A dog's age is supposed to be a simple fact: a birthday, or a shelter's best guess at one. A new international study says there may be a second answer to that question, and it's sitting in your backyard.
Researchers led by Prof. Elhanan Borenstein at Tel Aviv University, working with the Dog Aging Project, mapped the gut microbiomes of 922 pet dogs living in homes across the United States — then built a machine-learning model that can estimate a dog's age from the bacteria in a stool sample alone. The work was published in Nature Communications and picked up wide coverage this week.
The headline writes itself: a poop-based biological clock for dogs. But the genuinely useful part of this study isn't the clock. It's the enormous, boring, well-controlled picture it gives of what actually shapes a normal dog's gut — and how little of it has anything to do with the supplements marketed at you.
What the researchers actually did
The team worked with the Dog Aging Project's Precision Cohort — the subset of enrolled dogs that gets deep biological sampling on top of the annual owner health surveys. Of 976 dogs enrolled, 922 were successfully sampled and analyzed: 444 females and 478 males, 84% of them spayed or neutered, ranging in age from 8 months to 18 years.
Rather than the cheap 16S rRNA sequencing used by most consumer dog-gut kits, the researchers used fecal shotgun metagenomic sequencing, which reads the actual genetic material of the whole microbial community rather than one identifying barcode gene. That was paired with owner surveys about diet, environment and behavior, plus real clinical bloodwork and urinalysis.
That combination is what makes this dataset unusual. Most dog microbiome studies are a dozen research beagles eating an identical diet in an identical kennel. This one is nearly a thousand real dogs, in real houses, eating what their people actually feed them.
Finding 1: Age leaves a real, readable signature
Age turned out to be one of the strongest factors shaping gut bacterial composition. As dogs got older, the diversity of their gut bacteria declined and the overall community gradually shifted, with dozens of individual bacterial species rising or falling in abundance in a predictable direction.
The shift was strong and consistent enough that the model, trained on part of the cohort, could then significantly predict age in a separate group of dogs it had never seen. The researchers also compared their results against a large human cohort study and found that some — not all — of the age-related microbial patterns known in people were echoed in dogs.
The team is careful about this, and so should you be: this is a population-level model at an early stage, not a diagnostic test. It can tell you that dogs like this one tend to be around a certain age. It cannot yet tell you that your specific dog's gut is "three years older than it should be," and nobody has shown that nudging those bacteria makes a dog live longer.
Finding 2: What you feed your dog shows up clearly
Dogs eating commercial food had measurably different microbiome profiles from dogs eating home-cooked diets. That's not a verdict on which is better — it's a reminder that diet is one of the largest and most controllable inputs into gut composition, which is exactly why any gut-health advice that ignores the food bowl is incomplete.
Behavior mattered too. Coprophagy — poop-eating — showed up as a distinct signal in the data. If you've been treating your dog's habit as purely a gross-out problem, it's also a microbial event.
Perhaps the most striking result: when the team compared microbiome composition against the dogs' actual lab results, microbiome composition often explained more of the variation in blood and urine values than factors like breed or sex did. The individual effects were modest, but they were consistent.
So should you buy a dog gut test or a probiotic?
Here's where the science and the marketing part ways, and where most articles about studies like this quietly go off the rails.
Consumer "gut health" kits
The direct-to-consumer dog microbiome tests you can buy online are not the same technology used in this study, and there is currently no validated consumer test that tells you your dog's gut is "unbalanced" in a way that should change your care. The one analytically validated clinical assay is the Dysbiosis Index (DI) from the Texas A&M Gastrointestinal Laboratory — a quantitative PCR test that measures seven bacterial groups plus total bacterial abundance and returns a single number. It is ordered by veterinarians, usually for dogs with chronic diarrhea, weight loss or suspected chronic enteropathy — not for healthy dogs out of curiosity. A DI above 2 is considered abnormal, and results are meant to be read alongside the abundance of a specific organism, Peptacetobacter (Clostridium) hiranonis, whose loss is a major driver of an abnormal result.
Probiotics
Probiotic evidence in dogs is narrower than the shelf space suggests. A veterinary review from the University of Edinburgh concluded that despite widespread use, scientific evidence for pro-, pre- and synbiotics across canine conditions remains scarce, with strain-specific effects poorly evaluated. Key points worth carrying to your vet:
- Strains are not interchangeable. "Probiotic" is not a product category with shared evidence; each strain has its own (usually thin) file.
- Enterococcus faecium SF68 (the strain in Purina's FortiFlora) has the strongest single-strain canine evidence base, going back to a 2011 JVIM trial showing reduced diarrhea in shelter dogs. AAHA and ACVIM consensus statements rate it as moderate evidence for adjunctive use in acute idiopathic diarrhea, antibiotic-associated upset and stress-related dysbiosis.
- Multi-strain blends are mixed. A placebo-controlled trial of the seven-strain Proviable-Forte blend found it safe but no better than placebo or metronidazole for acute diarrhea; a different multi-strain blend did beat placebo in puppies with gastroenteritis.
- Effects fade. In a randomized trial of healthy household dogs given a synbiotic, the microbiome shifted significantly at four weeks — then largely returned to baseline two weeks after supplementation stopped.
- "Gut health" is not a tested claim. The real evidence is narrow: specific strains, for specific GI problems, with modest results.
Fiber
Dietary fiber is the least glamorous and best-supported lever. Feeding trials show that fiber sources measurably change fecal microbiota diversity, short-chain fatty acid production and fecal IgA in healthy dogs. If you want to influence your dog's gut, the food is a bigger dial than the jar of powder.
What to actually do with this
A sensible, evidence-shaped approach for a normal healthy dog:
- Do pick one complete, consistent diet and stop rotating on impulse — the study confirms diet is a dominant input.
- Do transition foods over 7–10 days rather than overnight.
- Do discourage coprophagy in the yard and on walks; pick up promptly and trade for a treat rather than scolding.
- Do ask your vet about SF68 specifically when your dog is on antibiotics, boarding, or has acute diarrhea — those are the situations with actual data behind them.
- Do get chronic diarrhea, chronic vomiting or unexplained weight loss worked up properly, where a vet-ordered Dysbiosis Index may genuinely help.
- Don't buy a consumer gut-microbiome kit expecting it to change your dog's care plan.
- Don't treat a probiotic as a permanent daily requirement for a healthy dog — the effect washes out when you stop.
- Don't read "biological age" headlines as proof anyone can slow your dog's aging with bacteria. That trial hasn't been run.
Why this study matters anyway
The real payoff is longer-term. The Dog Aging Project's whole premise is that companion dogs are the best available model for human aging: they share our houses, our air, our water, our couches and, increasingly, our sedentary habits — while compressing an entire lifespan into roughly a decade. Nine hundred and twenty-two dogs' worth of shotgun metagenomics, paired with real clinical labs and years of owner-reported health data, is the kind of baseline that makes it possible to eventually ask the interesting question: does a dog whose gut looks older than its birthday actually get sick sooner?
Nobody can answer that yet. But for the first time, somebody has the dataset to try — and your dog's next annual stool sample is a slightly more interesting object than it was last week.
For more evidence-checked dog health reporting, plus the best dog-friendly patios, parks and breweries in your area, browse our city guides at Sidewalk Dog — where we do the digging so you can get back to the walk.
