Rye bread, berries and fish against low mood: Finns who ate the most Nordic way reported fewer depressive symptoms. Followed for 11 years, the link faded once smoking, exercise and weight were counted

Most of what is known about diet and depression comes from the Mediterranean. Pooled studies the new paper cites put the risk of depression about a third lower in people who eat closest to that pattern. But olive oil and fresh tomatoes are not what grows in Finland, and researchers there have long asked whether a diet made from northern crops would do as well. The Nordic diet has the same outline as the Mediterranean one, with plenty of plants, some fish, not much red meat or sugar, and it swaps in rye, oats and barley, cabbage and root vegetables, bilberries and lingonberries, apples, salmon and rapeseed oil.

The study, published in the European Journal of Nutrition on 7 October 2026, comes from the University of Eastern Finland, the Finnish Institute for Health and Welfare and colleagues in Tampere, Kuopio and Australia. It uses Health 2000, a survey that drew a national sample of Finnish adults aged 30 and over in 2000 and 2001 and went back to them in 2011 and 2012. People filled in a 125-item questionnaire about what they had eaten over the past year, and each got a Nordic diet score from 0 to 22. Points came from fruit, vegetables, whole grains, low-fat milk, fish, a better balance of unsaturated to saturated fat, less red and processed meat, and keeping alcohol modest.

The first result is a snapshot. Among 4,697 adults with no clinical depression in 2000, 1,746 (37%) counted as having depressive symptoms: they said they had felt down in the past month, felt unhappy in the past few weeks, and scored at least one point on a depression questionnaire. People in the top quarter of the diet score had 25% lower odds of being in that group than people in the bottom quarter, after allowing for age, sex, calories, education, marital status, smoking, exercise and body weight. This is the paper's firmest finding. It is also the one that cannot say which came first. People who feel low cook less, shop differently and drink more, so a link measured at a single moment runs in both directions.

The 11-year follow-up was meant to sort that out. The researchers took the people who had no symptoms in 2000 and looked at who had them in 2011. Of 1,788 people, 994 (56%) scored at least 1 on a 13-question depression scale, 160 (9%) scored at least 5, and 46 scored 9 or more. A higher diet score went with lower odds at the two lower thresholds, but only until smoking, exercise and body weight were added to the sums. With those in, neither link could be told from chance (p = 0.061 and 0.138). At the highest threshold there was nothing in any version of the analysis. For clinical depression, diagnosed by interview or reported as diagnosed or treated by a doctor, there were 82 new cases among 1,862 people and no link with the diet score.

The authors read this as a sign that diet may matter more for the milder end of low mood, and they suggest the clinical result may be down to having only 82 cases. They are also candid in the discussion: the fading after adjustment 'suggests that these behaviors, such as physical activity, smoking, and BMI, may partly explain the association', and healthy habits tend to travel together. They note that more than 1,100 people had no follow-up data on symptoms, and that those who drop out of long studies are often the ones in worse health.

The rest is our reading. The headline word 'mild' needs unpacking. The lowest threshold is a score of 1 on a scale that runs to 39, and more than half of a group that was symptom-free in 2000 crossed it. That is closer to 'ticked one box' than to anything most people would call depression. The tables also need a careful eye. The rows labelled 'depressive symptoms (%)' show each quarter's share of all the cases, not how common symptoms were within the quarter. Working it out from the paper's own counts, 53% of the bottom quarter and 55% of the top quarter crossed the lowest threshold in 2011. The top quarter was about six years older on average, which is why the adjusted odds point the other way, but the raw gap is hardly there. In the 2000 snapshot the same sum gives 40% in the bottom quarter and 33% in the top, with the two middle quarters at 37% and 38% and neither clearly different from the bottom. So the benefit, such as it is, shows up only among the people with the very highest scores. And one check in the paper sits oddly with the rest: after removing just eight people who were on psychiatric medication, the fully adjusted estimate for the lowest threshold moved from 0.972 to 1.002, which is no link at all, and the paper describes this as similar to the main result. None of this makes rye bread and fish a bad idea. It means this study, with one diet questionnaire filled in a quarter of a century ago, shows that Finns with the best diets were somewhat less likely to feel low at the time, and does not show that the diet protected anyone over the following decade. The work was funded by two Finnish foundations and the authors declare no competing interests.