Healthy eating: what the evidence says and how to apply it

What the WHO means by healthy eating

The World Health Organization (WHO) points out that an unhealthy diet is one of the main risk factors for disease and disability, and that eating healthily throughout life helps prevent malnutrition in all its forms and noncommunicable diseases such as diabetes, heart disease, stroke and cancer.¹

It doesn't propose a single menu. According to the WHO, the exact makeup of a healthy diet varies with age, sex, lifestyle, physical activity, culture, the foods available and local customs, but it always rests on four principles:¹

• Adequacy: meeting nutrient needs, without overdoing it, to avoid deficiencies.
• Balance: matching the energy you eat to the energy you use, with an appropriate split between protein, fat and carbohydrates.
• Moderation: limiting the nutrients, ingredients and foods that can harm health.
• Diversity: eating a wide variety of nutritious foods, within each food group and across groups.

The foundation, the WHO adds, is a variety of foods that are minimally processed or unprocessed, with little unhealthy fat, few free sugars and little sodium. For the population it also gives some reference figures: from age 10, at least 400 grams of fruit and vegetables and at least 25 grams of fiber from foods a day; less than 10% of energy from free sugars; no more than 10% from saturated fat; and, in adults, less than 5 grams of salt a day.¹

The reference in Spain: the AESAN report

In 2022, the Comité Científico de la AESAN updated the dietary recommendations for the Spanish population, taking into account, in addition to health, the environmental impact of foods. It starts from a practical principle: people choose foods, not nutrients, so the recommendations are expressed mainly in terms of foods; and foods are combined within a pattern, in which they can reinforce each other.²

Its conclusion is that a varied and balanced pattern, based mainly on plant foods with a smaller share of animal foods, can improve the population's health and wellbeing while reducing environmental impact. The committee frames its recommendations within the traditional Mediterranean diet, because it's the predominant pattern in Spain and because, in its view, the evidence of its beneficial effect is very solid; it also takes into account the recipes of the Atlantic diet of northwestern Spain.²

The report also suggests that, of the protein-rich foods at lunch and dinner, one be plant-based, especially legumes, and the other fish, eggs, dairy or meat.² In the Mediterranean pattern, beyond olive oil we explain what characterizes that pattern beyond any particular food, what trials and cohorts have been able to show, and in which people it was studied.

Why the pattern matters more than a single food

For decades, nutrition was studied one nutrient at a time. But nobody eats nutrients in isolation: people eat foods in combination, day after day. That's why guidelines have gradually shifted toward dietary patterns, that is, the quantities, proportions, variety and combination of foods and drinks in the usual diet. That is the definition used by the scientific committee that advised on the Dietary Guidelines for Americans, 2020–2025; the committee justified the shift on the grounds that analyzing isolated nutrients or foods can't capture how they interact with one another.³

That committee carried out a systematic review to find out which dietary patterns were associated with all-cause mortality. It brought together 153 articles, with more than 6.5 million people from 28 countries with a high or very high Human Development Index, Spain among them: one randomized trial and 152 observational studies.³ The result was very consistent. The patterns associated with lower mortality had more vegetables, fruit, legumes, nuts, whole grains, unsaturated vegetable oils and fish, and lean meat or poultry when they included meat; and less red and processed meat, high-fat dairy, refined grains and sweets.³

The name mattered little. The patterns studied were called Mediterranean, “prudent,” DASH or plant-based, or were measured with various diet quality indices, and the authors conclude that what's associated with better health is the quality of the diet, whatever its label. They see this as an advantage: it leaves room to adapt food combinations to each person's tastes, culture and budget.³ The WHO puts it in its own way: healthy diets come in many forms.¹

Eating and aging in good health: what long-term studies show

Living longer isn't the only thing that matters: reaching old age still able to do everyday things counts too, an idea we develop in the guide to living more years in good health. Some recent studies measure exactly that.

One, published in 2025, followed 105,015 US nurses and health professionals for up to 30 years; in 1986 they were between 39 and 69 years old and had none of 11 major chronic diseases. Their diet was recorded several times with questionnaires. A person was considered to have aged healthily if they reached age 70 without those diseases and without decline in their cognitive, physical and mental function, according to validated questionnaires the participants answered themselves.⁴

Only 9.3% did. All eight healthy dietary patterns compared, one of them Mediterranean, were associated with greater chances of healthy aging. The strongest association was with the Alternative Healthy Eating Index (AHEI), designed to predict the risk of chronic diseases; the weakest, with a healthful plant-based diet index. By food, eating more fruit, vegetables, whole grains, unsaturated fats, nuts, legumes and low-fat dairy was associated with greater chances, and eating more trans fats, sodium, sugary drinks and red or processed meats, with lower chances.⁴

In the analyses that took those and other factors into account, the fifth whose diet most closely matched each pattern had 45% to 86% higher odds of healthy aging than the fifth whose diet matched it least (odds are a way of expressing probability that these studies use; they're not exactly the same as “86% more likely”). High consumption of ultra-processed foods, by contrast, was associated with 32% lower odds.⁴ These are relative figures, and a relative figure says little without the starting point; we explain how to read them in relative and absolute risk.

Why headlines about a single food are so confusing

If patterns have so much backing, why does a different food seem to save us or harm us every week? A 2013 study didn't look at headlines but at the scientific articles themselves, and it helps explain where they come from: its authors chose 50 ingredients from recipes picked at random from a cookbook and searched for recent studies on their relationship with cancer risk. They found at least one for 40 of those ingredients.⁵

Of 264 assessments made in individual studies, 72% concluded that the food was associated with higher or lower risk, but three out of four risk estimates had weak or no statistical support, and significant results made it into the article's abstract more often than non-significant ones, which more often appeared only in the full text. Meta-analyses, which combine several studies, gave more cautious results: on average, no effect.⁵ The authors add that many single studies show implausibly large effects.⁵

When you see a headline about a food, three questions help; they're a deduction from the above:

1. Is it a single study or a review of many? A conclusion about patterns that's repeated across dozens of studies carries more weight than an isolated result.
2. Was it done in people? A compound in a food having an effect in cells or in mice doesn't show a benefit from eating it; we explain why in what animal studies can and can't tell us.
3. What changes in absolute numbers, and at the cost of what? “20% lower risk” can be very little if the starting risk was low, and eating more of one thing usually means eating less of something else.

Packaged foods: compare rather than rule out

Much of what the recommendations ask us to limit comes in ready-made products. According to the WHO, in many countries most salt comes from processed foods, such as ready meals, processed meats, cheese or salty snacks, or from foods eaten often and in large amounts, such as bread. And diets with a large amount of highly processed foods, which tend to be high in sodium, sugar or unhealthy fats, are associated with worse health outcomes.¹

AESAN's committee, for its part, attributes the harm of replacing unprocessed foods with processed ones mainly to their ingredients, more than to processing techniques, and proposes two things: favoring home cooking and, when buying processed foods, choosing those with less salt, less sugar and less fat other than olive oil.² Choosing means comparing, and in how to read food labels we explain how to compare labels without judging a food by a single number.

Water: the go-to drink, with no single figure

AESAN recommends that water be the main drink and that people drink as much as they need. Its report points out that the water that enters the body isn't only what you drink: water from other drinks, from cooked dishes and from foods themselves also counts. As a general reference for adults it cites the adequate intakes set by the European Food Safety Authority (EFSA): 2.5 liters (about 85 fl oz) of total water a day for men and 2 liters (about 68 fl oz) for women, counting the water in food, not just what you drink.²

That figure doesn't apply equally to every day or to every person: you need more water with physical activity, in heat or humidity, with a fever or when you lose fluids. With age, moreover, the sense of thirst changes, and dehydration is a significant problem in older people.² We explain why neither “eight glasses” nor “two liters” works as a fixed rule in how much water to drink a day. If you've been told to limit fluids because of an illness or a treatment, follow that advice.

Bringing it into a real week

The recommendations say what to eat; everyday life sets the conditions: time, budget, tastes, who does the cooking. A few details from the AESAN and WHO documents themselves remove common obstacles:

• Legumes can be cooked in bulk and frozen. Canned or jarred legumes have an environmental impact similar to those cooked at home, and the report advises choosing low-salt ones. If you're not used to eating them, it's best to increase them gradually so you tolerate them well.²
• Frozen and canned fish have a nutritional value equivalent to fresh fish; among canned products, those with little salt are the better choice.²
• Frozen or canned fruit and vegetables are also a good option if they have no added sugars or excess sodium.¹
• Cooking larger amounts at once and freezing saves time and energy.²

The rest is organization, and here what follows is our own deduction, not a result of these studies. Instead of changing your whole diet at once, it seems more realistic to choose one small, specific change you can repeat—one more serving of legumes a week, fruit where you can see it, a pitcher of water on the table—and keep it up before adding another. How to choose that first change is in the guide to sustainable habits; what backing there is for placing foods closer or farther away, in the section on setting up your environment; and how to link a new habit to something you already do, such as having fruit when you finish a meal, in habit stacking.

What this guide doesn't do

This guide isn't a diet: it doesn't propose weight or calorie targets, menus, supplements or “superfoods.” Nor does it sort foods into “good” and “bad.” What has backing is the whole of what you eat over time, and that whole allows for many combinations.³¹

The WHO and AESAN references are for the general population. If you have a chronic illness, take medication, are pregnant or breastfeeding, have or have had a difficult relationship with food, or are considering a major change in your diet, talk to a health professional first. This guide is general information and doesn't replace an individual assessment.