Introduction

Olive oil is the cornerstone in the Mediterranean diet, which is also abundant in plant foods. High adherence to the Mediterranean diet has been associated with lower incidence and mortality from cardiovascular disease (CVD) (1) and cancer (2). For CVD, the association with the Mediterranean diet appears most attributable to olive oil, fruit, vegetables, and legumes (1). The PREDIMED (PREvención con DIeta MEDiterránea) randomized clinical trial, which enrolled 7,447 Spanish adults at high cardiovascular risk (3), showed that a Mediterranean diet supplemented with either extra virgin olive oil (the goal was to consume 50 g [approximately 4 tablespoons] or more per day) or mixed nuts significantly decreased the risk of the composite endpoint of cardiovascular death, myocardial infarction, and stroke by about 30% compared with the control diet (advice on a low-fat diet) (3). The associations were mostly driven by a reduction in stroke risk (3), but a post hoc analysis of the PREDIMED trial showed that the Mediterranean diet supplemented with extra virgin olive oil also conferred a reduction in atrial fibrillation risk (4). Even though the PREDIMED trial could not distinguish between the effects of olive oil or nuts from other foods that were recommended for the 2 Mediterranean diet groups, a main difference between those groups and the control group was the increase in extra virgin olive oil and nut consumption in the Mediterranean diet groups (3). In an observational analysis of the PREDIMED study, participants in the highest category of baseline consumption of total olive oil (mean 56.9 g/d) and extra virgin olive oil (mean 34.6 g/d) had a significant reduced risk of total CVD incidence and mortality but not cancer and all-cause mortality compared with those in the lowest category of olive oil consumption (5). Although the evidence is convincing regarding a beneficial role of the Mediterranean diet on CVD risk, data on olive oil consumption specifically in relation to all-cause and cause-specific mortality in populations with a low average olive oil consumption are limited.

In this issue of the Journal, Guasch-Ferré et al (6) report results from a study of olive oil consumption and risk of all-cause and cause-specific mortality in 2 cohorts of >90,000 U.S. women and men. In this well-designed study, with long-term follow-up and repeated measurements of dietary intake and other risk factors for diseases, participants who reported the highest olive oil consumption (>0.5 tablespoon/day or >7 g/d) had 19% lower risk of all-cause mortality, 19% lower risk of CVD mortality, 17% lower risk of cancer mortality, 29% lower risk of neurodegenerative disease mortality, and 18% lower risk of respiratory disease mortality compared with those who never or rarely consumed olive oil after adjustment for known risk factors and other dietary factors. The risk of all-cause mortality and mortality from CVD, cancer, and neurodegenerative diseases was significantly reduced already at a daily olive oil consumption of >0 to ≤1 teaspoon (median 1.5 g/d). This small amount of olive oil was associated with a 12% reduction in the risk of all-cause mortality. The authors subsequently performed substitution analyses and found that replacement of margarine, butter, mayonnaise, and dairy fat with olive oil was associated with a reduced risk of mortality. However, substituting olive oil for other vegetable oils (eg, canola, corn, safflower, and soybean oil) did not confer a reduced mortality risk. This suggests that vegetable oils may provide the same health benefits as olive oil.

A major challenge of this type of observational study is residual confounding. Despite adjustment for potential confounders, it cannot be inferred whether the observed associations of small to moderate amounts of olive oil consumption with reduced risk of all-cause and cause-specific mortality are causal or attributed to confounding. Furthermore, the biological mechanisms underpinning the observed associations are somewhat elusive, particularly for low levels of olive oil consumption and for non-CVD mortality. Olive oil and other vegetable oils contain high amounts of monounsaturated fatty acids (MUFAs), particularly oleic acid. Nevertheless, meta-analyses of observational studies have found no beneficial effects of increased circulating levels or intake of MUFAs or oleic acid on CVD (7-9), and instead found an increased risk of coronary heart disease with high circulating MUFA levels (9). Olive oil is also a source of phenolic compounds that could confer cardiovascular benefits and have been shown to possess anticarcinogenic properties in in vitro and animal studies (10). The PREDIMED trial showed that the group assigned to the Mediterranean diet supplemented with extra virgin olive oil had a significant lower risk of breast cancer compared with the control group (11). The inverse association between olive oil consumption and total cancer mortality in the current U.S. study was observed in both women and men, indicating that the association is not confined to a potential reduction in breast cancer mortality but might also apply to major causes of cancer-related deaths in men, such as deaths from lung, prostate, and colorectal cancer. A study in male rats with azoxymethane-induced colon cancer showed that dietary olive oil suppressed the development of colon carcinoma and that the effect may be explained by arachidonic acid metabolism and local prostaglandin E₂ synthesis (12).

The findings for CVD mortality in the study by Guasch-Ferré et al (6) are complementary but not directly comparable with the results of the PREDIMED trial on major cardiovascular events (3). First, the amount of olive oil consumed in the Spanish and U.S. populations differed remarkably. In the PREDIMED trial, participants had a mean baseline extra virgin olive oil and refined/mixed olive oil consumption of 20-22 g/d and 16-18 g/d, respectively (3), and participants assigned to a Mediterranean diet with extra virgin olive oil substantially increased their consumption of extra virgin olive oil (to 50 g/d) (3). In the U.S. study, the mean baseline consumption of any olive oil in the highest category (>0.5 tablespoon/day) was about 9 g/d (6). Second, in the PREDIMED trial, participants were supplied with polyphenol-rich extra virgin olive oil. The U.S. study could not distinguish between different olive oil varieties. This distinction is important because refined olive oil has much lower levels of phenolic compounds than extra virgin olive oil and may therefore have fewer health benefits. Third, participants of the PREDIMED trial were at high cardiovascular risk, whereas the U.S. study included nurses and health professionals with a relatively low cardiovascular risk. In the U.S. study, results were, however, largely consistent in subgroups with and without major CVD risk factors, suggesting that olive oil consumption might be beneficial independent of cardiovascular risk.

A novel finding of the study by Guasch-Ferré et al (6) is the inverse association between olive oil consumption and risk of neurodegenerative disease mortality. Alzheimer’s disease is the major neurodegenerative disease and the most common cause of dementia. In a sensitivity analysis, the authors found a significant 27% reduction in risk of dementia-related mortality for those in the highest vs lowest category of olive oil consumption. Considering the lack of preventive strategies for Alzheimer’s disease and the high morbidity and mortality related to this disease, this finding, if confirmed, is of great public health importance. Another novel finding of the current study was the inverse association of olive oil consumption with risk of respiratory disease mortality. Because the mechanism behind this association is unclear and residual confounding from smoking cannot be ruled out, this finding is tentative and requires confirmation in a study that is less susceptible to confounding, such as a randomized trial.

To summarize, the current study and previous studies have found that consumption of olive oil may have health benefits. However, several questions remain. Are the associations causal or spurious? Is olive oil consumption protective for certain CVDs (eg, stroke [3,13] and atrial fibrillation [4]) only or also for other major diseases and causes of death? What is the amount of olive oil required for a protective effect? Is the potential effect related to MUFAs or phenolic compounds, ie, is the protective effect confined to polyphenol-rich extra virgin olive oil or are refined olive oil and other vegetable oils as beneficial? More research is needed to address these questions.