather than other confounding factors.

• Blinding: The abstract does not mention blinding. It is highly unlikely that participants were blinded to their intervention (they knew whether they were eating breakfast or fasting). It is also unlikely that the researchers directly interacting with participants could be blinded. However, the researchers analyzing the data (e.g., blood samples, DEXA scans, CGM data) could have been blinded to the group assignments, which helps reduce observer bias in data interpretation. The lack of participant blinding is a common limitation in dietary intervention studies, as it's difficult to conceal whether someone is eating or not. This means participants' expectations or beliefs about breakfast could potentially influence their self-reported outcomes or even their activity levels.

• Washout Periods: Not applicable here, as it was a parallel-group RCT (participants stayed in their assigned group for the duration) rather than a crossover design (where participants switch groups after a washout).

• Duration: The intervention lasted for 6 weeks. This duration is generally considered sufficient to observe short-to-medium term metabolic adaptations and changes in energy balance components. However, it might not be long enough to detect subtle changes in body composition or long-term health markers that develop over months or years.

• Statistical Approach: The abstract mentions "95% CI" (confidence intervals) and "CV" (coefficient of variation), indicating that statistical analyses were performed to compare the groups and assess the precision of the estimates. The use of confidence intervals is good practice, as it provides a range within which the true effect likely lies, rather than just a single point estimate.

• What this design can and cannot prove:

  • Can prove: The RCT design allows the researchers to establish a causal link between eating breakfast (or fasting) and the observed changes in energy balance components and glycemic responses. Because of randomization, the differences observed between the groups are highly likely to be caused by the breakfast intervention.
  • Cannot prove:
    • Long-term effects: A 6-week study cannot definitively prove the long-term health consequences of daily breakfast or fasting over months or years.
    • Generalizability to other populations: The study was conducted on lean adults. The findings may not directly apply to individuals who are overweight, obese, have metabolic disorders (e.g., type 2 diabetes), or are from different age groups or ethnic backgrounds.
    • Mechanism of action: While it shows what happened, the abstract doesn't delve deeply into the precise biological mechanisms (e.g., hormonal changes, gut microbiome shifts) that might explain why these effects occurred.
    • Impact of breakfast composition: The study defined breakfast as "≥700 kcal before 1100," but did not specify the macronutrient composition (e.g., high-carb vs. high-protein). The specific type of breakfast might influence outcomes differently.

• Major methodological weaknesses:

  • Lack of blinding for participants: As participants knew their assigned intervention, their expectations or beliefs about breakfast could have influenced their behavior (e.g., activity levels, dietary choices outside the specified intervention) or self-reported data.
  • Reliance on self-reported energy intake: Dietary records, while common, are prone to under-reporting or misreporting of calorie intake, which could affect the accuracy of the energy intake findings.
  • Limited generalizability: The study's focus on lean, healthy adults means its findings cannot be directly extrapolated to other populations, especially those with metabolic health challenges where breakfast might have different effects.

The study revealed several important findings regarding the causal role of breakfast in lean adults:

• Resting Metabolic Rate (RMR):
  • There was no metabolic adaptation to breakfast. RMR remained stable within 11 kcal/d between the breakfast and fasting groups. This means that eating breakfast did not "kick-start" metabolism or significantly increase the calories burned at rest over the 6-week period.
• Energy Intake:
  • Participants in the breakfast group had a greater overall dietary energy intake, consuming 539 kcal/d more than those in the fasting group (95% CI: 157, 920 kcal/d). This indicates that eating breakfast