Effective fat loss strategies are necessary for individuals with obesity and other conditions in which excessive fat tissue may pose a health risk. One commonly recommended strategy is to eat breakfast, and there has been a tremendous amount of research investigating the effects of eating various types of breakfasts on satiety, metabolism, and health.

Yet, there has also been an increasing number of studies showing that breakfast is not king. For instance, in August of 2014, David Levitsky from Cornell University published an editorial where he rightly pointed out that breakfast is the most important meal of the day only if you are selling breakfast cereals.

This editorial was spurred in part by research examining the effects of breakfast consumption or omission on all components of energy balance and health in lean men and women over the course of six weeks. The results of which showed that there are no metabolic adaptations or adverse cardiometabolic effects of eating or omitting breakfast.

It would be silly to assume that these findings are generalizable to the obese population, however, which is why Chowdhury et al, the same group of researchers that conducted the aforementioned lean individual breakfast research, repeated their study in participants with obesity.

The set-up

During this 6-week intervention, 23 obese men and women were randomly assigned into a group that was required to consume at least 700 kcal before 11:00 am daily, with at least half eaten within 2 hours of waking up (breakfast group), or into a group that could not consumed anything except water until noon daily (fasting group). Importantly, the participants were stratified before randomization according to their habitual breakfast habits. This way, an equal number of individuals who regularly eat or skip breakfast were in both groups.

Intensive laboratory-based measurements were performed during the first and last week of the intervention to determine resting metabolic rate (via indirect calorimetry) and body composition (via dual-energy X-ray absorptiometry), as well as concentrations of key metabolites and hormones (via blood sampling), glucose tolerance (via an oral-glucose tolerance test and 24-hour continuous glucose monitoring), and physical activity levels (via combined heartrate and accelerometry).

Dietary intake was assessed from weighted food logs. Notably, packaging from convenience items was kept by the participants for analysis by the researchers using the manufacturers’’ information.

During the 4 weeks between the measurement weeks, the participants were free-living and not monitored.

Breakfast vs fasting

Starting with the broad picture of energy balance, which is illustrated in figure 1 below, there are few significant differences between the breakfast and fasting groups. Energy intake in terms of calories and macronutrient composition was not significantly different between the groups. However, this may be a consequence of the small sample size as the average intake in the breakfast group was 14% greater than in the fasting group (2719 vs 2381 kcal/d).

Regarding energy expenditure, no significant effect of breakfast consumption or skipping was noted for resting metabolic rate or diet-induced thermogenesis. Additionally, there was no significant differences between the groups in physical activity energy expenditure or intensities of activity over the entire day (figure 2). Again though, this may be a consequence of the small sample size as the breakfast group had ~28% greater physical activity energy expenditure than the fasting group. Additionally, the difference in physical activity energy expenditure was significant before lunch, with the breakfast group being more active. However, after lunch there was no difference between groups.

Thyroid hormones that regulate resting metabolic rate (free-T3 and free-T4) were unresponsive to either treatment. Similarly, appetite-regulating hormones such as ghrelin (total and acylated), peptide YY, active glucagon-like peptide-1, and adiponectin were not significantly different between groups. There was a trend for leptin to increase in the breakfast group (+20%) and decrease in the fasting group (-13%), but the consequences of this change remain unknown.

Neither group showed any significant changes in bodyweight or body composition, although there was a significant increase in bodyweight (+1.3 lbs) when all participants from both groups were analyzed together. Similarly, blood lipids and inflammatory markers remained unchanged in both groups, but total cholesterol and LDL-c significantly increased when the groups were combined.

Fasting blood glucose and insulin, the Matsuda and HOMA-IR indexes of insulin sensitivity, and the glycemic response to the oral-glucose tolerance test were not affected by the intervention in either group. Additionally, the insulin response to the oral-glucose tolerance test was not significantly different between groups and was not significantly changed over time within each group. However, the change over time between groups was significantly different, with the breakfast group showing an overall reduction and the fasting group showing an overall increase. Notably, these changes appeared to be driven by a few individuals only, for whom the change was markedly noticeable.

Continuous glucose monitoring also showed little difference between groups. Average peak glucose concentrations from waking to lunch, from lunch until sleep, and during sleep were similar between groups and did not change over time. Similarly, glucose variability (accumulated hyper- and hypoglycemic episodes) and average blood glucose levels over the 24-hour day were not responsive to either intervention and were not different between the groups.

Breakfast is not king in those with obesity

The current study set out to investigate how breakfast consumption or omission impacted all aspects of energy balance. No differences in reported total caloric intake was noted between the groups, despite the breakfast group being required to consume an additional 700 kcal before lunch. In fact, the average difference in caloric intake was only 338 kcal, indicating that the fasting group did somewhat compensate for the missed meal. The bottom line remains: they still ate less.

Energy expenditure in the morning was greater in the breakfast group. Perhaps not eating promotes feelings of lethargy, or perhaps eating gives more energy than normal. Either way, total daily physical activity energy expenditure was not significantly different between groups, indicating that the morning activity was minimal. Indeed, the difference between groups for the day was only 272 kcal, which is less than the 338 kcal food intake difference noted above.

When we combine the above, the overall impact of breakfast on energy balance is minimal. Still, in view of the fact that 10 of the 11 individuals assigned to eat breakfast gained weight, it appears that promoting breakfast consumption as a weight-loss aid may be misguided. This is certainly in line with a 16-week weight loss intervention that found no effect of breakfast consumption or omission on weight loss.

The above said, breakfast consumption did appear to lower the insulin response to an oral glucose-tolerance test without changing the glucose response, indicating increased insulin sensitivity. However, the lack of difference between groups in fasting glucose and insulin, the Matsuda and HOMA-IR indexes of insulin sensitivity, blood lipids, inflammatory markers, and 24-hour glucose variability raises questions as to the relevance of this finding.

Recommending breakfast consumption to facilitate weight loss may be misguided when done on a public level. There will be individuals who respond well to eating breakfast, but population-level advocacy is not supported. There may be other health benefits to breakfast consumption, which future research will need to discern, but it will be important to discriminate between the types of potential benefits. Similarly, the type of breakfast being eaten will likely play a huge role on health and weight management. The current study used a free-living population, making the results more applicable to the general public given blanket advice, but more specific recommendations could lead to a different outcome.