A reworked Mediterranean diet cuts type 2 diabetes risk by 31%, study finds

A large new observational study has found that a refined, plant-leaning version of the Mediterranean diet lowers the risk of developing type 2 diabetes by 31 per cent. The findings, summarised by Science Daily, represent one of the strongest numerical effects seen in dietary research in recent years.

The trial enrolled 16,000 adults and followed them for seven years. Participants were aged 40 to 65 at enrolment, had an average BMI of 27.4, and none had a type 2 diabetes diagnosis at baseline. Across the follow-up window, 1,247 new diagnoses were recorded.

Compared with the traditional Mediterranean pattern, the reworked plan in the study tilted further toward legumes, whole grains, leafy greens and oily fish, and pulled back more visibly on red meat and wine intake.

The authors summarise the approach as "more plants, less plate." The classical olive-oil base was preserved, but daily vegetable variety was increased and oily fish was prescribed at least twice a week.

Lead author Dr Maria Rosario Carmona said in the published commentary: "The takeaway here is direct: small plate adjustments can produce major metabolic effects. A 31 per cent relative risk reduction over seven years is meaningful from both a clinical and a public-health standpoint."

The risk reduction was especially pronounced in participants with a baseline BMI above 28. In that subgroup, new diagnosis incidence was 14.2 per 1,000 person-years compared with 20.6 in the control arm.

Mechanistically, the researchers point to the role of low-glycaemic-index foods in insulin sensitivity. The omega-3 fatty acids in oily fish are also flagged for a possible protective effect on pancreatic beta cells, though that association is correlational rather than causal at this stage.

An independent commentary from Prof Helene Almqvist, a scientific committee member of the European Association for the Study of Diabetes, called the work "important evidence that dietary intervention can reach effect sizes comparable to metformin."

Limitations are acknowledged. Eighty-six per cent of the participant pool was of European ancestry, so generalisation to global populations needs caution. The study also relied on self-reported dietary intake, with the attendant measurement error.

The study was published in the BMJ in its May 2026 issue. Its potential effect on clinical guidelines is still under review. The EASD has said it will discuss possible dietary-recommendation updates at its September congress.