What the Dog Aging Project Is and What It Has Found So Far

A twelve-year-old Labrador mix stalls halfway up the porch steps, then finishes the climb a beat slower than she did last spring. Her owner mentions it almost in passing at a routine visit — not a limp, not pain, just three weeks of “she’s just older now.” That offhand line is exactly the kind of signal the Dog Aging Project was built to capture at scale, because a single exam room only ever sees one dog at a time, and slow decline hides well in a fifteen-minute appointment.

The Dog Aging Project is a nationwide, NIH-funded longitudinal study tracking tens of thousands of companion dogs over their lifetimes. As the National Institute on Aging - NIH: National Institute on Aging describes it, the project exists to define what aging actually looks like in dogs, identify the genetic and environmental factors that push it faster or slower, and test interventions that might extend both lifespan and healthspan — the years a dog spends healthy, not just alive. Dogs share our homes, our couches, and often our exposure to pollution and stress, but they age roughly seven times faster, which makes them an unusually efficient window into aging biology that also happens to matter directly to the people reading about it.

Why researchers built the study around dogs, not mice

I spend most of my clinical time thinking about the nervous system, and the honest answer is that mouse models of neurologic aging have limits — inbred strains, short lifespans, artificial housing. Dogs live in the same environment we do, get diagnosed with the same imaging and lab tools our hospitals use, and develop naturally occurring diseases, including cognitive decline that looks a great deal like human dementia on a cellular level. That overlap is the whole rationale. The National Institute on Aging notes that data from the project is meant to generate hypotheses about human aging biology, not just canine care, which is why a study this large needed to happen at all.

Structure: a giant community cohort, plus a deep-data subgroup

The bulk of participants make up the “Pack,” an open community cohort where owners fill out annual health and behavior surveys on dogs of every breed, size, and background. Layered underneath is the Precision cohort, a smaller group built for depth rather than breadth. A GeroScience paper describing that cohort’s design targeted 1,000 dogs and ultimately processed biospecimens — blood, urine, stool — on 976 of them, generating the kind of multi-omic data (genomics, metabolomics, microbiome) that a survey alone can’t produce. That’s the tradeoff every longitudinal study makes: broad self-reported data tells you what’s happening across the population, while a smaller deep-phenotyped group tells you why. Neither one alone would have been enough.

What the project has actually found

Two findings stand out, and neither is what most owners would have guessed going in.

First, social connection appears to matter more than money. Analysis published in Nature as part of the project’s open-science dataset found that the effect of social support on a dog’s health was roughly five times stronger than the effect of household financial resources. A dog with steady companionship and engagement outperformed a dog with a bigger budget but less social contact. That’s a striking result for a field that often assumes healthcare spending drives outcomes.

Second, blood chemistry tells a story that looks familiar from human aging research. Researchers identified patterns of small metabolic molecules — tied to metabolism, inflammation, and cellular stress — that correlated with shorter or longer expected lifespan, and did so in ways that echo similar biomarker patterns already documented in people. That’s a mechanistic thread worth naming plainly: a metabolic signature is not a diagnosis and not a treatment target on its own, but it’s a real clue about which biological pathways age faster in some dogs than others, and it’s the kind of data point that eventually turns into an actual clinical test.

The rapamycin trial, and where the gap between promise and proof actually sits

The project’s highest-profile intervention arm is TRIAD — the Test of Rapamycin in Aging Dogs — a placebo-controlled, double-blinded trial examining whether low-dose rapamycin, a drug already used in human transplant medicine, can improve heart function, mobility, and cognition in middle-aged and older dogs. According to AVMA reporting, the trial had enrolled around 170 dogs across 20 sites before a new five-year, $7 million NIH grant allowed it to expand toward a target of 580 dogs at additional locations. In lab animals, low-dose rapamycin has shown effects on heart muscle function, cognition, and even lifespan — but TRIAD’s own investigators are careful to say the trial is ongoing and full-cohort results in dogs are not yet published. That distinction matters clinically: laboratory promise in mice is not the same evidentiary weight as a completed, blinded trial in the species you actually treat, and until TRIAD reports its results, rapamycin belongs in the “being studied” column, not the “proven longevity drug for dogs” column.

I saw a version of this gap play out with a different family last year — an eleven-year-old Border Collie, 42 pounds, whose owner had read about NAD+ precursors and cellular-aging research and wanted to start “whatever the science dogs are on” immediately, before her own vet had even run baseline bloodwork. Six weeks in in, nothing had measurably changed except the owner’s confidence that something was being done, which is its own kind of risk: it can quietly replace the diagnostics — bloodwork, joint exams, cognitive screening — that would have caught something earlier. The lesson wasn’t that supplementation is wrong; it’s that population-level research findings and an individual dog’s workup are two different kinds of evidence, and one doesn’t substitute for the other.

That gap between “supports normal cellular function” and “treats a disease” is one I come back to constantly in supplement conversations, because it’s where marketing tends to outrun the data. Ingredients studied in this space — nicotinamide riboside, quercetin, resveratrol — have legitimate, if early, evidence behind them as structure/function support. Nicotinamide riboside is a NAD+ precursor studied for its role in normal cellular function; quercetin is a flavonoid with antioxidant properties studied for immune and cellular support; resveratrol is an antioxidant researched for cellular health and vitality. None of that is the same claim as reversing aging or curing a condition, and an owner survey (Boops Pets, March 2025, n=193) reporting that 97.8% of respondents saw visible improvement after the first jar, with 94% reporting a cognitive-function benefit and 98% reporting an energy benefit, is useful anecdotal signal about a specific product — but it’s observational owner-reported data, not a controlled trial, and it should be read as exactly that: encouraging, not conclusive. Some of the underlying ingredient research is even earlier-stage than that. A 2024 study in Pharmaceuticals testing quercetin formulations in a nematode model found that quercetin, N-acetylcysteine, and ascorbic acid didn’t affect feeding behavior at the concentrations tested — a reminder that a lot of “longevity ingredient” evidence still lives in cell and invertebrate models, several steps removed from a dog on your couch. And dose and context can cut both ways: a 2026 Phytomedicine study found that embryonic exposure to resveratrol at 50 μg/ml accelerated development but shortened adult lifespan in the model organism tested, which is a useful caution that “antioxidant” doesn’t automatically mean “more is better” or “safe at every life stage.”

For an owner trying to translate any of this into a Tuesday-afternoon decision: don’t wait for TRIAD’s final results to prioritize the basics the project has already validated — social engagement, activity, and consistent veterinary monitoring — because those are the levers with the strongest data behind them right now. If you’re also comparing antioxidant or NAD+-support options, our breakdowns of curcumin versus quercetin, fisetin versus quercetin, and astaxanthin versus resveratrol walk through what’s actually backed by evidence versus what’s still marketing-forward, and our apoaequorin versus NAD+ piece covers the cognitive-decline angle specifically.

An edge case worth knowing before you generalize any of this

Not every dog in this dataset ages on the same curve, and the project’s own design accounts for that — which is why a single anecdote can mislead as easily as it can illuminate. I once fielded a call about a nine-year-old Great Dane, already showing gray around the muzzle and slower to rise after lying down for more than 20 minutes, whose owner assumed the dog was tracking behind schedule compared to a friend’s nine-year-old terrier that still ran full loops around the yard. Giant breeds compress their whole aging timeline into a shorter life; by year nine, that Dane was already deep into what the project would classify as its senior life stage, while the terrier had years of middle age left. Population data has to be read through breed size and life stage, not treated as one universal clock — otherwise an owner ends up either panicking early or missing real decline because they benchmarked against the wrong dog.

Quick recap: the Dog Aging Project is real, NIH-backed, and already producing findings — social connection outweighing money, metabolic biomarkers echoing human aging patterns, and an ongoing rapamycin trial that hasn’t reported final results yet. None of that licenses skipping your vet’s own bloodwork and exam in favor of a study finding or a supplement claim; it’s context for that visit, not a replacement for it. Individual dogs vary, breed size changes the aging timeline, and the project’s biggest current value to owners is confirming that the boring fundamentals — company, movement, and monitoring — are doing more work than most of us assumed.