Can Three Tablespoons of Extra Virgin Olive Oil Change Gestational Diabetes Care?

Study Overview

The paper, “OLIDIAG Study: Extra Virgin Olive Oil Supplementation in the Diet of Women with Gestational Diabetes Mellitus—A Randomized Clinical Trial,” was published in Nutrients in 2026 by Jawerbaum and colleagues. It was a multicenter, parallel randomized clinical trial conducted across eight centers in five Argentinian provinces and registered as NCT05120388.

The investigators enrolled 190 pregnant adult women with gestational diabetes before week 29 of pregnancy and randomized them 1:1 to standard care or standard care plus EVOO. The intervention was food-level, not capsule-level: three 13 mL tablespoons of extra virgin olive oil daily, about 36 g/day, taken uncooked with main meals. Both groups still received standard nutritional counseling, glucose monitoring, and insulin when clinically indicated.

Attrition matters here. The final analysis included 140 women with third-trimester data: 66 controls and 74 in the EVOO group. The paper states that a strict intention-to-treat analysis was not applied because participants who dropped out had no third-trimester data. That weakens the trial, but the baseline groups that remained were well matched for age, prepregnancy BMI, obesity, gestational age, fasting glucose, HbA1c, fructosamine, and triglycerides.

Key Findings: The Numbers That Matter

The cleanest primary result was not total pre-delivery insulin use, which fell from 54.6% to 41.9% but did not reach significance (RR 0.758, 95% CI 0.534–1.085, p = 0.135). The stronger result appeared after excluding women who had already been told to start insulin at the first clinic visit, before EVOO could plausibly work. During the actual intervention window, insulin initiation was 24.3% in the EVOO group versus 40.9% in controls.

Lipids moved more clearly. Triglycerides were 257.6 ± 72.9 mg/dL in controls and 214.3 ± 73.6 mg/dL in the EVOO group at term, a −43.3 mg/dL difference. The TG/HDL cholesterol ratio also improved, falling by roughly one unit after adjustment. Neonatal outcomes were exploratory but notable: neonates with more than one complication dropped from 19.7% to 6.76%, and NICU requirement fell from 18.2% to 6.8%.

Mechanism: Why EVOO Could Matter in GDM

Gestational diabetes is not just “high glucose during pregnancy.” It is a state of insulin resistance, placental metabolic stress, oxidative pressure, and inflammatory signaling. EVOO plausibly acts on several of those pathways at once. Oleic acid can improve dietary fat quality by replacing refined carbohydrate or saturated fat, while EVOO phenolics such as hydroxytyrosol, tyrosol, oleocanthal, and oleacein can reduce lipid oxidation and inflammatory tone.

The trial’s pattern fits that biology. Fasting glucose, fructosamine, and HbA1c did not meaningfully change at term, so EVOO did not behave like a glucose-lowering drug. Instead, the signal was metabolic load: less progression to insulin, lower triglycerides, a better TG/HDL ratio, and fewer complex neonatal complications. In pregnancy, triglycerides naturally rise; in GDM, excessive triglyceridemia is linked to fetal overnutrition and neonatal risk. Reducing that lipid burden is a credible route by which a dietary fat swap could matter.

Context: How This Compares With Previous Research

This study sits beside, not above, the broader Mediterranean-diet pregnancy literature. Earlier trials have suggested that Mediterranean-style diets enriched with EVOO and nuts can reduce the incidence of gestational diabetes. OLIDIAG asks a narrower downstream question: once GDM is already diagnosed, does adding EVOO to standard care improve maternal and neonatal markers?

It also complements the 2026 IMPACT BCN secondary analysis, where a Mediterranean-diet intervention with free EVOO and walnuts was associated with better fetal cardiac-function markers in high-risk pregnancies. Together, these papers make pregnancy one of the more interesting emerging lanes for olive-oil research. The caveat is important: these are diet interventions inside clinical systems, not isolated “olive oil cures.”

Limitations

• • Open-label design: participants knew whether they received EVOO, so behavior and reporting could shift.
• • Attrition: 50 of 190 randomized women lacked third-trimester data, and strict intention-to-treat imputation was not used.
• • Food intervention complexity: EVOO may have displaced other fats or foods, so the result is a dietary-pattern effect, not just a molecule effect.
• • Secondary neonatal outcomes: NICU and multiple-complication findings are clinically interesting but should be confirmed in larger trials.
• • Generalizability: control participants were not instructed to use EVOO and generally did not use it for cultural and economic reasons, which may differ from Mediterranean populations.

Our Take

This is a strong practical nutrition paper with one major weakness: attrition. The randomized multicenter design, predefined primary outcomes, clear dose, and clinically meaningful endpoints make it far more useful than another observational “Mediterranean diet is good” association. The insulin and triglyceride results are also directionally coherent, which increases confidence that the signal is not random noise.

Is it game-changing? Not alone. But it is unusually actionable. A 36 g/day EVOO intervention is feasible, cheap relative to many medical interventions, and aligned with broader cardiometabolic evidence. The responsible headline is not “olive oil treats gestational diabetes.” It is: in women already receiving GDM care, daily EVOO looked like a meaningful adjunct that reduced insulin initiation during the treatment window and improved lipid-linked risk. That is bookmark-worthy — especially because the downside of replacing poorer fats with high-quality EVOO is likely low when calories and pregnancy care are managed properly.

Reference

Jawerbaum A, et al. OLIDIAG Study: Extra Virgin Olive Oil Supplementation in the Diet of Women with Gestational Diabetes Mellitus—A Randomized Clinical Trial. Nutrients. 2026;18(7):1120. doi: 10.3390/nu18071120. PMID: 41978170. Full text: PMC13074905.