Full-Fat Dairy Didn’t Worsen Weight or Cholesterol in a Small Trial

For decades, dietary advice steered people toward low-fat dairy. A new randomized trial adds to a growing body of evidence questioning that — though its size, length and funding call for caution.

Researchers at the University of Toronto, led by nutrition professor Harvey Anderson, ran a 12-week randomized controlled trial in 74 adults with overweight or obesity. Participants followed one of three diets: low-dairy with calorie restriction; an energy-neutral diet with three daily dairy servings; or an unrestricted diet with three daily servings. Most of the dairy was full-fat.

What they found

Full-fat dairy did not worsen weight, body composition, or cholesterol and lipid levels. Those eating three servings a day showed improved blood pressure and higher intake of calcium, protein and vitamin D.

“Those that had three servings of dairy didn’t have adverse levels of blood cholesterol or lipids or evidence of insulin resistance,” Anderson said.

Where the low-fat advice came from

The recommendation followed a reasonable chain of reasoning that has since been questioned at several points.

Saturated fat raises LDL cholesterol; LDL cholesterol is associated with cardiovascular disease; therefore reducing saturated fat should reduce cardiovascular disease. Dairy fat is substantially saturated, so low-fat dairy became standard advice.

Each link holds individually. The difficulty is that the chain has proved harder to demonstrate end to end than expected, particularly for dairy specifically, and observational studies of dairy consumption have repeatedly failed to show the harm the reasoning predicts.

The dairy matrix idea

Researchers point to the dairy matrix — the proposition that the way dairy’s fats, proteins and nutrients are packaged together affects the body differently than isolated saturated fat.

The underlying claim is that food is not simply the sum of its nutrients. Dairy fat exists within a structure of milk fat globule membranes, is accompanied by calcium, phosphorus and specific proteins, and is fermented in cheese and yoghurt — and these features may alter how the fat is absorbed and metabolised.

Calcium is the most concrete proposed mechanism: it can bind fatty acids in the gut, forming compounds that are excreted rather than absorbed, which would mean less of the fat consumed actually enters circulation.

The concept is plausible and not fully established. It is also invoked frequently by the dairy industry, which is a reason to examine it carefully rather than to dismiss it.

Reading it with caution

Two caveats matter.

The study was short — 12 weeks — and limited to adults with overweight or obesity, so it cannot speak to long-term heart-disease risk or to other groups. Cholesterol responds to diet within weeks, so measuring it over 12 weeks is legitimate; cardiovascular events accumulate over decades, and no trial of this length can address them.

And it was funded by dairy-industry sources (Dairy Research Cluster 3), a conflict of interest to weigh.

How much the funding should matter

Industry funding warrants scrutiny without automatically invalidating a result, and the distinction is worth being precise about.

Research consistently shows industry-funded nutrition studies more often produce conclusions favourable to the sponsor. The mechanism is usually not fabrication but design and framing: which comparison is chosen, which outcomes are emphasised, whether unfavourable findings are published at all.

A randomized controlled trial is more resistant to those pressures than an observational study, since the design is specified in advance. The appropriate response is to weight the finding less heavily than an equivalent independently funded trial, and to look for whether independent research points the same way — which, for dairy, it broadly does.

What a 74-person trial can and cannot establish

Sample size determines what questions a study can answer, and 74 participants across three arms means roughly 25 per group.

That is adequate for detecting reasonably large differences in measures that vary predictably — body weight, blood pressure, cholesterol. Those outcomes change measurably in most individuals who change their diet, so a real effect of meaningful size would be visible.

It is inadequate for detecting small differences, and this is where the result requires care. A finding of “no worsening” in a small trial is not the same as a finding of “no effect” — a modest adverse effect could easily be present and undetectable at this size.

The distinction matters because the low-fat recommendation was never premised on dairy fat producing large harm. It was premised on a modest effect on cholesterol accumulating over decades across a population, which is precisely the magnitude a 74-person 12-week trial is least able to rule out.

Why nutrition questions stay unresolved

The persistence of this debate reflects methodological constraints rather than a lack of research.

The definitive study — randomising thousands of people to different dairy intakes and following them for decades to count heart attacks — has never been done, and probably never will be. It would be extraordinarily expensive, and people cannot be kept on assigned diets for that long.

What exists instead is observational data, which cannot separate dairy consumption from everything correlated with it, and short mechanistic trials like this one, which measure intermediate markers over weeks. Both are informative; neither is conclusive; and the gap between them is where dietary controversy lives.

It adds to the case that full-fat dairy may not be the villain it was once thought — but it is not the final word. This summarises one short study and is not dietary or medical advice.