Which Foods Are Associated With Longer Life?

1. Five food groups showed consistent associations with lower mortality.

Higher consumption of nuts, whole grains, fruits, vegetables and fish was associated with lower all-cause mortality in both higher-versus-lower comparisons and dose-response analyses.

The most recent and comprehensive analyses reported:

nuts: 20% lower mortality with higher versus lower intake

whole grains: 12% lower mortality

fruits: 11% lower mortality

vegetables: 7% lower mortality

fish: 5% lower mortality

These percentages should not be interpreted as the effect of simply adding one food to an otherwise unchanged diet.

They describe associations observed across prospective populations.

2. Nuts and whole grains showed particularly strong associations.

For nuts, each standardized 28 g daily serving was associated with approximately 24% lower all-cause mortality.

For whole grains, each additional 30 g per day was associated with approximately 6% lower mortality.

The underlying meta-analyses for both food groups were rated as high methodological quality.

These findings do not mean that increasing intake by exactly these amounts will produce the corresponding reduction in an individual person’s risk.

They demonstrate the direction and magnitude of the associations observed across populations.

3. Processed meat, red meat and sugar-sweetened beverages showed associations in the opposite direction.

Higher consumption of processed meat, unprocessed red meat and sugar-sweetened beverages was associated with higher all-cause mortality.

Higher versus lower consumption was associated with approximately:

14% higher mortality for processed meat

10% higher mortality for unprocessed red meat

11% higher mortality for sugar-sweetened beverages

Dose-response analyses showed the same overall direction.

4. Not every food group showed a clear association.

Higher consumption of legumes and white meat tended to be associated with lower mortality, although the evidence was less consistent.

For dairy products and refined grains, there was no clear association with all-cause mortality.

Eggs and added sugars showed tendencies toward higher mortality in some analyses, but these findings were not statistically significant in the principal higher-versus-lower comparisons.

Absence of a clear mortality association should not be interpreted as evidence that a food is either beneficial or harmful.

The findings provide an important bridge between broad dietary patterns and everyday food choices.

Mediterranean, Nordic and other healthy dietary patterns differ culturally, but many share the same underlying foods.

Whole grains, vegetables, fruits, nuts and fish repeatedly appear within dietary patterns associated with better long-term health.

At the same time, higher consumption of processed meat, red meat and sugar-sweetened beverages was associated with greater mortality.

This suggests that healthy aging may depend less on following one named diet and more on the combination of food groups that make up the diet over many years.

This umbrella review provides one of the broadest recent evaluations of individual food groups and all-cause mortality.

It was prospectively registered, followed PRISMA methodology and evaluated the methodological quality of the included reviews using AMSTAR-2.

An important strength is the consistency seen for several food groups across both high-versus-low and dose-response analyses.

However, the evidence has important limitations.

Although the umbrella review considered systematic reviews that could potentially include different study designs, the mortality evidence ultimately included in the analysis came from cohort studies rather than randomized trials.

The findings therefore show associations, not proof of causation.

People who regularly eat more nuts, whole grains, fruits or vegetables may also differ in physical activity, smoking, socioeconomic circumstances, body weight and other health behaviors.

Statistical adjustment can reduce these differences but cannot completely eliminate residual confounding.

There was also substantial heterogeneity in many analyses. More than 80% showed an I² above 50%.

This may reflect differences between populations, dietary assessment methods and the foods included within broad categories.

This is particularly relevant because a food group is not a single food. Different vegetables, dairy products, grains or meats can differ considerably in composition and preparation.

The review therefore provides strong evidence about patterns of association across food groups, but it should not be used to assign a precise survival effect to an individual food.