Foods linked to higher dementia risk (and smarter swaps to make instead)

Short answer: No single food causes dementia, and no food prevents it. But large studies repeatedly link a few eating patterns to a higher risk of dementia: ultra-processed foods and processed meats most consistently, and sugary drinks and added sugar on a more mixed evidence base. The good news is that these are modifiable — and the research suggests the biggest wins come not from banning anything forever, but from swapping the riskiest items for minimally-processed ones a few times a week. Below is what the evidence actually shows, and one realistic swap for each.

This is educational, not medical advice. Diet is one of several modifiable factors in dementia risk — it can't guarantee any outcome, and food changes should fit your own health needs. Talk to your clinician or a dietitian before making big dietary changes, especially if you manage diabetes, blood pressure, or cholesterol.

First, the honest framing

Nutrition headlines love the phrase "worst foods for your brain." It's catchy, but it overpromises in both directions. The studies below are observational — they follow large groups of people over years and find associations, not proof that a food directly damages neurons. People who eat a lot of ultra-processed food also tend to differ in many other ways, and researchers can only adjust for so much.

So read this as: these patterns are repeatedly linked to higher risk, the associations are strong enough to act on, and the "action" is cheap and low-risk — eat a little less of these, a little more of the alternative. That's it. No fear, no forbidden list.

1. Ultra-processed foods (the biggest pattern)

Ultra-processed foods (UPFs) — packaged snacks, sodas, instant noodles, mass-produced bread, many ready-meals — are the most consistent dietary signal in recent research. In a UK Biobank cohort of 72,083 adults aged 55 and over, followed for a median of ten years (518 of whom developed dementia), each 10% higher share of ultra-processed food in the diet was associated with a 25% higher relative risk of dementia — and, importantly, replacing 10% of that ultra-processed food with unprocessed or minimally-processed food was associated with a 19% lower risk (Li et al., Neurology 2022). A 2024 systematic review pooling ten observational studies (867,316 adults) found high versus low UPF intake associated with a 44% higher relative risk of dementia — but the studies disagreed with each other a great deal, so that pooled figure is better read as a rough signal than a precise number (Henney et al., Journal of Neurology 2024). That review also includes the UK Biobank cohort above, so it isn't independent confirmation of it.

Smarter swap: you don't have to cook from scratch every night. Pick one UPF you eat daily — the afternoon packaged snack, the soda, the breakfast pastry — and replace it with a minimally-processed version (fruit and nuts, sparkling water, plain yoghurt with oats). The substitution studies suggest that's where the benefit lives.

2. Processed meats

Processed meats — bacon, ham, sausages, deli slices, hot dogs — are one of the most-studied ultra-processed categories, with their own dedicated evidence base. In a UK Biobank cohort of 493,888 people followed for eight years, each 25 g of processed meat per day — about one rasher of bacon — was associated with a 44% higher relative risk of dementia, and a 52% higher relative risk of Alzheimer's disease specifically (Zhang et al., Am J Clin Nutr 2021).

Keep those numbers in proportion. Of those 493,888 people, 2,896 developed dementia over the eight years — roughly 0.6% — so a 44% higher relative risk is a small shift in absolute terms. It's the kind of difference that matters across a whole population and stays modest for any one person.

The same study also found unprocessed red meat (per 50 g/day) associated with lower risk. Treat that as a caution flag rather than a green light: an apparent protective effect in observational data often reflects other ways the people eating it differ, and the study couldn't rule that out. What it does suggest is that the signal here is specific to processed meat rather than to meat in general.

Smarter swap: move deli meat and sausage from "daily default" to "occasional." On sandwich days, swap in tinned fish, roast chicken, hummus, egg, or beans. Twice-a-week beats never — sustainable change is the point.

3. Sugary drinks and added sugar

Where the sugar sits seems to matter more than the sugar itself. In a UK Biobank analysis of 186,622 adults followed for about a decade, free sugars from drinks — soda and fruit drinks in particular — were associated with higher dementia risk, while free sugars from solid foods showed no clear association (Schaefer et al., Nutrition Journal 2023). Brain-imaging work points the same way without proving anything: in the Framingham Heart Study, higher sugary-drink intake was associated with lower total brain volume and poorer episodic-memory scores (Pase et al., Alzheimer's & Dementia 2017) — a cross-sectional snapshot taken at a single point in time, not a change anyone watched happen.

Not every cohort agrees, and it's worth saying so: a separate Framingham analysis found no association between sugar-sweetened beverages and dementia (Pase et al., Stroke 2017). Sugar's clearest route to risk may in fact be indirect — it drives weight gain, type 2 diabetes, and blood-pressure changes, all of which are themselves on the modifiable-risk list. If you manage diabetes, blood pressure, or cholesterol, talk to your clinician or a dietitian before making big changes to how you eat and drink.

Smarter swap: for most people the daily sweet drink is the easiest place to start (full detail: Sugary drinks and dementia risk). Sparkling water with a squeeze of citrus, unsweetened tea, or coffee replaces it without the sugar load. Diet soda isn't a clear win either, but the evidence there is genuinely unresolved — the well-known Framingham finding weakened once diabetes was taken into account, which points at least partly to confounding rather than to a direct effect. Water is the low-regret default.

4. Fried food (the weakest signal here)

This is the thinnest section on the page, and it's worth being upfront about that. Frying starchy foods at high heat forms acrylamide and other compounds, which has been raised as a possible concern for the brain — but that's a proposed mechanism, not something shown to drive dementia risk in people. We couldn't find a strong prospective study tying fried food specifically to dementia, so we won't imply one exists. What is fair to say is that fried food tends to travel with the patterns above, which do have an evidence base behind them.

Smarter swap: keep fried food as a treat, not a habit. Roasting, air-frying, grilling, and baking get you most of the texture with far less of the downside.

5. Excess saturated fat and refined carbs (the softer signals)

One meta-analysis of nine prospective cohorts (23,402 participants) linked the highest category of saturated-fat intake to a higher risk of cognitive impairment and of Alzheimer's disease, though the authors urged caution because the studies varied so much (Cao et al., J Prev Alzheimers Dis 2019). Refined carbohydrates (white bread, sugary cereals, many packaged baked goods) spike blood sugar in ways that may matter for the brain over decades. These signals are noisier and shouldn't drive fear — but they point the same direction: whole and minimally-processed beats refined and packaged.

Smarter swap: trade some refined grains for whole ones (oats, whole-grain bread, brown rice, beans) and let olive oil, nuts, and fish carry more of your fat intake — the pattern behind the MIND diet. If you take medication for cholesterol or blood sugar, run bigger dietary changes past your clinician first.

The pattern matters more than any single food

Notice the theme: every "smarter swap" above nudges you toward the same place — more whole, minimally-processed foods; fewer packaged, processed ones. That's the MIND-style eating pattern, and eating patterns are what the research actually supports. You don't need a perfect diet or a banned-foods list. You need a handful of repeatable swaps that stick.

And diet is only one lever. Dementia risk is shaped by a whole set of modifiable factors — physical activity, blood pressure, hearing, sleep, blood sugar, and more — which is why what matters most differs from person to person.

Want the full picture on eating for brain health? Start with our pillar guide: The MIND diet and dementia risk, and see the complete list of modifiable risk factors.

How we research and review our content →

This article is educational and is not medical advice, diagnosis, or treatment. The studies cited — the UK Biobank ultra-processed food and processed meat cohorts, the 2024 Journal of Neurology systematic review, the Nutrition Journal free-sugars analysis, the Framingham sugary- and artificially-sweetened-beverage analyses, and the saturated-fat meta-analysis — are observational and describe associations in general populations, not proof that any food causes dementia and not findings about Solenna specifically. Solenna is not a medical device and does not diagnose, treat, prevent, or cure Alzheimer's disease or any other condition; risk reduction means lowering probability, not eliminating it, and no outcome is guaranteed. Always discuss health decisions — including diet changes, any medication you take for diabetes, blood pressure or cholesterol, and any concerns about memory — with a qualified clinician or dietitian.