Why Did an Olive-Oil Polyphenol Pattern Track With Higher Heart Risk?

Study Overview

The paper, “Sex differences in (poly)phenol intake patterns and cardiovascular risk in Spanish adult population at high risk of cardiovascular disease: Cross-sectional findings,” was published in Nutrition in 2025 by Rubín-García, Vitelli-Storelli, Toledo, Álvarez-Álvarez, Martínez-González and colleagues from the PREDIMED-Plus consortium.

This was not a randomized olive-oil feeding trial. It was a cross-sectional baseline analysis in 6,633 Spanish adults already at high cardiovascular risk. The researchers estimated intake of 26 polyphenol classes and subclasses, then used factor analysis to identify dietary polyphenol patterns in the total sample and separately in men and women. Estimated absolute cardiovascular risk was calculated using the Framingham equation.

The statistical model then asked whether people in higher quintiles of each polyphenol pattern had different estimated cardiovascular risk than people in the lowest quintile. The key olive-oil-relevant pattern was “pattern three,” dominated by olives and olive oil. This is clinically useful because it tells us how olive-oil-associated eating clustered in a real high-risk population, not how pure extra virgin olive oil behaves under controlled conditions.

Key Findings: The Actual Numbers

On the surface, the result is uncomfortable: the olives-and-olive-oil polyphenol pattern was positively associated with estimated cardiovascular risk in the whole sample and in men. The authors also report that this pattern exhibited higher diabetes prevalence, more smoking and higher salt consumption. In men, the same pattern again clustered with diabetes and higher sodium intake.

That surrounding context is not a footnote. It is probably the story. In a high-risk Spanish cohort, people with diabetes, hypertension or existing medical advice may have consciously shifted toward traditional Mediterranean foods such as olive oil and olives. If so, the diet pattern is partly a marker of already being high risk, not a cause of the risk.

Mechanism: What Could Be Biological, and What Is Probably Behavioral?

Biologically, extra virgin olive oil remains plausible as a cardioprotective food. Its oleic acid can improve replacement-fat quality; phenolics such as hydroxytyrosol and tyrosol support antioxidant defenses; and higher-quality EVOO has been linked in controlled studies to lower LDL oxidation, improved endothelial function and better postprandial lipid handling. Those mechanisms do not vanish because one cross-sectional paper reports a positive association with estimated risk.

But olives are not only polyphenol packages. Table olives can be salty. Olive oil can add calories when poured on top of an already energy-dense diet. And a person can eat plenty of olive oil while still smoking, eating excess sodium, having diabetes or carrying abdominal obesity. In this analysis, the olive-related pattern did not travel alone; it traveled with a riskier cardiometabolic profile.

The most convincing mechanism here is therefore not “olive oil raises risk.” It is reverse causality plus clustering. Higher-risk people may adopt olive-oil-forward dietary advice after diagnosis, while their Framingham score remains high because age, diabetes, smoking, blood pressure and lipid history still dominate the equation. This is why randomized trials and prospective biomarker analyses matter.

Context: Does This Contradict PREDIMED?

It does not overturn PREDIMED-style evidence. The original PREDIMED randomized trial tested Mediterranean diet advice supplemented with extra virgin olive oil or nuts against control dietary advice and found fewer major cardiovascular events in the Mediterranean groups. Later PREDIMED analyses using urinary polyphenol signatures and cumulative EVOO intake have generally supported the idea that phenolic-rich Mediterranean eating patterns are associated with lower cardiovascular risk.

This new paper answers a different question: among people who are already high risk, what do polyphenol intake patterns look like at one point in time, and how do those patterns correlate with estimated Framingham risk? A cross-sectional pattern can easily capture treatment-seeking behavior. Someone with diabetes may be told to eat more olive oil and vegetables, but the dietary change does not erase the fact that diabetes itself raises Framingham risk.

In that sense, the study is valuable because it pushes back against simplistic “more polyphenols equals lower risk” storytelling. Polyphenol source, dietary context, sex, baseline disease and behavior all matter. Olive oil is not magic dust; it is one food inside a whole pattern.

Limitations

• • Cross-sectional design: diet and cardiovascular risk were assessed at the same broad time point, so causality cannot be inferred.
• • Estimated risk, not events: the outcome was Framingham-estimated absolute cardiovascular risk, not heart attacks, strokes or cardiovascular deaths.
• • High-risk population: these were Spanish adults in PREDIMED-Plus, so results may not generalize to younger or lower-risk consumers.
• • Dietary measurement error: polyphenol intake estimates depend on dietary reporting and food-composition databases.
• • Pattern interpretation: factor-analysis patterns are useful but not the same as testing one food in isolation.
• • Confounding and reverse causality: diabetes prevalence, smoking and salt intake clustered with the olive-related pattern, making causal interpretation especially risky.

Our Take

This is not a “gotcha” paper against olive oil. It is better than that. It is a reminder that observational nutrition can punish foods that high-risk people start eating after they become high risk. If a doctor tells a patient with diabetes to follow a Mediterranean diet, that patient may report more olive oil while still carrying the diabetes, blood-pressure and lipid burden that drives their Framingham score.

The study is strongest as a cautionary epidemiology paper. It shows that polyphenol patterns are sex-specific and behavior-specific, and that “olive oil intake” can mean very different things depending on the person, the rest of the plate and the medical context. It is weak as evidence about olive oil causing harm, because the design simply cannot answer that question.

Our bottom line: still choose fresh extra virgin olive oil, preferably high-polyphenol, as your main culinary fat. But do not use it as a health halo. If the meal is salty, calorie-heavy, low-fiber or paired with smoking and unmanaged diabetes, olive oil will not rescue the risk profile by itself.

Reference

Rubín-García M, Vitelli-Storelli F, Toledo E, Álvarez-Álvarez L, Martínez-González MÁ, Corella D, Salas-Salvadó J, et al. Sex differences in (poly)phenol intake patterns and cardiovascular risk in Spanish adult population at high risk of cardiovascular disease: Cross-sectional findings. Nutrition. 2025 Dec;140:112919. doi: 10.1016/j.nut.2025.112919. PMID: 40876090.