The Gut Bacteria Pattern Found in People Who Live Past 90, and What It Does and Doesn't Tell Us
A 2026 study of 301 samples found people over 90 had gut diversity like adults decades younger, with more Akkermansia and Bacteroidota. Here is what that means, and what it doesn't.
If you wanted to know what a very long life looks like from the inside, you could ask the oldest person you know about their habits. You would get a story about walking, or a garden, or one particular cup of tea. A different approach is to stop asking and start sequencing: take a stool sample, read the bacterial DNA, and see what the gut of a 94-year-old has in common with other 94-year-olds.
That is roughly what a 2026 study did. Researchers used 16S rRNA sequencing on 301 fecal samples from three groups of people in China, and compared those aged 90 and over with adults aged 45 to 59 and with older adults aged 60 to 89. I am relaying the findings as published; the paper is the place to check the detail. What it found is more interesting than "eat more probiotics," and also easier to over-read.
What the study found
Three things stood out. First, the people aged 90 and above had gut microbial diversity comparable to the much younger 45 to 59 group, rather than looking like typical 60-to-89-year-olds. Second, two groups of bacteria were more abundant in the long-lived: Bacteroidota and Akkermansia. Third, two were less abundant: Prevotella_9 and Megamonas.
A separate 2026 review in Biomedicines frames the gut microbiome as one of the things that modulates immune regulation, metabolic balance and neuroendocrine signaling in aging. That is a framing paper rather than a result, but it explains why researchers keep looking here: the gut is a place where several slow-aging systems meet.
Why Bacteroidota and Akkermansia are interesting
Bacteroidota is a large group of gut bacteria, and many of its members are specialists in breaking down complex plant carbohydrates into short-chain fatty acids, which the cells lining the colon use as fuel. That is the part that links it to gut-barrier health.
Akkermansia is more particular. It lives close to the mucus layer that coats the gut wall and feeds on that mucus. This sounds alarming until you learn that the layer is constantly renewed, and that in many studies a moderate Akkermansia population goes along with a sturdier barrier and better metabolic markers. In the long-lived group, more of it was present. Whether it did anything for them is another matter, which brings us to the main caveat.
Diversity itself, not a list of heroes
The detail I find most useful is that the long-lived group did not simply have more of a few good bacteria. Their overall diversity looked younger. That fits a view that a gut ecosystem resembles a garden more than a pharmacy: a varied community seems to handle disturbance better than one dominated by a few species, however respectable those species are.
It also quietly complicates a popular story. Prevotella is a genus often found in people eating plant-heavy, high-fiber diets, yet Prevotella_9 was lower in the oldest group here. So "more fiber, more of everything good" cannot be the whole explanation. Different bacteria rise and fall for reasons that are not a simple function of one nutrient.
Correlation versus causation
This is the section I would not skip. A person who has reached 90 has also had a particular life: a particular diet over decades, a particular level of activity, probably fewer courses of antibiotics, a particular region, income, family. Any of those might shape both the gut and the lifespan.
There is also a reverse possibility. A body that is aging well may produce a gut environment that supports these bacteria, rather than the other way around. And a single sample from each person is a snapshot, not a film. The study can say that these gut signatures occur together with extreme age. It cannot say that acquiring them will extend anyone's life.
What is known to shift Akkermansia
From separate research (not from this study), a few things have been linked to Akkermansia abundance: dietary fiber, polyphenol-rich foods such as berries, pomegranate and tea, and in some animal and small human studies, caloric restriction or fasting windows. The human evidence is smaller and less consistent than the headlines suggest, and individual starting points differ a lot.
Supplement versions of Akkermansia exist. I would treat them as a separate question from "foods that go along with a gut that looks like this," and as one where I would want to see long-term trial data before drawing a conclusion.
A table of associations
The table below maps the named bacteria to foods and habits that other research has plausibly linked to them. It is associative, not prescriptive. These are not instructions, and none of the links has been shown here to cause longevity.
| Bacteria | Direction in the 90+ group | Foods and habits plausibly linked | Confidence |
|---|---|---|---|
| Bacteroidota | More abundant | A varied range of plant fibers, whole grains, legumes | Moderate, mostly observational |
| Akkermansia | More abundant | Dietary fiber, polyphenol-rich foods (berries, pomegranate, tea), possibly fasting windows | Mixed; small studies |
| Prevotella_9 | Less abundant | Often higher with plant-heavy, high-fiber diets; no clear reason it fell here | Low; unresolved |
| Megamonas | Less abundant | No reliable dietary lever known; sometimes reported higher in metabolic conditions | Low; inconsistent |
How this differs from the "fiber diversity" idea
The popular version says to eat thirty different plants a week and your microbiome will thank you. That advice is reasonable and low-risk, and this study does not contradict it. What it adds is a note of humility: the signature of the very old is not simply "the most fiber-loving bacteria." It is a particular balance, and we do not yet know how to steer toward it on purpose.
If I had to turn this into one small, honest habit, it would be variety rather than any single food. A different grain this week, a bean you do not usually cook, tea instead of a second coffee on some afternoons. Boring advice, but at least it is advice that does not depend on one bacterium being the hero.
FAQ
Can I test my gut for these bacteria?
Consumer stool tests exist, but their reporting is not standardized and the long-life pattern from a single study is not something a home test can meaningfully score you against. I would not change my diet based on one result.
Should I take an Akkermansia supplement?
This study does not support that step. It observed the bacteria in people who were already very old, and did not test supplements. Talk to a clinician if you have a specific health reason to consider one.
Does a lower Prevotella mean I should eat less fiber?
No. The study did not look at diet, and fiber has a large body of evidence behind it independent of this result. The finding is a reminder that the microbiome is complicated, not a reason to cut plants.
Is gut diversity the same as eating a "good" probiotic?
Not really. Diversity describes the whole community. A probiotic adds a few strains, usually temporarily. Food variety is the more likely lever for the broad picture, though the evidence is still developing.
Maybe the quiet lesson of a very old gut is that it looks less like an achievement and more like a long, undramatic collaboration with a great many small things we mostly never meet.