Your Mouth Bacteria May Reveal How Fast You’re Aging

A swab of the inside of your cheek might tell you more about how fast you're aging than a birthday candle count ever could. Researchers analyzing oral bacteria from thousands of American adults have built a score that tracks with mortality, frailty, and kidney decline, sometimes years before those problems become obvious. It's not a clinical test yet, but it's the strongest evidence so far that the microbial community living on your gums and tongue keeps a kind of running tally of your health.

Oral Microbiome Aging Score Tracks Mortality and Frailty Risk

The team, led by Zhao and colleagues and published in Nature Communications, started with 16S rRNA sequencing of oral rinse samples collected as part of the National Health and Nutrition Examination Survey (NHANES). They identified 64 bacterial genera whose abundance shifts predictably with age, then trained a machine-learning model to predict a person's chronological age purely from their oral bacterial profile. The gap between a person's predicted "microbial age" and their real age became the core metric: the Oral Microbiome Aging Acceleration score, or OMAA.

People whose oral bacteria looked older than they actually were, a high OMAA score, had measurably worse outcomes. Each one-unit increase in OMAA was linked to a 4 to 5 percent higher risk of death (hazard ratio 1.05 in the discovery group, 1.04 in validation) and a similar bump in frailty risk. The score also tracked with declining kidney function, measured by estimated glomerular filtration rate, in both the discovery and validation cohorts.

A Mouth Swab Could Flag Cancer and Heart Attack Risk Early

The practical hook is prediction. Adding the OMAA score to standard risk models improved how well researchers could forecast two major health events: cancer risk prediction accuracy rose from an AUC of 0.67 to 0.70, and heart attack prediction rose from 0.76 to 0.79. Those are modest but real gains, and they came from a single mouth swab layered on top of the usual clinical variables. Specific bacteria did most of the work in the model: Capnocytophaga, Actinobacillus, and Stomatobaculum were among the strongest predictors, while Rothia tended to rise with age and Scardovia tended to fall.

For anyone who has watched a parent or grandparent age faster than their years, or slower, the idea that a swab could someday flag that trajectory before symptoms appear is the appeal here. It reframes the mouth as a genuinely diagnostic window into the rest of the body, not just a place where cavities happen.

Three Cohorts, Over 6,000 People, Validated the Score

What separates this from a typical single-sample microbiome study is the scale and the built-in replication. The researchers used a discovery cohort of 2,029 adults from NHANES 2009 to 2010, then confirmed the pattern in a separate validation cohort of 2,646 adults from NHANES 2011 to 2012, and again in an external, multi-country validation set of 1,293 adults aged 30 to 70. That is more than 6,000 people total, sampled and modeled independently, with the core mortality and frailty associations holding up each time. People with periodontal disease, pregnant participants, and anyone on recent antibiotics were excluded, so the score reflects a relatively "clean" oral microbiome signal rather than one confounded by active gum disease or recent drug exposure.

The age-prediction model itself was moderately accurate on its own terms, correlating with true age at a Spearman correlation of 0.44 in discovery, 0.35 in validation, and 0.22 in the external cohort, with a typical error of roughly 9 to 13 years. That is not precise enough to replace a birth certificate, but the goal was never predicting age for its own sake; it was capturing whose oral microbiome looks "older" than it should, and whether that gap predicts real health outcomes. On that count, the signal held.

Cross-Sectional Design Means No Proof of Cause Yet

The biggest caveat is also the simplest: this is a cross-sectional study, a single snapshot of each person's oral bacteria and their health outcomes over time, not a trial where oral bacteria were deliberately changed to see what happened. That means the OMAA score could reflect people's health influencing their oral microbiome just as easily as the reverse, or some third factor, like chronic inflammation or diet, driving both. The 16S sequencing method used also only identifies bacteria down to the genus level, not the exact species or strain, so the study can't yet say which specific microbes are doing the damage, or the protecting. And because the underlying NHANES cohorts are US-based, it's not yet clear how well the score would generalize to populations with very different diets, oral hygiene norms, or dental care access, though the external multi-country validation cohort is a reassuring first step in that direction.

Diet and medication use, two obvious confounders, showed only weak associations with OMAA in this analysis, for what that's worth; the one exception was a signal from cardiovascular medication use, which the researchers suspect reflects people already being sicker rather than the drugs themselves aging the microbiome.

What Comes Next for Oral Microbiome Aging Research

The researchers are explicit that OMAA isn't ready for clinical use. The next step is longitudinal data, tracking the same people's oral microbiomes over years to see whether OMAA changes before health declines, rather than alongside it, and whether interventions, better oral hygiene, dietary changes, or targeted probiotics, can actually lower a person's score. Interventional trials would also help settle whether the oral microbiome is a cause worth treating or simply a very sensitive bystander, reading out damage that's already happening elsewhere in the body. Given how cheap and noninvasive an oral swab is compared to blood draws or imaging, there's an obvious incentive to keep pushing this research toward something a primary care doctor could actually order.

The finding adds to a broader picture connecting microbes beyond the gut to whole-body aging; a related Nature Aging analysis of skin, oral, and gut microbiomes in older adults found similar links between microbial "reservoirs" and frailty, and a separate Scientific Reports study of oral microbial signatures in Canadian adults has reported comparable age-related shifts in oral bacterial composition. Taken together, they suggest the mouth may be one of the more accessible places to look for early signs of biological aging, if researchers can nail down the causal story.

References

1. (2026). Your mouth could hold the secrets to a longer and healthier life – The Australian. https://www.theaustralian.com.au/health/wellbeing/your-mouth-could-hold-the-secrets-to-a-longer-and-healthier-life/news-story/92166a0aa5ab6f6100159e79dbaac691
2. Jia-Jun Zhao, Ming Hu, Siyan Li, Qianqian Wang, Qiufen Mo, Huilin Yu (2026). Oral microbiome signatures predict biological age and host health. Nature Communications. https://doi.org/10.1038/s41467-026-72096-2
3. Peter J. Larson, Wei Zhou, Alba Santiago, et al. (2022). Associations of the skin, oral and gut microbiome with aging, frailty and infection risk reservoirs in older adults. Nature Aging. https://doi.org/10.1038/s43587-022-00287-9
4. Vanessa DeClercq, Robyn J. Wright, Jacob T. Nearing, Morgan G. I. Langille (2024). Oral microbial signatures associated with age and frailty in Canadian adults. Scientific Reports. https://doi.org/10.1038/s41598-024-60409-8