Is olive oil really good for you?

7 min read

Olive oil is often called "the star of the Mediterranean diet" and "a healthy oil."

So how much has actually been confirmed in research on people?

The short answer: the strongest evidence is for preventing cardiovascular disease. But when reading that evidence, you need to keep three things apart:

• The effect of the whole "Mediterranean diet + olive oil" diet
• Observational studies following "people who consume a lot of olive oil"
• The effect of "replacing butter or margarine with olive oil"

Mix these up, and you get the misreading that "as long as you consume olive oil, you'll be healthy."

The strongest evidence: an RCT of the Mediterranean diet

The biggest piece of evidence is the PREDIMED trial in Spain. It divided 7,447 people at high risk of cardiovascular disease into three groups.

The hazard ratio for major cardiovascular events (heart attack, stroke, cardiovascular death) was 0.69 (95% CI 0.53–0.91) in the EVOO group and 0.72 (0.54–0.95) in the nut group [1]. Note that the original 2013 paper was reanalyzed after methodological problems and republished in 2018 [1].

What matters here is that this was not a trial of olive oil alone. It was an intervention on the whole Mediterranean diet, adding vegetables, legumes, fish and more, so "the effect of olive oil alone" can't be separated out. Still, it's valuable as a large RCT showing that an olive-oil-centered dietary pattern may reduce events through a change in diet rather than medication.

Large US cohorts found associations too

Is this only about people around the Mediterranean? Two US cohort studies by a Harvard research group addressed that question.

But both are observational studies. People who consume a lot of olive oil may differ in other lifestyle habits. The researchers adjusted for major dietary and lifestyle factors, but cause and effect can't be claimed.

Also, people in the highest group (over 7 g a day) were only about 5% of the total at the start of the study [3].

"Replacement" is the most practical angle

What I most want to highlight in the 2020 study is the replacement analysis.

• Replacing 5 g/day of margarine, butter, mayonnaise, or dairy fat with the same amount of olive oil was associated with a 5–7% lower risk of cardiovascular and coronary heart disease [2]
• But comparing olive oil with "other vegetable oils combined" showed no significant association [2]

In other words, olive oil's strength lies less in being "better than other vegetable oils" and more in being usable "in place of animal fats and processed fats."

"Good for the brain" is still at the observational stage

In 2024, a study on dementia using the same cohorts was published. Following about 92,000 people for 28 years, those consuming more than 7 g of olive oil a day had a 28% lower risk of dementia-related death than those who rarely consumed it (adjusted hazard ratio 0.72) [4]. The same trend was seen regardless of overall diet quality [4].

But the outcome was "death with dementia as the cause," not "developing dementia," and it's observational, so we can't say "olive oil prevents dementia." In the replacement analysis, replacing margarine and mayonnaise was significant (8–14% lower), while replacing butter or other vegetable oils was not [4].

Are EVOO and refined olive oil the same?

Extra virgin olive oil (EVOO) is said to contain more polyphenols. So are the effects specific to EVOO?

The US cohort studies couldn't separate EVOO from regular olive oil in their analysis [2]. A report analyzing PREDIMED participant data as an observational study found that the group with the highest EVOO intake had a 39% lower risk of cardiovascular disease [5]. But this too is an observational analysis based on participants' intake, not the trial's random assignment.

So for now, the accurate summary is: "EVOO may have an advantage, but the evidence isn't enough to say definitively how it differs from refined olive oil."

What to look at on the label when buying is in how to choose olive oil.

Is it okay to heat it?

You often hear that "olive oil has a low smoke point, so it's not suited to heating."

In 2018, a lab study comparing ten cooking oils under heating was reported. EVOO was among those producing the fewest polar compounds, considered harmful, even at high temperature (180°C for a long time), while refined seed oils such as canola produced more [6]. A high smoke point didn't necessarily match stability under heating.

But this study has limitations.

• What was heated was the oil itself, not food being cooked [6].
• The study was done at a testing lab that is a subsidiary of an olive oil producer. A possible conflict of interest should be kept in mind.
• It didn't measure effects on human health.

So the claim that "EVOO isn't suited to cooking with heat" is, at the very least, weakly supported. On the other hand, we can't go as far as "it's healthy even for deep-frying."

How much should you use?

The optimal amount can't be stated definitively. The amounts used in research were:

PREDIMED's amount is a lot by the standards of many diets, including Japan's. And oil is energy-dense, about 9 kcal per gram. One tablespoon is about 120 kcal. Thinking "the more, the healthier" is risky.

Realistically, the use most in line with the research is swapping olive oil in where you'd been using butter, margarine, mayonnaise and the like.

So what's actually special about olive oil?

Olive oil isn't "a magic oil that makes you healthy if you drink it."

What's still worth noting is that the small everyday choice of replacing fats like butter and margarine with olive oil is associated with lower risk in large studies.

Summary

Much of the information about olive oil emphasizes health benefits. But look at the papers, and it's a bit plainer, and a bit more practical.

For cardiovascular disease, a Mediterranean diet RCT and large US cohorts point in the same direction. For dementia and mortality, the evidence stays at the observational stage. And the heart of the effect may lie less in "olive oil itself" than in "what you use olive oil in place of."