What can you learn about a person’s physiology without breaking through their skin?
The answer, for most of medical history, has been: not much, and not continuously. Breath carries a narrow slice of volatile chemistry. Urine integrates hours of activity into one delayed number. Saliva is contaminated by whatever a person last ate, drank, or said. Tears carry real information in a volume too small to work with. Every non-invasive fluid trades something away.
Sweat is the exception, and the reason isn’t one property. It’s three, stacked together. It’s secreted continuously. It’s secreted under a known, controllable stimulus. And it carries a genuinely broad chemistry (electrolytes, metabolites, proteins, and hormones) rather than a single class of molecule. No other fluid you can access without a needle combines all three. That combination is what makes sweat a category of its own, not just another biofluid on the list.
Continuous and on-demand is a rare combination
Blood is rich but episodic. You get a value at the moment of the draw, and nothing about the six hours before it. Urine is continuous in the sense that it accumulates, but it reports an average, not a moment; a change at 2pm doesn’t show up until the next void. Interstitial fluid is chemically similar to blood and genuinely promising, but reading it requires a microneedle, which puts you back through the skin barrier the whole premise was supposed to avoid.
Sweat does something none of those do: it can be actively stimulated (thermally, through exertion, or pharmacologically via iontophoresis) and then read while it’s being produced. That’s the difference between a snapshot and a signal. You’re not waiting for a fluid to accumulate somewhere and hoping to catch it in time. You’re triggering production and reading it as it happens.
The chemistry gets in through at least four different doors
The oldest misconception about sweat is that it’s a dilute filtrate of blood, plasma minus some water. It isn’t. Analytes reach sweat by mechanistically distinct routes, and the route determines what a given analyte is actually good for measuring.
Active transport with reabsorption. Sodium and chloride are secreted into the duct, then partially reclaimed on the way out through dedicated channels. This reabsorption step is precisely the mechanism that malfunctions in cystic fibrosis, which is why sweat chloride became a diagnostic test in the first place.
Passive diffusion. Potassium, cortisol, creatinine, urea, and ethanol cross membranes along a concentration gradient, arriving in sweat roughly in proportion to their level elsewhere in the body.
Ionization trapping. Ammonia diffuses into sweat’s more acidic environment, gets protonated, and then can’t diffuse back out. The result: sweat ammonium concentrations run 20 to 50 times higher than plasma, not because the body is producing more, but because the fluid itself is a chemical trap for that molecule.
Local generation. The sweat gland doesn’t just pass chemistry through. It produces some of it. Sweat lactate is measurable at concentrations far above blood lactate, generated in part by the gland’s own metabolic activity rather than reflecting systemic levels alone.
Four different mechanisms, four different kinds of information, in one fluid you can access from the surface of the skin. That’s the actual argument for sweat’s density. It is not that it happens to contain a lot of molecules, but that those molecules arrive by pathways rich enough to be individually informative.
Decades of Research Support Sweat Sensing
It is worth being precise about what “information-dense” does not mean: it does not mean that sweat sensing is new or unproven. Sweat chloride has been a standardized, regulated diagnostic test for cystic fibrosis since Gibson and Cooke formalized pilocarpine iontophoresis in the late 1950s. More than six decades later, it remains the clinical gold standard, with defined thresholds and a defined minimum sample volume.
A biofluid rigorous enough to anchor a diagnostic test for more than six decades was never the uncertain part of this story. What is uncertain is how much further that same fluid can be pushed now that the constraint has shifted from the biology of sweat to the engineering of how we miniaturize biosensors and collect, measure, and interpret sweat.
The chemistry was never the bottleneck. Access is.
If sweat has been diagnostic-grade since 1960s, the obvious question is why it produced exactly one mainstream clinical use in the six decades that followed. The answer has little to do with the molecules themselves. The problem was everything downstream of the chemistry: stimulation, timed collection, adequate sample volume, transport, instrumentation, and a technician trained to execute the entire process. It all remained manual, labor-intensive, and largely confined to the hospital. The fluid was validated. The method of reading it never left the clinic.
That is an engineering problem, not a biological ceiling. And it is precisely the problem that has begun to be solved over the last several years: microfluidic collection small enough to wear, chemistry that can be read directly on the skin rather than in a laboratory, and a data platform that turns a patch into a continuous stream rather than a single laboratory value.
The molecules didn’t change. What changed was what a person could do with them.
The company built to read it
This is where Epicore fits, and deliberately not before now, because the case for sweat as a category doesn’t need a vendor to make it. It needs recognition that access, not chemistry, was always the constraint.
Epicore has spent the years since its founding solving the access problem specifically: microfluidic wearables that survive a twelve-hour shift instead of a clinic visit, collection engineering that accounts for the flow-rate dependence described above rather than ignoring it, and an enterprise data platform that turns a patch into a continuous, individual-level stream rather than a single lab value. That infrastructure now runs two ways: deployed today, at scale, measuring sweat sodium and chloride across industrial, defense, and athletic performance settings through Connected Hydration, and extended into clinical and translational research through Discovery Patch, whose own sweat-collection work has independently catalogued more than 800 proteins and 32,000 endogenous peptides in human sweat.
Breadth in the literature was never the constraint. The constraint was building something that could read a fluid this information-dense outside of a hospital, continuously, on a real person, at scale. That’s the part of the sweat story that just changed, and it’s the part worth paying attention to.
FAQ
FREQUENTLY ASKED QUESTIONS
Sweat is the only fluid you can access without a needle that is produced continuously, can be stimulated on demand, and carries broad chemistry across electrolytes, metabolites, proteins and hormones. Most non-invasive fluids offer one or two of those properties. Sweat offers all three, which is what makes it a category rather than another item on a list.
Sweat reports how much fluid and salt your body is losing in real time, and it carries markers reflecting activity elsewhere in the body. Sodium and chloride indicate hydration and electrolyte status. Research has identified markers that distinguish why a stress response is occurring, whether from heat, humidity or exertion, rather than simply that one is underway.
Sweat contains electrolytes such as sodium, chloride and potassium, metabolites including lactate, urea, creatinine and ammonia, and hormones such as cortisol. Research has catalogued more than 800 proteins and 32,000 endogenous peptides in human sweat. Breadth of chemistry has never been the limiting factor.
By at least four distinct routes, and the route shapes what each analyte is useful for. Sodium and chloride are actively transported into the duct then partly reabsorbed. Potassium, cortisol, creatinine and urea diffuse passively. Ammonia is chemically trapped, reaching 20 to 50 times plasma concentration. Lactate is generated by the gland itself.
Blood carries more chemistry, but sampling it means breaking the skin and you get one instant per draw. Sweat carries less, but it can be read continuously while it is being produced, and its rate of production is itself a useful signal. They answer different questions rather than competing to answer the same one.
Each of the alternatives trades something important away. Saliva is readily contaminated by food and drink. Urine reports an average over hours rather than a moment, and reports it late. Tears carry real information in volumes too small to work with reliably. Interstitial fluid is chemically close to blood but needs a microneedle.
The sweat chloride test is the clinical gold standard for diagnosing cystic fibrosis, with defined diagnostic thresholds and a defined minimum sample volume. Sweat measurement is also used to assess hydration and electrolyte status in athletic, industrial, defence and clinical research settings.
Iontophoresis uses a small electrical current to drive a stimulating agent, usually pilocarpine, into the skin so that a specific patch of skin sweats on demand. It allows sweat to be collected without exercise or heat exposure, and it is the mechanism behind the standard cystic fibrosis diagnostic test.
Since the late 1950s, when Gibson and Cooke formalised pilocarpine iontophoresis for cystic fibrosis diagnosis. More than six decades later it remains the clinical gold standard for that condition. The chemistry has long been settled; what changed recently is the ability to read it outside a hospital, continuously, on a moving person.
References
- Zhou, M., Yang, D.S., Kim, J., et al. A flexible wearable system for integrated sweat stimulation and chloride analysis in cystic fibrosis diagnosis. npj Digital Medicine, 2026. doi:10.1038/s41746-026-03056-9
- Spinelli, J.C., et al. Wearable microfluidic biosensors with haptic feedback for continuous monitoring of hydration biomarkers in workers. npj Digit. Med. 8:76, 2025.
- Ghaffari, R., et al. The Gx Sweat Patch for personalized hydration management. Nat Rev Bioeng 1:5–7, 2023.
- Brasier, N., et al. The potential of wearable sweat sensors in heart failure management. Nat Electron 7:182–184, 2024.
- Yang, D.S., Ghaffari, R., Rogers, J.A. Sweat as a diagnostic biofluid. Science 379:760–761, 2023.