In a world where the overconsumption of food is increasing, which diet would you put your money on to promote eating less?

According to a recent landmark study, something about ketogenic, low-carbohydrate diets make us more likely to overeat – but the story is more complicated than it may seem!

‘I’ll take a brief overview, please!’
If you’ve been on social media recently, you’ve probably seen competing support for low-carbohydrate, ketogenic diets versus plant-based diets. Supporters argue that the benefits of “their” diet  include helping to maintain your weight while the other leads to increased hunger, overconsumption of calories, and ultimately, excess body weight. But who’s right? Until now, no one actually knew. This study is one of the first to actually compare the two diets under very strict conditions.

The researchers found that participants ate more on an animal-based, ketogenic, low-carbohydrate (ABLC) diet compared to a plant-based, low-fat (PBLF) diet -even though the two diets were matched on calories, protein and non-starchy vegetables. These components of the diets obviously differ from the higher proportion of carbohydrates (carbs) that largely characterise plant-based diets to the elevated proportion of fat that makes up the majority of ketogenic diets.

What was even more interesting was that participants on the ABLC ate more calories even though this diet led to a greater increase in blood ketone levels and a greater reduction in glucose and insulin levels compared to the PBLF diet.

Ah, now we’ve got your attention! So what did researchers of this study do to shed more light on these contrasting diets?
They admitted 20 adults into a clinic for four weeks, where the researchers could meticulously control participants’ environment and food intake. Metabolic outcomes were also measured down to the finest detail, including blood ketones, glucose, insulin and lipids (e.g. cholesterol), blood pressure, body weight and composition and physical activity.

Participants were advised that this was not a weight loss study, so they should not try to lose weight. They also wore loose clothing and were not privy to their daily weight, ketone and glucose measurements. These strategies reduced the chance of participants altering their behaviour (i.e. how much they ate) to be more in-line with expected outcomes (aka expectation bias).

It is important to emphasise here that this type of study is considered the Rolls Royce of clinical research, especially in the nutrition space where it is often difficult to fully capture what participants are eating versus what they say they are eating.

After being randomly assigned to one of the two diets, participants were instructed to eat as much or as little (ad libitum) as they desired of either the ABLC diet or the PBLF diet for a period of two weeks. After this initial two-week period, participants swapped groups to consume the alternative diet for another two weeks.

What did the participants eat?
The percentage of protein was similar between the diets at ~14%. However, the ketogenic diet contained ~10% carbs versus ~75% fat, which was the opposite to the plant-based diet with ~75% carbs and ~10% fat.

Perhaps unsurprisingly, over 80% of calories came from animal products in the ABLC diet, as well as containing elevated levels of salt, whereas the PBLF diet was primarily made up of plant-based products and contained greater levels of fibre.

Example images of an animal-based, ketogenic, low-carbohydrate meal (zucchini pasta with meat sauce) versus a plant-based, low-fat meal (spaghetti with vegan pasta sauce)
Image Source: Hall KD, et al. (2020) osf.io/preprints/nutrixiv/rdjfb

What did the researchers more specifically find?
The ABLC diet led to greater blood ketone levels (~3mM) (which is thought to suppress appetite), whereas the PBLF diet led to elevated after-meal (or postprandial) glucose and insulin levels (which is thought to induce food cravings).

Surprisingly, however, the ABLC ketogenic diet prompted participants to eat ~700 more calories (~2800 kilojoules) per day than the PBLF diet. More calories consumed in the ABLC diet occurred despite a greater mass of food being eaten in the PBLF diet and despite there being no reported difference in participants’ ratings of appetite or enjoyment of meals.Image Source: https://twitter.com/KevinH_PhD/ status/1258102121246806017  KevinH_PhD (2020, May 7th 2020). Retrieved June 8th 2020 

While the ABLC diet resulted in increased calorie consumption, the macronutrient composition of the consumed foods and beverages was as expected, with ABLC consumption being 9.9% carbs, 74.6% fat and 15.5% protein and PBLF consumption being 10.5% fat, 75.5% carbs and 14.0% protein.

Although carb and fat intake obviously differed between the diets, protein intake was higher during the ABLC diet, both in absolute terms (+136 calories per day) as well as when expressed as a fraction of the energy consumed (15.5% with ABLC versus 14.0% with PBLF).

In terms of body weight, even though participants on the ABLC diet had more rapid weight loss during the first week compared to the PBLF diet, and increased calorie consumption overall, total weight loss was similar in both groups at the end of the diets.

Results of body composition assessments help explain how total weight loss between the ABLC and PBLF diets were similar despite differences in calorie consumption. Body composition assesses all body components, such as fat and ‘fat-free mass’ (i.e.water, bone, organ and muscle content).

Only the PBLF diet resulted in significant loss of body fat, whereas only the ABLC diet led to loss of fat-free mass. The researchers suggested that weight loss from cutting carbs in the ABLC diet may be largely due to the excretion of body fluids, such as water, gastrointestinal contents and glycogen (i.e. a form of glucose that serves as an energy store in the body).

Image Source: Hall KD, G. J., Courville AB, Boring J, Brychta R, Chen KY, et al. (2020). A plant-based, low-fat diet decreases ad libitum energy intake compared to an animal-based, ketogenic diet: An inpatient randomized controlled trial [Internet]. NutriXiv. Retrieved from Available from: osf.io/preprints/nutrixiv/rdjfb

During both diets, participants also experienced beneficial reductions in fasting blood glucose, insulin and markers of inflammation. However, participants on the PBLF diet experienced a reduction in blood cholesterol levels, blood pressure and the rate at which our heart beats (or pulse rate), whereas those on the ABLC diet did not.

Interestingly, from the start to the end of both diets, it was only the ABLC  ketogenic diet that led to lower levels of fasting triglycerides, whereas the PBLF diet gave rise to increased concentrations. A greater level of triglycerides (both in the fasting and after-meal state) is thought to increase the risk of cardiovascular disease.

What does all this mean?
These results go some way to challenging assumptions about these diets, including the premise that PBLF diets contain high-glycaemic carbs that increase after-meal insulin levels, which in-turn causes increases in body fat, hunger and overeating.

Although the PBLF diet in the current study was made up of foods with higher glycaemic load that increased after-meal glucose and insulin levels compared to the ABLC diet, participants on the plant-based diet consumed significantly fewer calories.

Image Source: https://twitter.com/KevinH_PhD/ status/1258102121246806017  KevinH_PhD (2020, May 7th 2020). Retrieved June 8th 2020

It is also important to note that although the ABLC diet did not promote body fat loss (compared to the PBLF diet), it did not lead to body fat gain either. This questions the notion that high-fat foods intrinsically promote body fat accumulation.

What can’t this study tell us?
Because the two diets were short-term (two weeks each), it is still unclear whether long-term intake of either diet would lead to different results.

It is also hard to know whether the differences in body fat and blood levels were due to the different make-up of macronutrients between the diets (i.e. high-fat, low-carb versus low-fat, high-carb) or due to the difference in calories consumed per day (~700 calories more during the ABLC ketogenic diet). Also, it remains unclear why higher levels of fasting blood triglycerides resulted after the PBLF diet.

Participants were also instructed to eat as much or as little of each diet as they wanted, so it remains unknown whether results would hold up if individuals were pursuing active weight loss. This is particularly unknown in a context where individuals may undertake such diets in a home setting rather than clinical environment.

What should we do?
This impressively robust study suggests that PBLF diets may benefit appetite control, whereas ABLC ketogenic diets may benefit after-meal glucose and insulin levels.

But, before you hedge your bets, if you see something on social media saying one diet is better than another, it may be worth keeping in mind that it is often much more complex than simply pitting macronutrient profiles, such as high-fat, low-carb versus high-carb, low-fat, against one another. More evidence from rigorous studies such as this is what is needed to inform our ideas about which diet might be best for ourselves or others.

Melissa is a PhD candidate within Deakin University’s Food & Mood Centre. She holds a Bachelor of Nursing (Pre-registration), Graduate Diploma in Psychology and Bachelor of Psychological Science (Honours). Melissa has a particular interest in the link between diet, gut microbiota and metabolic as well as mental health outcomes.