Pee could be a health goldmine. And to make the most of it, the world needs smart toilets.
That’s what a new, small-scale study published in Nature this month seeks to establish: whether regular urine collection and analysis of the thousands of telling, changing indicators in our pee can reliably serve up information about a person’s health.
The study’s University of Wisconsin authors, Dr. Joshua Coon and Dr. Ian Miller, say more research is still needed as the 10-day experiment only had two subjects: Coon and Miller themselves. However, they also think that a non-invasive device that could collect and analyze a person’s urine — like, say, a smart toilet — could work as a revolutionary tool for personalized, predictive health care.
Oh, and wouldn’t you know it, they’re making one such smart toilet themselves. But they’re not the only ones.
How Coon and Miller envision the smart toilet of the future.
Image: Dasom (Somi) Hwang
All that glitters is a golden shower
Recently, health-tech companies and even big tech players like Apple and Google have gotten into the business of harvesting health data, mostly from AI assistants, fitness trackers, smartwatches and even the smartphone in your hand. The current drumbeat in healthcare is that big data will be able to help us monitor our health in real time, as well as predict disease and health trends at a societal level in the future. And in this coming era of data-driven medicine, the study’s authors think that urine is an untapped resource.
Coon and Miller research metabolic systems at the University of Wisconsin and Morgridge Institute for Research (with work partially funded by the National Institute for General Medical Sciences). They analyze the molecules that reflect what our body has metabolized (called “metabolites”), whether that’s something we ingest, or is the consequence of an internal bodily process.
Researchers have already established that there are thousands of metabolites in urine, with many of them linked to our own habits like smoking, drinking, sleep, exercise, and even disease. As our habits change, so do the levels of these metabolites.
“There aren’t that many bodily fluids that you can get access to that have a window into health,” Dr. Michael Snyder, a Stanford University professor who studies biomedical data for healthcare, and is not involved in the study, said. “Urine is just one of those that it makes sense to be monitoring.”
The theory is that, on an individual level, regularly keeping track of your urine could give you information you can act on. For example, it could eventually tell if you have a urinary tract infection brewing before you even display symptoms, or if you’re getting enough exercise, drinking too much alcohol, or are registering inflammation that could cause disease down the line.
It could also serve as a compliment to the sort of digital healthcare many people are already doing, such as monitoring heartbeats with Apple Watches. Coon says that analyzing urine for lifestyle information could give a person a fuller vision of the why behind having heart troubles — something an Apple Watch alone can’t do.
Your pee could turn out to be somewhat of a “fingerprint.”
Even beyond what this data can tell about an individual, Coon says we don’t yet know what a large batch of this data from many people could tell us about disease outbreak, or other types of medical predictions down the line.
To tap into this potential, Coon and Miller tried to see if they could measure how information in urine corresponds with lifestyle choices, such as sleep or alcohol consumption. By keeping track of their habits, they were able to establish that specific metabolites do fluctuate based on lifestyle.
Your pee could also turn out to be somewhat of a “fingerprint.” The metabolic makeup of our urine varies from person to person — it looked different even just between the study’s two subjects! Knowing a person’s “baseline” levels could make personalized healthcare more precise. For example, if a person’s baseline changes, Miller says that that could serve as a warning sign, enabling “preventative treatment, instead of dealing with symptoms as they show up.”
Regularly comparing that baseline to changes is what particularly intrigues Snyder.
“This frequent measurement lets you really follow changes in [a person’s] health state, and that’s rarely done in medicine,” Snyder said. “Knowing your own personal health state is very, very important. And then finding shifts from that is the key to early detection of disease long before it’s symptomatic.”
The study presented a third finding: Collecting pee is not only time-consuming and resource-intensive — it’s annoying. Which is why they say the technology needs to catch up.
“It turns out it’s not convenient to collect every urine sample you make — we found that out,” Coon said. “But if you had a toilet that did that for you, that would make a difference.”
The smart toilet cometh
Coon’s research group is working on a smart toilet that does just that. They say they will have a prototype to demonstrate their ideas in Spring 2020, though they declined to share any preliminary images with Mashable. After that, it will be about getting funding, conducting clinical trials, and bringing it to market. It may sound like a long way off, but their vision is clear.
“The device that we are working toward is just the toilet,” Coon said. “It looks like a toilet, and it samples [urine] without you having to do anything. That is a way I think to be able to acquire data that will be helpful to people.”
Coons’ device would sample urine every time an individual uses the toilet, and has a built-in lab tool (called a mass spectrometer) to measure samples. That reading would go to a patient’s smartphone, where they could monitor their health in real time, or even send that info to doctors, or healthcare analysis companies.
An early model of Toto’s discontinued smart toilet.
Image: YOSHIKAZU TSUNO / AFP via Getty Images
Another company, Toto, built a smart toilet in the early 2000s that would allegedly “analyze sugar levels in urine, check blood pressure, body weight and even measure body temperature and hormone balance.” However, Toto discontinued it, owing to low demand.
Other companies including Panasonic and even Google have made moves to develop smart toilets. But they were all limited in function, collecting measurements like blood pressure, body fat, and blood in the urine. Coon sees his device as more far-reaching, with consequences for both the individual and society.
“The fascinating thing for me about this is that there are so many possibilities that one could imagine,” Coon said. “It really comes down to if one could build this device, and make these measurements on a large population, it would really, we think, open the doors to changing how we think about tracking health and lifestyle.”
Pushing the limits of pee?
Is Coon’s vision for his smart toilet realistic?
While Snyder agrees with the theory that regularly monitoring metabolites in urine can deliver health insights, he is skeptical that Coon’s 10-day, two-person study lays sufficient groundwork.
“I think we need larger studies to demonstrate utility,” Snyder said. “I don’t doubt in the long run they will be useful. The question is just how quickly, and I don’t know the answer to that.”
Other experts (and competitors) are even less gung-ho.
“I just don’t think it’s very practical,” Vik Kashyap, the founder of human waste analysis company Toi Labs, said. “They’re exploring interesting things, but I don’t know if they’re ever going to make their way into the real world.”
At Toi Labs, Kashyap and his team have been studying the best way to bring health tech into the bathroom for years. His company is currently conducting clinical trials for their own smart toilet product, True Loo. It uses AI to analyze pictures of stool and urine for health insights (what your poo looks like can apparently tell you a lot about your health!). Kashyap said that the choice to use photography instead of the “mass spectrometer” lab tools like Coon was intentional because he says that the tools and science Coon is relying upon have some major flaws in a commercial setting.
“We’ve been working on a smart toilet for more than a couple decades combined,” Kashyap said. “For this kind of approach to reach the market, it would be a very slim chance.”
Kashyap views mass spectrometers as large, expensive tools that work in the lab, “but don’t necessarily work in an everyday toilet environment.”
Kashyap is also not sold on Coon’s science. He says that mass spectrometers are good at looking for specific things, like in drug testing. But does not believe they are proven as well as they need to be for a true at-home smart toilet.
Other competitors and experts share that skepticism. Raja Dhir, the co-founder and co-CEO of microbiome research company Seed Health, doesn’t see the potential for urine testing without also testing blood and poop.
“The concept of continuous measurement of biological data is intriguing,” Dhir told Mashable. However, he’s not convinced that people will actually get much use of out of it in the real world.
Still, Coon, Miller, and even Snyder are optimistic that, with enough study, time, and investment they can prevent the future of healthcare from just being flushed down the drain.
“I’ve been saying that someday, we need a smart toilet,” Snyder said. “I think the day is here.”