Sweets, Pizza, and Processed Meat May Be Linked to Faster Cognitive Decline in Some Adults, Research Finds

A new Australian study on aging and the brain puts pizza, processed meat and sweets in the spotlight. Yet the real finding is smaller, stranger and more interesting than the food list suggests.

Researchers in Perth followed 185 older adults and found that a Western-style eating pattern was linked to faster declines in attention, but only among people carrying a gene variant tied to Alzheimer’s risk.

There’s a catch, though, and it doesn’t appear in the study’s public announcement. When the researchers subjected their own findings to a stricter test, the result changed. The study makes that clear, even if the detail is less attention-grabbing than a list of familiar foods.

So what did the study actually find? How much confidence does it deserve? And what do larger studies say about the same gene and the foods on our plates?

What the Perth Team Actually Found

One gene, one thinking skill, and one broad eating pattern make up the core finding.

The study, published in August in the journal Nutrients, used data from the Western Australia Memory Study. Participants averaged 70, and none had a dementia diagnosis when they joined, although some already showed signs of memory problems.

They completed detailed food questionnaires, underwent genetic testing, and took cognitive tests every 18 months.

Rather than isolate individual foods, researchers identified two eating patterns. The Western pattern included red and processed meat, chips, pizza, sweets, refined grains, and beer. The “Prudent” pattern emphasized fruit, vegetables, leafy greens, garlic, and whole grains.

About 30% carried APOE ε4, the strongest common genetic risk factor for Alzheimer’s disease. It raises risk but does not guarantee the disease. Among carriers, a stronger Western-pattern diet was associated with faster declines in attention.

Non-carriers showed no significant association, and the Prudent pattern showed none in either group. Those results concern attention scores, not Alzheimer’s diagnoses.

The Result Wobbled Under a Stricter Test

The authors ran a second check, and the finding didn’t hold up.

Nineteen of the 185 participants scored below 23 on the Montreal Cognitive Assessment when they joined. Researchers treated that as a marker of possible cognitive impairment, not a clinical diagnosis. After excluding those 19 people, the Western-diet association among APOE ε4 carriers was no longer statistically significant.

That raises an awkward possibility: early cognitive changes might influence what people eat, or how well they remember what they ate. The authors point to this themselves. Known as reverse causality, it could mean that changing cognition helped shape the dietary results, rather than the other way around.

The authors describe the finding as fragile and exploratory. That’s a long way from a pizza-night warning label.

Seven Years Sounds Longer Than It Was

The “more than seven years” figure reflects the longest follow-up, and most people were followed for far less.

Even the time span needs context. Some participants were followed for 7½ years, but the average was about 50 months, closer to four years. Participants averaged fewer than three follow-up visits, and only 36 reached the final assessment. The researchers say that left the genetic-subgroup analysis underpowered.

Volunteers were recruited between 2006 and 2009, so the new part is the analysis, not the cohort. Each person also answered only one food questionnaire, covering the previous 12 months.

That leaves no way to tell whether their diet stayed the same for years or whether changing eating habits might have affected their cognitive scores. The authors want future research to track diets repeatedly.

Related: Coffee and Kidney Health: Risks, Benefits, and What Studies Say

A Bigger Chicago Study Saw a Similar Pattern

Different fats pulled in opposite directions for people with the same gene variant.

A larger study gives the gene angle some company, though not a direct replication. In 2024, researchers studied 3,360 older adults in the Chicago Health and Aging Project, including 1,097 APOE ε4 carriers.

Among carriers, higher saturated fat intake went with faster cognitive decline. Every 5-percentage-point rise in the share of calories from saturated fat was associated with a 21% faster decline in global cognition. That’s about the speed of decline, not a 21% increase in Alzheimer’s risk.

Long-chain omega-3 fatty acids from food went the other way, accompanying slower decline. Neither association appeared in non-carriers.

However, an important distinction remains. Chicago examined fats and overall cognition, while Perth examined eating patterns and attention. And like the Australian team, Chicago’s researchers caution that an association is not proof of cause and effect.

Healthy Eating Hasn’t Delivered a Clean Win Either

Image Credit: fachrulrevlindi/123RF

When researchers actually assigned people a brain-friendly diet, the answer got muddier.

A randomized trial complicates the picture further. Over three years, researchers in the New England Journal of Medicine studied 604 older adults with a family history of dementia, a less-than-ideal diet, and a body-mass index above 25.

One group followed the MIND diet with mild calorie restriction; the other followed a control diet with the same restriction. Cognitive scores improved in both groups, with no significant difference between them. Brain MRI results also showed no significant difference.

Both groups received counseling and weight-loss support and lost similar amounts: about 5 kilograms in the MIND group and 4.8 in the control group.

Subsequent analyses added wrinkles. Among 590 participants, those who increased their MIND score by an average of four points on a 14-point scale improved global cognition more than those whose scores stayed the same. But researchers didn’t randomize individual score changes.

In another analysis of 442 participants, higher baseline blood carotenoid levels went with slower decline among 134 APOE ε4 carriers, while associations in the others were weaker or non-significant. These were blood measurements, not supplements.

Where the Evidence Stops

Larger datasets haven’t produced a tidy list of brain-saving and brain-harming foods.

Even larger datasets resist a simple rule. A pooled analysis of 17,368 people across eight European and U.S. cohorts linked several healthier eating patterns to better cognitive performance at one point in time, but far fewer to changes over time. The favorable exception was MIND adherence and executive function in three U.S. cohorts.

That helps explain why the Perth team wants more research into individual differences before genetic risk informs specific dietary advice. The studies don’t support using a genetic test to decide what to eat, ban a particular food, or set a serving limit.

There’s also a question left unanswered. None of the Perth study’s public summaries explain why the association appeared in carriers but not non-carriers. Establishing whether that difference is real, and what might explain it, will take further investigation.

When a headline links a food to brain health, what makes you stop and read the details instead of just scrolling past?