The Wellness Brew - Issue 31

Ask what a healthy gut microbiome looks like and you'll get a confident answer from almost anyone selling something. Ask the researchers who study it and the answer is that nobody knows yet. There's no agreed set of bacteria, no target ratio, no diversity score that marks a gut as healthy. That sounds like a reason to tune the subject out, and it isn't. Some of what we know about the gut is settled well enough to act on this week. The work is telling that apart from everything stacked on top of it.

But first, we filtered the noise — here's what's worth knowing this week.

Rhonda Patrick's position, which she revisited in a blueberry solocast two weeks ago, is that she skips bananas because they're unusually high in polyphenol oxidase, an enzyme that starts breaking down polyphenols (which are what make blueberries to good for you) the moment the blender runs. This took the internet by storm. The study behind it is real: eight men at UC Davis, a crossover design, and a high-enzyme banana drink produced an 84% lower peak of circulating flavanol metabolites. What gets lost is what the researchers put in the drink. They tested epicatechin from a cocoa extract, and never measured berry polyphenols at all. The study's lead author has since said plainly that bananas are still a great fruit for a smoothie. If you blend cocoa or matcha, keep the banana out. For berries, this isn't the problem it's become. — Ottaviani JI, et al., Food & Function, 2023.

A meta-analysis pooled nine randomized trials covering 513 older adults and found that daily blueberry intake improved episodic memory and language memory by small but statistically real margins. The catch is who it worked in. The benefit appeared in people who already had mild cognitive impairment or self-reported cognitive decline, not in healthy adults, and four of the six measures tested, including working memory, processing speed, recognition memory and visuospatial learning, came back with nothing. So this isn't evidence that blueberries sharpen a healthy brain. It's decent evidence that in people whose memory is already slipping, roughly a cup a day shows up as a small improvement on memory testing. The authors want multicenter trials before calling it settled. At the price of a cup of blueberries, there's no reason to wait for them. — Nardelli da Silva AB, et al., Biogerontology, 2025.

Walk through any supplement aisle and the gut is everywhere. Probiotics for mood. Protocols to repair a leaky barrier you didn't know you had. Mail-in stool kits that score your "gut age." Few corners of health have been turned into products this quickly.

The biology underneath the marketing is real, and parts of it are settled well enough that you can act on them. The trouble is that the settled parts and the speculative parts get sold at the same volume, and it's almost always the speculative parts that come with something to buy.

So here's what actually holds up.

Your large intestine holds trillions of bacteria and they're doing work your own cells can't. Bear with me as the below gets a bit sciency. 

The biggest job is finishing digestion. Your body produces enzymes for starch, fat, and protein, but it has no way to break down most of the fiber in plants. That material arrives in the colon essentially intact, and the bacteria there ferment it. What they produce in the process are short-chain fatty acids, mainly butyrate, acetate, and propionate.

Butyrate matters most. The cells lining your colon use it as their primary fuel source, which is an unusual arrangement in human physiology: an entire tissue running on a byproduct of bacterial metabolism. Those same cells form the barrier between what's in your intestine and what's in your bloodstream. Keep them fed and the barrier holds together better.

The second job is training your immune system. Approximately two-thirds of your immune tissue sits in and around the gut, and a large part of what it does early in life is learn the difference between a threat and a tenant. Animals raised in sterile conditions, with no gut bacteria at all, develop immune systems that are visibly underbuilt.

Then there's the more mundane work. Gut bacteria synthesize vitamin K and several B vitamins. They occupy physical space and consume available nutrients, which makes it harder for harmful organisms to establish a foothold. And they help regulate how quickly things move through the intestine.

Now that we know how a “healthy” gut supports us, let’s dive into how you’d identify and define a healthy gut vs poor gut health. 

In 2021, three researchers at University College Cork published a review in Gastroenterology with a question for a title: what is the definition of a healthy gut microbiome? Their answer, after surveying the field, was that there isn't one. No specific set of bacteria, no ratio, no diversity score has been established as the signature of a healthy gut. [1]

Part of what makes this difficult is the sheer variation between people. A large population study in Science examined gut microbiome composition across two European cohorts and identified 69 separate factors that tracked with it, from stool consistency to medication use to diet. Those 69 factors, taken together, accounted for only a modest fraction of the differences between individuals. Most of the variation remains unexplained. [7] Two healthy people can have gut communities that look almost nothing alike.

The usual fallback to defining gut health is diversity. More types of bacteria equal a better gut. That idea has real support in some conditions and falls apart in others. A 2026 paper in Science made the point directly: approaches built on diversity scores assume that reduced diversity is a universal marker of disease, and the evidence doesn't support that. Diversity-based measures behave inconsistently across diseases and across study populations. [2] Reduced diversity shows up reliably in inflammatory bowel disease and in C. difficile infection. In conditions like irritable bowel syndrome, the picture is mixed.

Which is why the field has been shifting toward a functional definition. The useful question isn't which species are present, it's whether or not the bacteria that are present are doing their jobs: fermenting fiber into short-chain fatty acids, supporting the barrier, keeping pathogens suppressed. Two people with entirely different species lists can both have a gut that works.

So what are the actual risks of a gut community that isn't working? The honest answer separates into three tiers.

Established and causal: C. difficile infection, which happens when the good bacteria are wiped out and the C. diff bacteria take over and release toxins. This is the clearest case of gut disruption directly causing disease, and it can be severe. This is usually associated with antibiotics wiping out healthy gut bacteria. 

Strongly established, with direction of causality still unsettled: inflammatory bowel disease. The microbiome of someone with Crohn's or ulcerative colitis looks different from a healthy control's, consistently. Whether that difference helps drive the disease or results from it is still being worked out.

Associated, mostly from observational research: obesity, type 2 diabetes, depression, autoimmune conditions, colorectal cancer. These links are real in the sense that they show up repeatedly in the data. They are also exactly the kind of finding that can't tell you which way the arrow points. Sick people eat differently, take more medication, and move less, and all three of those reshape the gut. The mouse work is more suggestive, but mice are not people, and gut transplant studies in humans remain early.

The honest summary: a disrupted gut is a legitimate medical problem in a handful of specific, well-defined situations. Beyond those, it's an active research area with a lot of promising leads and not much that's nailed down.

If there's no agreed definition of a healthy microbiome, then how would a test score yours accurately? 

In February 2026, researchers at the National Institute of Standards and Technology and the University of Maryland ran a straightforward experiment. They took a single standardized human fecal sample, developed at NIST specifically as a reference material, and sent portions of that identical material to seven direct-to-consumer gut microbiome testing companies. Three kits from each company. Twenty-one tests, all on the same stool. [4]

The results came back as if they'd tested twenty-one different people.

The number of bacterial groups each company reported ranged from 34 to 906. Across all the analyses, 1,208 unique organisms were named. Only three of them appeared in every single result. When the researchers narrowed to the 18 groups that showed up at least once from every company and compared each company's numbers against the consensus, no company matched on all 18. Three of the seven reported C. difficile present in the sample. The other four reported it absent.

The clearest illustration involved three kits from one company. Two of its kits came back with the microbiome scored as healthy, while the third came back scored as unhealthy.

The variation between providers was on the same scale as the biological variation between completely different donors. For 17 of the 18 groups examined, the differences introduced by testing method were as large as or larger than the differences between actual human beings. Which means the signal you're paying for is buried under the noise of how it was measured.

This isn't a rogue-operator problem. These services generally aren't regulated as medical devices, and there's no standard they're required to meet. An international consensus panel of microbiome researchers and clinicians, publishing in The Lancet Gastroenterology & Hepatology in 2025, reviewed microbiome testing in clinical practice and found the evidence for clinical usefulness scarce enough that their main output was a call for a regulatory framework. [3]

Stool testing has genuine clinical uses, ordered by a physician for a specific question. But paying out of pocket to find out your gut age or health score is not one of them.

Fiber, in quantity. This is the most direct lever you have, because fiber is the raw material for the whole short-chain fatty acid process described earlier. Most adults in the US eat roughly half of the recommended intake, which sits around 25 grams a day for women and 38 for men. Closing that gap does more than anything else. Beans, lentils, oats, whole grains, vegetables, fruit with the skin on, nuts and seeds.

Variety across plants, not just volume. Different fibers feed different bacteria, so eating thirty grams of fiber from one source supports a narrower community than the same amount from a dozen. The "thirty different plants a week" guideline that circulates widely is a reasonable goal, though it's worth knowing it functions as a target rather than a finding from a trial.

Fermented foods. This one has a good controlled trial behind it. Stanford researchers randomized 36 healthy adults to either a high-fiber diet or a high-fermented-food diet for ten weeks, publishing in Cell in 2021. The fermented food group, eating about six servings a day of things like yogurt, kefir, kimchi, and kombucha, increased their microbial diversity, and the increase held for four weeks after the study ended. They also showed decreases in 19 separate inflammatory markers, including interleukin-6. The high-fiber group did not increase diversity, though their existing bacteria shifted toward producing more fiber-digesting enzymes. [5]

Thirty-six people is a small trial, and the diversity finding needs replication at scale before anyone treats it as settled. But it's a randomized controlled trial with a biologically coherent result, which puts it well above most of what gets cited in this space.

Probiotic supplements. Much weaker than the marketing suggests. Specific strains have decent evidence for specific situations, most notably reducing antibiotic-associated diarrhea. The broader claims, that a general-purpose probiotic improves mood, immunity, or body composition in a healthy adult, mostly rest on small studies. Strain specificity is the catch: evidence for one strain says almost nothing about another, and the labels frequently don't make clear which you're getting.

Antibiotics flatten the gut community. That's not a side effect to work around, it's what the drug does, and when you need one, you need one. Nothing here argues otherwise.

What's worth knowing is how the rebuild goes, because the standard advice may be backwards.

A 2018 study in Cell followed people after a course of antibiotics and compared taking a probiotic against doing nothing. The group that did nothing recovered their original gut community within about three weeks. The group taking probiotics recovered more slowly, with the delay extending for months. The probiotic bacteria colonized the emptied gut and appear to have inadvertently gotten in the way of the native community coming back. [6]

That finding has been contested, including by the trade association representing probiotic manufacturers, which is worth weighing in both directions. The study also used one specific multi-strain product in a small group of people, and hasn't been widely replicated.

Still, it's the best direct test anyone has run on the question, and it points away from reflexively taking a probiotic after antibiotics. Feeding the bacteria you still have, with fiber and fermented foods, is the lower-risk approach.

The things you can do to support your gut health with the strongest evidence are the ones with no product attached: eat more fiber, eat a wider range of plants, eat fermented foods, and don't take antibiotics you don't need. The parts being sold hardest, the tests and the general-purpose probiotics, are where the evidence is thinnest.

And the biggest question, whether the state of your gut is driving your metabolic health or your mood or your immune function, is genuinely open. 

Four things you can do this week:

1. Find out how much fiber you’re eating before changing anything.
Log everything you eat for three days in a tracking app and look at the fiber line. If you land in the low teens, which is where most Americans sit, you have the single biggest lever in this whole subject sitting untouched. Work toward 25 grams a day for women and 38 for men, and get there over a few weeks rather than in one day, because a sudden jump is uncomfortable enough that most people quit. Where the fiber comes from matters nearly as much as the total: a dozen sources across the week feeds a wider range of bacteria than the same grams from oatmeal alone.

2. Put one fermented food into the daily rotation.
Yogurt with live cultures, kefir, kimchi, sauerkraut from the refrigerated section, miso, kombucha. The Stanford trial found it took six servings a day to increase microbial diversity, which is more than most people will sustain, so be honest with yourself with how many servings you can reasonably eat each day. This is the intervention with an actual randomized trial behind it, which is more than the supplement say.

3. Don't buy the mail-in gut test.
Seven companies analyzing the identical stool sample returned results that looked like twenty-one different people, and one company's three kits disagreed with each other about whether the sample was healthy. There is no number on that report you can act on. If you have symptoms that are actually bothering you, ongoing pain, blood, unexplained weight change, or a change in bowel habits that has lasted weeks, that is a conversation with a physician who can order a test aimed at a specific question. 

4. After a course of antibiotics, feed what survived instead of reseeding.
The reflex is to take a probiotic, and the best direct test of that reflex found it slowed recovery rather than speeding it up. The group that did nothing was back to their original community in about three weeks. Eat fiber and fermented foods during and after the course and let the rebuild happen. The exception worth knowing: specific probiotic strains do have reasonable evidence for reducing antibiotic-associated diarrhea, so if that is the problem you are solving, it is a different question from rebuilding your microbiome.

If you found this useful, please forward it to your friends and family. If you had this forwarded to you — sign up here.

Wellness, filtered.

The Wellness Brew

Sources:

Disclaimer: The Wellness Brew is for informational purposes only and does not constitute medical advice. The content published here is not intended to diagnose, treat, cure, or prevent any disease or health condition. Always consult a qualified healthcare professional before making any changes to your diet, supplement routine, or lifestyle.