PoshanSense
Food choices · 5 min read

Does cheese affect cholesterol the same way as butter?

The food surrounding saturated fat can change the result. But better than butter is not the same as best for cholesterol—or unlimited cheese.

Plain cheese, butter and seeded bread on a mustard-coloured table.
Original food illustration; the portions do not reproduce the feeding trial.

Complementary human studies · Published 2015 & 2017

Two foods can supply similar saturated fat and still behave differently in a feeding study. That is the interesting part of the cheese-and-butter question. The misleading part is turning it into “cheese is good, butter is bad” and forgetting what the experiment actually compared.

Better than butter is a comparison, not a health badge

If one food leads to a lower LDL result than another, that is useful information. It does not tell you that the first food is the best possible choice, that larger portions are harmless, or that adding it to your current diet will improve a blood test.

A replacement and an addition are different actions. A slice of cheese taking the place of a butter-heavy ingredient is not the same as adding more cheese to an unchanged meal. Before applying a finding, name which of those you mean.

What did the Canadian comparison show?

In the complete primary report, LDL cholesterol after the cheese diet was about 3.3% lower than after the butter diet. The menus supplied similar amounts of saturated fat; researchers controlled several other nutrient features and aimed to keep weight stable.

But the same paper supplies the essential second half: LDL after cheese was higher than after the monounsaturated-fat and polyunsaturated-fat diets. So the statement “cheese performed better than butter” cannot be stretched into “cheese beats every alternative.”

HDL was the comparison used for the primary sample-size calculation, and HDL did not differ significantly between cheese and butter. LDL and HDL are different measures; the study is not accurately described as a clear victory on its prespecified primary comparison.

Original explanatory infographic

“Compared with what?” changes the answer

01

Cheese versus butter

Matched saturated fat did not produce identical LDL responses in the Canadian trial.

02

Cheese versus unsaturated-fat diets

The cheese diet still produced higher LDL in that comparison.

03

One marker versus disease

A short lipid experiment cannot count heart attacks prevented.

04

Food versus portion

The food matrix matters; it does not make quantities irrelevant.

Comparisons from specified trial diets, not a ranking of all dairy products.

What is the food matrix?

It means the surrounding structure and ingredients of a food. Cheese contains protein, calcium and a physical structure that butter does not share in the same way. Looking at saturated fat alone therefore does not describe every difference between the foods.

The Canadian paper discusses possible mechanisms, including calcium and fat absorption. These are explanations to investigate, not a proven reason that every cheese must have the same effect. The trial did not isolate each mechanism by changing it alone.

That is why nutrient labels remain useful while being incomplete. They tell you what is present in a serving; they do not capture every feature of how an entire food is digested. Neither perspective has to replace the other.

Does European research say the same thing?

A Norwegian randomized trial compared two particular cheeses with a control group over eight weeks. Its overall interpretation was a neutral effect on cholesterol and metabolic-syndrome measures, with some differences appearing in selected baseline subgroups.

It was not another matched cheese-versus-butter feeding experiment. The participants largely kept their habitual diets, and the cheeses differed substantially. The European result adds context, but pooling the headlines would hide differences in foods, comparison groups and endpoints.

The practical lesson is to resist a sweeping dairy verdict from one product. A hard cheese, a fresh cheese and a cheese-containing snack may differ in serving size, salt and other ingredients. These trials do not validate all of them individually.

What should change on your plate?

Start with the usual meal rather than a photograph of a research diet. How much cheese or butter do you actually use? Is it a garnish, a spread, the main filling or one ingredient in a larger dish? Which food would change if you made a swap?

The same questions work with toast in the US, a sandwich in Europe or a familiar dish elsewhere. There is no need to reproduce a Canadian menu. Equally, the study does not establish identical effects for every locally available dairy product.

If your aim is managing a high LDL result, look at the broader routine and discuss suitable changes with your clinician or dietitian. A food experiment can inform that conversation; it cannot choose treatment or make a personal cardiovascular prediction.

Keep the interesting finding in proportion

The evidence makes an important point: whole foods matter, and equal saturated fat does not guarantee identical short-term lipid responses. It also makes a quieter point that is just as useful: a favourable comparison against butter is not a licence to ignore quantities or other alternatives.

You can enjoy a food without asking it to become a treatment. Keeping the portion visible and understanding the replacement gives this question a more useful answer than a simple good-food/bad-food list.

Evidence context: the Canadian study had dairy-sector/public funding and several author industry relationships. The Norwegian study had public/agricultural and TINE funding and supplied cheese. Both report funders did not direct the analysis or writing.

Sources & further reading

  1. Canada · cheese, butter and replacement diets (2017)
  2. Author-hosted complete Canadian primary paper
  3. 2015 Norwegian cheese trial

Study years differ from this article’s publication date. We reviewed full primary papers and relevant supporting material; access limitations are kept out of efficacy claims. Original illustrations and diagrams; no publisher figures reproduced. Educational writing, not clinical review.

Published 11 October 2026 · Sources checked 11 October 2026. Suggest a correction.