Study Finds Keto, Mediterranean & Plant Diets Produce Similar Weight Loss

Highlights

• A five-month randomized trial compared ketogenic, Mediterranean, and plant-forward eating in 55 adults.
• All three groups lost about the same weight, roughly a tenth of what they started at, and all three improved metabolic health.
• Liver fat fell furthest on the low-carbohydrate plan, by about two thirds rather than just under half.
• The findings apply to one specific group, and the researchers are clear that weight loss itself did most of the work.

Nutrition studies usually carry a quiet caveat: nobody really knows what the participants ate. People forget. People round down. A plan followed loosely for five months is not the plan written into the protocol, and the gap between the two is where a lot of dietary research quietly loses its footing.

A trial published on 27 August in the journal Cell Metabolism went at that problem from the other end. Researchers at WashU Medicine supplied participants with every single meal for the length of the study, then measured what happened. No food diaries, no self-reporting, no guessing.

What came back is an unusually clean answer to a question people ask constantly. Does the type of diet matter, or only the weight that comes off? The answer turns out to be: it depends entirely on what you are measuring.

The scale said the three diets were equal. The liver measurements said something else.

The short version

What the Trial Actually Did

Fifty-five adults were randomly assigned to one of three eating patterns and stayed on it for around five months. Everyone entered with the same starting profile: obesity alongside raised blood sugar and extra fat stored in the liver, a combination the researchers describe as metabolically unhealthy obesity.

The design is what makes the results worth reading. Providing food rather than advice is expensive and logistically awkward, which is why so few trials do it, and it is also the only way to know with confidence that the three groups actually ate three different diets.

1. Random assignment. Participants were placed into one of the three diets rather than choosing, which keeps personal preference from shaping the result.
2. Every meal supplied. All food came from the study for the full five months, so what was eaten was known rather than reported.
3. Weekly dietitian meetings. Each participant met a study dietitian every week, which kept adherence high across all three arms.
4. Measurement beyond the scale. Liver fat, insulin sensitivity in muscle and liver, and 24-hour glucose and insulin were all tracked, not just body weight.

The three plans differed sharply in their make-up, which is the point. They ranged from very low carbohydrate to very high, covering most of the territory the popular diets occupy.

On the Scale, a Three-Way Tie

Weight loss came out the same across all three groups. Participants shed roughly ten percent of their starting weight, whichever plan they had been assigned.

Insulin sensitivity in muscle told the same story. It improved by about half again from where each participant started, and it improved by that much on every diet. Whether the calories arrived mostly as fat or mostly as carbohydrate made no measurable difference to how muscle responded.

That finding is worth sitting with, because it supports something people have said for years without much hard evidence behind it. If the goal is losing weight and improving how muscle handles blood sugar, the plan you can actually keep is a defensible plan to choose. Three quite different approaches arrived at the same destination.

Where the Diets Parted Company

The liver told a different story. All three groups improved, which is worth saying plainly. But measured against the other two, the low-carbohydrate plan produced a noticeably larger change across several markers at once.

That last figure has the widest spread of any result in the study. Half of the low-carbohydrate group moved out of the prediabetes range, against 29 percent on the Mediterranean plan and 7 percent on the high-carbohydrate one. Same weight lost, three quite different outcomes.

Why Muscle and Liver Responded Differently

The split between the two tissues is the most interesting thing in the paper. Muscle responded to the weight loss and appeared indifferent to how that weight loss was achieved. The liver responded to the composition of the diet on top of the weight loss.

The research team points to two hormones. The low-carbohydrate group saw the largest rise in glucagon, which prompts the liver to release stored fat, and the largest fall in circulating insulin. Both push in the same direction where liver fat is concerned.

The drop in insulin has a straightforward explanation behind it. With very little carbohydrate arriving, there is far less glucose for insulin to manage, so the pancreas is not called on to produce as much of it. The 74 percent figure is less a treatment effect than a description of a system that has been asked to do considerably less work.

Same weight off, same muscle response. The difference showed up in an organ nobody was weighing.

Worth remembering

Why Liver Fat Was the Marker to Watch

Choosing liver fat as a primary measure was not arbitrary. Fat accumulation in the liver is common among adults carrying excess weight, and it has become the fastest-growing driver of chronic liver disease worldwide.

What makes it a useful thing to measure is that it moves. Unlike some markers that shift only slowly, liver fat responds to dietary change over a matter of months, and it responded in every group in this trial. The 45 percent reduction on the Mediterranean and plant-forward plans is a substantial result in its own right, and it is easy to overlook next to the larger number.

That is the reading the researchers themselves emphasise. All three diets worked. One of them did something additional.

The Result Nobody Expected

The finding that most surprised the team had nothing to do with the liver. The group eating by far the most fat did not see their blood fats or cholesterol climb across the five months.

That runs against the intuition many people bring to a high-fat diet, and it is the sort of result that earns a study a second look. It is one trial, in one carefully chosen group, over five months, so it settles nothing on its own. It does make the question more interesting than it was a week ago.

Where the Weight-Loss Medicines Fit

The researchers raise a point that gives the study a longer shelf life than most diet trials. Many people who match this profile are also candidates for the newer weight-loss medicines, and those medicines are very effective at the job of taking weight off.

Their argument is that this does not make the diet question redundant. If the composition of a diet changes specific outcomes independently of how much weight is lost, then it still matters what someone eats while the weight is coming off, whatever else is helping it along. Understanding that interaction is on the team's list for future work.

Who the Findings Apply To

The boundaries here are worth repeating, because they are easy to lose in a headline. Everyone in this trial had obesity together with raised blood sugar and fat in the liver. Whether someone without that particular combination would see the same pattern is an open question the study was not designed to answer.

There is also the matter of how the results were produced. Every meal was supplied and every participant had a weekly appointment. That is a long way from choosing a plan at home in January and hoping it holds through February.

What to Take From It

The headline is not that one diet beat the others outright. Weight loss helped everyone in this study, on all three plans, and that remains the main event. The senior author makes the point directly: three diets with vastly different make-ups all improved metabolic health through weight loss alone.

The practical read: if weight is the goal, pick the plan you can live with. If a specific marker is the goal, what the plan is made of starts to matter as well.

What the trial adds is a second layer to that advice. Two people can lose the same ten percent and end up in measurably different places on things that never appear on a bathroom scale. For anyone tracking a specific marker, that distinction is a useful one to carry into the conversation.

The old advice survives, then, with a footnote attached. The best diet may still be the one you can stick to, and the details of what is on the plate are not beside the point.