Lying about your age might not do you any good if your microbiome gives you away. New research out in the journal mSystems lays out how the crowds of bacteria teeming on and in our bodies change as we get older. The researchers found that skin microbes in particular change so consistently over time that a sample of them can be used to predict someone’s age within about four years.

For now, the correlation is a fun tool. But down the line, whether or not someone’s microbiome actually aligns with their real age might be useful when it comes to assessing their overall health, says study co-author Shi Huang, an engineer at the University of California San Diego. “Our assumption is that if age is a key driver of certain disease, then the [microbiome] markers we can ID here can indicate diseases,” he says.

Your Microbial Age

Our microbiomes are changing all the time. Dietary shifts can overhaul what lives in your gut, and aging can change what lives on your skin, too. It even shifts after we die: Bacteria fluctuate so predictably on bodies that forensic anthropologists study the microbiome of the deceased to calculate time since death. 

Those precise post-mortem estimates got Huang and his coworkers thinking: What if microbiome samples from a living person could indicate how old they are? The team put this theory to the test with nearly 9,000 fecal, saliva and skin swab samples from people in the U.S., U.K., Tanzania and China collected as part of other microbiome projects. They used data on what and how many bacteria were there in the samples to build a computer model that compared each person’s age to the types and number of microbes they had. 

Working backwards, the team tested the data to see how well microbiome alone predicted age. Poop microbes predicted real age within about 11.5 years, while skin microbes did the best by predicting within about four years.

Diagnosing With Microbes

Though the accuracy window could be narrower, Huang says that he and his team aren’t as interested in making the estimate more precise. The team is more interested in what it means for people who have a microbiome of someone nowhere near them in age. “The big question here could be, if you know your microbiome, what is the difference between yours and what’s normal?” Huang asks.

With these comparisons in hand, Huang and his colleagues plan to study if microbiomes change in response to aging-related illnesses. If so, then a microbiome analysis could reveal a health condition someone might have. 

That possibility is in the distant future, Huang says. There are all kinds of conditions influencing the microbiome that the team couldn’t control for in this study, like whether or not participants had traveled recently.

Additionally, this research team didn’t include individuals with health conditions in their dataset. To see if microbiomes change with age-related disease, then the team has to first find out what the "age" of a microbiome is on people with health problems, Huang says.