Diet for brain health — an evidence-based guide.
There are two reasons people arrive at this question, and I see both. The first is that you have watched someone you love lose their memory — a mother who stopped recognising the kitchen she cooked in for forty years, a father who asked the same question six times in one afternoon — and somewhere inside that exhaustion, a quieter fear settled in about yourself. The second is nearer and smaller. You are in your forties or fifties, a colleague's name went missing mid-sentence last week, you walked into a room and could not retrieve why, and you have begun a private tally.
For deeper context, see: 5 Nutrition Changes That Actually Help with PCOS (From a Dietitian Who Gets It).
Let me take some weight off that tally. Word-finding lapses and a slower search for names are extremely common through midlife and beyond, and on their own they predict very little. Stress, broken sleep, perimenopause, an underactive thyroid, low B12, alcohol, depression and many ordinary medications all affect memory, and each is worth investigating before you conclude anything about your brain. Searching for a diet for brain health was a sensible instinct, and I will be straightforward about what eating can and cannot do. One thing belongs before any advice: memory change that is clearly worsening, that others have begun to notice, or that interferes with ordinary tasks — managing money, following a familiar route, using the cooker safely — belongs with your doctor rather than with a supplement purchase. That assessment is warm, unremarkable and worth having early, because several causes are treatable.
What a Diet for Brain Health Can Offer, and Why It Starts With Your Heart
I want to be straight about the ceiling first, because most of what is written on this topic sits above it. No diet prevents dementia. No food reverses it, and anyone selling a capsule with a glowing brain on the label — nootropics, memory formulas, blends of whatever is trending — is selling marketing, not medicine.
What the evidence supports is risk reduction, and I use those words carefully. In large observational research, people eating a particular broad pattern show slower cognitive decline and fewer dementia diagnoses over time. That is an association, not proof; it works over decades, not weeks; and how someone eats is never fully separable from how they sleep, move and live. The signal is consistent enough that I take it seriously anyway.
Here is the part that disappoints people and should not. The pattern with the most credible evidence is the familiar one: vegetables and plenty of them, especially leafy greens; berries and other deeply coloured fruit; pulses; wholegrains; olive oil; nuts; oily fish; and considerably less ultra-processed food than most of us eat. It is the same pattern I would recommend for your blood pressure, cholesterol and blood glucose — the version marketed online as the MIND diet is essentially this, renamed. That overlap is reassuring. You are not being asked to run two projects, one for your heart and one for your head.
The reason is mechanical. Your brain is fed by very small blood vessels, and what damages small vessels elsewhere damages these too. A meaningful share of what is diagnosed as dementia involves vascular damage, so the measurable, treatable things matter more than any single food. Blood pressure is probably the most important modifiable factor here, and treating it properly — with medication where your doctor advises it — does more for your brain than any supplement; the food side is in my guide to lowering blood pressure with the DASH approach. Long-standing high blood glucose is associated with higher dementia risk, which is one reason I take prediabetes seriously, and lipids and smoking belong in the same conversation. None of those is a food: they are systems, measured in a clinic, and eating supports the treatment without replacing it — alongside your doctor's care and never in place of it. [TK: confirm how you prefer to phrase the midlife blood-pressure-to-dementia-risk link, without a figure]
Greens, Berries and the Case for Colour on the Plate
Green leafy vegetables carry the most consistent signal of any single food group here. In observational research, people eating them most days show slower cognitive decline over the following years than people who rarely eat them. Spinach, rocket, kale, chard, cabbage, lettuce, purslane, and the parsley and dill most of us treat as decoration — all of it counts. The proposed mechanisms involve folate, vitamin K, lutein and nitrates, which is not a reason to swallow any of them as an isolated capsule.
Berries are the other group that keeps reappearing. Blueberries, blackberries, strawberries, sour cherries and pomegranate are rich in the pigments that give them their colour, and those pigments are associated in observational work with better memory performance over time. Frozen is fine and usually cheaper.
Beyond those two, the principle is colour and variety. Red cabbage, beetroot, aubergine skin, peppers, tomatoes, black grapes, walnuts, olives and pulses each bring different families of plant compounds, and variety across a week does more than perfecting any one item.
Oily Fish and Omega-3: The Food Evidence and the Capsule Gap
This is the most interesting nutrient story in the field, and I have to be honest about a gap in it. Oily fish appears repeatedly in research on cognitive ageing: people who eat it regularly tend to show slower decline than people who never do, and the long-chain omega-3 fats it contains are structural components of brain tissue. The mechanism is credible and the food-based association reasonably consistent. [TK: confirm whether you want to name a weekly oily fish portion target here]
The capsules are a different story. Trials giving omega-3 supplements to older adults to slow cognitive decline have been far less impressive than the food evidence — some neutral, some showing small effects in subgroups, and nothing that justifies the confidence with which they are sold. The supplement may arrive too late in life to matter, or the benefit may never have belonged to the fat in isolation. Nobody selling you a bottle knows either.
So eat the fish if you like fish — sardines, mackerel, anchovies, salmon, herring, trout, fresh or tinned, and tinned is no lesser version. If you do not eat fish at all, an omega-3 supplement is a reasonable conversation, particularly if you are vegetarian or vegan, and better had with a clinician than with an algorithm. My guide to omega-3 and healthy fats covers the food sources, and which supplements are worth taking is where I triage questions like this.
B12, Alcohol and the Gut: Three Places Where I Stay Careful
Some nutrient deficiencies genuinely affect memory, and this is where testing earns its place, because correcting a real deficiency helps while topping up an adequate level does not. Low vitamin B12 can produce confusion, poor concentration, low mood and fatigue, and it is reversible when caught. It is worth asking your doctor about it if you have eaten vegetarian or vegan long term, are older, take certain acid-reducing or diabetes medications, or have absorption problems such as coeliac disease. Folate is usually tested alongside it, and vitamin D is low in a great many people at northern European latitudes through winter, though the evidence tying it to dementia risk is weaker than the enthusiasm suggests. I am giving no doses deliberately: supplementing B12 blindly can obscure what is happening, and the right amount depends on your result and your medications. [TK: confirm which of B12, folate, ferritin, thyroid function and vitamin D you routinely ask clients to request from their GP or huisarts when memory is the complaint] [TK: clinical observation — what you most often find on bloods when a client comes to you worried about memory and brain fog]
Alcohol next, and without a sermon, because moralising changes nobody's drinking. Heavy, long-term drinking causes measurable cognitive harm, including forms of memory damage with their own diagnostic names, partly through direct effects on brain tissue and partly through thiamine deficiency. The middle of the range is far less settled: the old suggestion that a modest daily glass protects the brain has not held up, and the more carefully the research is done, the less of that effect survives. My summary for a reader in midlife is that less is better than more, that no intake can be promised as protective, and that drinking which has crept upward over a stressful decade is worth noticing without shame.
The gut, briefly and modestly. The work on the gut-brain axis is real and interesting, and the conclusions are thin: the leap from a shift in bacterial populations to a difference in someone's memory at seventy-five has not been made. What I am comfortable saying is that the pattern which appears to support cognitive ageing — fibre, variety, plants, fermented foods, less ultra-processed food — also supports a diverse microbiome, so you can eat for both at once. What I will not do is recommend a probiotic for your brain. [TK: confirm your current position on fermented foods and probiotics in cognitive health]
A Sample Day of Brain-Supportive Eating
Here is how these ideas come together on an ordinary day — illustrative rather than prescriptive, with portions that are yours.
- Breakfast: porridge with berries and a spoon of walnuts — or, on a savoury morning, eggs with spinach wilted in olive oil, wholegrain bread, tomatoes and olives.
- Lunch: a lentil or bean soup with wholegrain bread, and a salad of rocket and herbs with olive oil and lemon. Or last night's leftovers, which is a legitimate strategy.
- Afternoon: yoghurt with fruit, or fruit and nuts.
- Dinner: tinned sardines or grilled mackerel with a large plate of vegetables cooked in olive oil and a portion of bulgur, barley or brown rice. On other nights, a bean stew or chicken with greens.
Notice the pattern rather than the items: something green every day, colour at most meals, oily fish regularly, pulses often, olive oil throughout, wholegrains where you can manage them, and far less that arrives in a packet. That is the whole strategy, and it bends to your budget and week.
The Levers That Are Not Food At All
Food is not the only lever here, and several of these carry more weight than any nutrient above.
- Sleep. Poor, fragmented sleep is associated with worse cognitive outcomes, and untreated sleep apnoea especially — loud snoring, pauses in breathing and daytime exhaustion deserve a doctor's attention.
- Hearing loss. One of the most consistently identified modifiable risks, and the most overlooked. Having your hearing tested, and using aids if you need them, is a brain-health intervention rather than a vanity question.
- Physical activity. Movement supports blood pressure, glucose, sleep, mood and muscle at once, and it need not mean a gym. Walking, gardening and carrying your shopping count, and resistance work protects the muscle that keeps you independent — the subject of my post on nutrition for healthy ageing.
- Social and mental engagement. Conversation, learning, languages, music, volunteering, a card game with the same four people every week. Isolation is a risk factor in its own right.
- Treating what is treatable. Blood pressure, diabetes, depression, vision, thyroid function — unglamorous, and far better evidenced than anything in a supplement aisle.
Eating Well With Dementia or Mild Cognitive Impairment
If you are reading this because you are cooking for someone else, this section is for you, and I imagine you are tired.
Nutrition changes its job here. It is no longer about risk over decades but about weight, strength, comfort, dignity and pleasure this month. Unintended weight loss is very common as dementia progresses, for several reasons at once: meals forgotten, hunger no longer registered, changes in taste and smell, difficulty with cutlery, dental pain, and medications that blunt appetite. Raise weight loss and reduced intake with the medical team: some causes are fixable, and difficulty with swallowing needs assessment.
- Make food available instead of waiting for hunger. Meals and snacks offered at the same times, in the same place, work better than asking whether someone is hungry.
- Keep mealtimes calm and unhurried. One thing on the plate at a time, the television off, contrasting plate and food colours so the meal is visible as a meal.
- Let go of nutritional perfection. Someone losing weight needs energy and protein in a form they will accept: full-fat yoghurt, olive oil over everything, cheese, eggs, milk in soups, tahini. This is not the moment for a low-fat plate.
- Use the fact that sweet is often the last taste standing. Many people keep accepting sweet foods long after refusing savoury ones. Milk puddings, semolina, stewed fruit with cream, tahini with molasses, a milkshake made with fruit and full-fat yoghurt — all of it is legitimate nutrition, and a sweet dish with real energy in it beats nothing.
- Simplify the mechanics, and sit down with them. Finger foods that need no cutlery, softer textures, a weighted cup, food cut before it arrives. People eat more in company, and eating alongside someone changes the tone of the meal.
A word for you, the person doing the cooking. You are not failing when someone refuses the meal you made. Refusal is part of the condition and not a verdict on your care, and your exhaustion is proportionate to the work. Ask for help earlier than you think you need it. [TK: clinical observation — how you typically support family carers, and whether you see partners or adult children more often] [TK: confirm which memory-clinic and dementia-support routes you want named for the Netherlands, Belgium, Germany and Turkey]
Common Brain Health Diet Myths
"There are supplements that prevent dementia if you start early enough."
There is no credible evidence for this, for any product on the market. Some ingredients have plausible mechanisms and thin human data, a long way from prevention, and confident packaging is a marketing decision rather than a scientific finding. Correcting a deficiency found on a blood test is different.
"Coconut oil, turmeric or a specific superfood reverses memory loss."
Individual foods do not reverse cognitive decline, and claims like these are built from animal or laboratory findings that never translated. Turmeric is delicious, coconut oil is a fat, and neither is a treatment.
"Sugar causes Alzheimer's disease."
The honest version is more nuanced than either the claim or its dismissal. Long-standing high blood glucose and insulin resistance are associated with higher dementia risk, which is a real reason to care about your overall pattern of eating. But one food does not cause one disease, and framing it that way mostly produces guilt.
"If both my parents had dementia, nothing I do will matter."
Family history does raise risk and I will not pretend otherwise, but risk is not destiny. A meaningful share of dementia risk across populations is attributed to modifiable factors — blood pressure, hearing, smoking, activity, alcohol, social contact, and the eating pattern behind the vascular ones. [TK: confirm whether you want to reference the proportion of dementia risk considered modifiable, and in what phrasing]
Working With Hanzi Nutrition
I am a dietitian, and my honest role here is narrower and more useful than the internet's version. I cannot promise you protection from dementia, and I would not trust anyone who did. What I can do is help you eat in the way with the most credible evidence behind it for your heart and your head at once, support the treatable things — blood pressure, glucose, nutrient status, alcohol, sleep — and take the noise out of the subject.
At Hanzi Nutrition, I work fully online with clients across the Netherlands, Belgium, Germany and Turkey, in English and Turkish, and this sits inside my healthy lifestyle counselling service. We begin with a first consultation covering your eating pattern, your bloods, your medications, your sleep and your family history, without judgement. From there I build a personalised plan around your kitchen, your culture, your budget and your schedule, built from food you recognise rather than a list of powders. Where something needs medical assessment, I say so and coordinate with your doctor. And if you are caring for a parent, we can make that the focus instead, because keeping weight on someone you love is its own clinical work.
If you want to eat in a way that supports your brain for the next thirty years, or you are trying to keep a parent eating well today, book a consultation with me for evidence-honest nutrition support that works alongside your medical care.
Hanzi Nutrition offers dietitian-led nutrition counselling across the Netherlands, Belgium, Germany, and Turkey, fully online, in English and Turkish. This article is general education and not a substitute for individual medical care — no eating pattern prevents or treats dementia, and memory change that is worsening or affecting daily life needs assessment by a doctor. Please coordinate any changes to your nutrition, supplements, or treatment with your doctor.
Related Reading
More on this topic from the journal:
