Can Wearables Detect Cortisol Levels? Understanding Stress Biomarkers
Wearables can detect cortisol when they include a dedicated biochemical sensor, and research devices have demonstrated cortisol measurement in human sweat. However, familiar smartwatch and smart-ring stress features typically estimate physiological stress from signals such as heart rate, heart-rate variability and skin conductance; they do not directly measure cortisol. Sweat-based cortisol monitoring is promising, but reliable everyday use requires validated sampling, selective sensors and careful interpretation of hormone timing. A stress score and a cortisol concentration answer different questions.
Your wearable flags a stressful afternoon. You remember a difficult meeting, so the result seems convincing. But did the device detect a rise in cortisol, or notice changes in your cardiovascular system?
That distinction matters.
“Stress tracking” can describe several different measurements, and a hormone result is not automatically a complete explanation of how someone feels. For users, understanding the difference helps make wearable data more useful. For product teams, it determines which sensors, algorithms and studies a stress-monitoring system actually needs.
What is Cortisol, and why is it called the Stress Hormone?
Cortisol is a steroid hormone produced by the adrenal glands. It helps regulate energy availability, immune activity and the body’s response to challenges. Its release is controlled through the hypothalamic-pituitar-adrenal axis, usually shortened to the HPA axis.¹
Calling cortisol the “stress hormone” is convenient, but incomplete. Cortisol supports normal daily functioning; it is not simply a substance that appears when something goes wrong. An increase can be part of an appropriate response to a demanding situation.
Cortisol also follows a daily rhythm. In people with conventional sleep schedules, levels are generally higher around the morning and lower at night. The cortisol awakening response describes a further rise during roughly the first 30–45 minutes after waking.²
Consequently, a morning reading cannot be interpreted in the same way as an evening reading. The useful question is often how a measurement relates to someone’s sleep schedule, baseline and recent activity.
What are Stress Biomarkers?
A stress biomarker is a measurable biological characteristic used to study some aspect of the stress response. Different biomarkers observe different systems, so they should not be treated as interchangeable.
Measurement | What it Captures | Main Interpretation Limit |
Cortisol concentration | A biochemical component of HPA-axis activity | Depends on timing, sample type and physiological context |
Heart rate and HRV | Patterns in cardiac activity and autonomic regulation | Influenced by exercise, breathing, posture and other factors |
Electrodermal activity (EDA) | Changes in skin conductance related to sweat-gland activity | Indicates sympathetic arousal rather than a specific emotion |
Skin temperature and movement | Peripheral and behavioural context | Neither independently identifies psychological stress |
These signals can complement one another. They do not provide a universal numerical scale for how stressed someone feels. The intended question should determine the measurement strategy.
How do Smartwatches and Smart Rings Measure Stress?
Many wearable stress features analyse heart rate and heart-rate variability, sometimes alongside temperature, movement or EDA. Optical PPG sensors record pulse-related changes in light at the skin. ECG sensors record cardiac electrical activity. Algorithms then derive features and compare them with patterns associated with physiological stress.
For example, Oura describes its Daytime Stress feature as using heart rate, HRV, motion and temperature, assessed against a personal baseline. Its documentation distinguishes physiological responses from emotions and explains that different stressors can produce similar results.³
EDA adds another perspective. Electrodes detect changes in skin conductance associated with sweat-gland activity. This is electrical sensing: it does not identify cortisol molecules in the sweat.⁴
A device can therefore register physiological activation during an enjoyable event, after exercise or during illness. That does not necessarily mean the person feels anxious.
These features can help users notice patterns. Their meaning depends on the signal quality, algorithm and context behind the label “stress”.
Can a Smartwatch Measure Cortisol Directly?
An ordinary optical heart-rate sensor does not directly measure cortisol. Neither does an HRV calculation. Direct cortisol sensing requires a measurement system that responds selectively to the hormone in a biological sample.
The wrist form factor itself is not the limitation. Researchers have built wrist-worn and skin-mounted systems incorporating biochemical sensors. The important distinction is the sensing mechanism and evidence behind the claim, rather than whether a product looks like a watch or patch.
For any claimed cortisol wearable, ask:
Does it measure cortisol molecules, or estimate a cortisol-related value from other signals?
Which fluid does it sample?
Has it been compared with a laboratory measurement of that fluid?
What conditions and participants were included in validation?
Predicted cortisol should be labelled as predicted. A correlation between HRV and cortisol in one study does not turn a pulse sensor into a hormone assay.
How do Wearable Sweat Cortisol Sensors Work?
Sweat is attractive because it can be accessed at the skin without drawing blood. However, collecting a useful sample and identifying cortisol within it requires more than placing an electrode against the body.
A typical research system brings together three functions.
Collect and route the sweat
A patch or microfluidic channel collects sweat and directs it towards the sensor. Some approaches use naturally produced sweat; others stimulate local sweating through iontophoresis, which applies a controlled electrical current with a sweat-inducing agent.⁵ The collection method affects when readings are available. A sensor demonstrated during exercise may not automatically work during a quiet office day.
Recognise cortisol selectively
Researchers have explored antibodies, aptamers and molecularly imprinted polymers. Antibodies bind target molecules. Aptamers are selected nucleic-acid sequences that recognise a target. Molecularly imprinted polymers contain binding sites formed around a template molecule. Each approach aims to distinguish cortisol from other substances in the sample. Selectivity must be tested, especially against structurally related compounds.
Convert binding into a usable result
Cortisol interacting with the sensing layer changes an electrical response. The device measures that response and uses calibration to estimate concentration. The system also needs fluid handling, electronics and data processing. A sensitive sensing material is an important starting point, but an everyday wearable must repeatedly deliver interpretable measurements under changing conditions.
Why is Continuous Cortisol Monitoring Difficult?
Detecting cortisol in a prepared sample is one challenge. Reliably following hormone changes while someone works, sleeps and moves is another.
Sweat availability and sample freshness
People do not continuously produce convenient volumes of sweat at every body location. Collection systems must account for low flow, evaporation and the mixing of older and newer samples. A 2024 study specifically investigated sweat self-collection for mental-stress evaluation, illustrating why fluid collection is an engineering problem in its own right.⁶ A device’s reporting interval should therefore reflect how quickly a fresh sample reaches the sensor, alongside the sensor’s own response time.
Sweat cortisol is not a direct substitute for blood cortisol
A sweat result describes the sampled fluid. Translating it into a blood concentration requires separate evidence. The pilot relationship reported by Torrente-Rodríguez and colleagues is encouraging, but correlation alone does not prove interchangeable measurements across individuals, activities or time points.⁷ For developers, it is helpful to separate two claims: accurately measuring cortisol in sweat, and using that measurement to infer another physiological quantity.
Sensor stability and repeated measurements
Real sweat contains salts and other compounds, while skin contact introduces contamination risks. Binding layers and electrodes must maintain performance across the intended wear period. Repeated tracking also requires the sensor to respond appropriately when concentrations fall. A device that accumulates a signal cannot automatically provide a reversible, continuously updated measurement.
Timing and interpretation
Even an analytically accurate measurement needs context. Cortisol’s daily rhythm and awakening response mean that a concentration without a timestamp and sleep context can be misleading. There is no single cortisol number that can universally label every user “calm” or “stressed.”
Does High Cortisol always mean Psychological Stress?
No. Cortisol contributes to normal physiology, and its interpretation depends on the sampling circumstances. A hormone result cannot identify the cause of a response on its own.
It is also important to distinguish everyday stress monitoring from investigating endocrine disease. Clinicians evaluating suspected Cushing’s syndrome may use tests such as late-night salivary cortisol, 24-hour urinary free cortisol and dexamethasone suppression testing. Selection and interpretation follow clinical protocols.⁸
A consumer stress graph cannot substitute for that assessment. Nor should an experimental sweat result be interpreted using a blood or saliva reference range. For readers, the practical message is to avoid diagnosing a hormone problem from a wearable score. Persistent symptoms deserve appropriate clinical evaluation.
What should Teams Validate before Building a Cortisol Wearable?
For product teams, a useful development plan separates three questions:
Does the device measure the intended analyte? Compare sensor results against an appropriate laboratory assay of the same sample type. Evaluate bias, repeatability, interference and performance across the intended concentration range.
Does it work during the intended activity? Test sample availability, skin contact, motion, temperature, wear duration and missing data. Include the settings in which the product will actually be used.
Does the result support the intended conclusion? If the claim concerns psychological stress, measuring cortisol accurately is only part of the evidence. The study must also establish how the output relates to that construct.
These are proposed development principles, not a universal certification checklist. Protocols should follow the intended use.
A stronger study might combine time-stamped cortisol samples, wearable physiology, activity records and brief self-reports. Participant-level evaluation matters: testing on data from people used during training can make a model appear more generalisable than it is.
The Future of Stress Monitoring: Combining Signals with Context
The most useful direction may be to combine biochemical information with cardiovascular signals, movement, sleep and personal context. Multimodal research already explores this approach. Each measurement can contribute a different perspective. Product design should preserve those distinctions and explain uncertainty, rather than compressing every response into an unexplained warning.
For consumers, wearable stress information can support reflection on recurring patterns. For researchers, cortisol sensing could enable new studies of hormone dynamics outside conventional sampling settings. Both applications depend on evidence that matches the claim.
Explore-Stress Biomarker Research with Sensio
Sensio welcomes research collaborations exploring new stress biomarkers and wearable monitoring approaches. If your team is investigating how biochemical signals could complement physiological sensing, connect with us to discuss the research question, device concept and validation pathway.
References:
professional, C. C. medical. (2026a, March 11). What does cortisol do?. Cleveland Clinic. https://my.clevelandclinic.org/health/articles/22187-cortisol
Tobias Stalder, Henrik Oster, James L Abelson, Katharina Huthsteiner, Tim Klucken, Angela Clow, The Cortisol Awakening Response: Regulation and Functional Significance, Endocrine Reviews, Volume 46, Issue 1, February 2025, Pages 43–59, https://doi.org/10.1210/endrev/bnae024
Daytime stress. Oura Member Care. (n.d.-a). https://support.ouraring.com/hc/en-us/articles/21205822135315-Daytime-Stress
Kyriakou, K., Resch, B., Sagl, G., Petutschnig, A., Werner, C., Niederseer, D., Liedlgruber, M., Wilhelm, F. H., Osborne, T., & Pykett, J. (2019). Detecting Moments of Stress from Measurements of Wearable Physiological Sensors. Sensors, 19(17), 3805. https://doi.org/10.3390/s19173805
Wang, B., Zhao, C., Wang, Z., Yang, K.-A., Cheng, X., Liu, W., Yu, W., Lin, S., Zhao, Y., Cheung, K. M., Lin, H., Hojaiji, H., Weiss, P. S., Stojanović, M. N., Tomiyama, A. J., Andrews, A. M., & Emaminejad, S. (2022). Wearable aptamer-field-effect transistor sensing system for noninvasive cortisol monitoring. Science Advances, 8(1). https://doi.org/10.1126/sciadv.abk0967
Wang, C., Wang, Z., Wei, W. et al. High-precision flexible sweat self-collection sensor for mental stress evaluation. npj Flex Electron 8, 47 (2024). https://doi.org/10.1038/s41528-024-00333-z
Torrente-Rodríguez, R. M., Tu, J., Yang, Y., Min, J., Wang, M., Song, Y., Yu, Y., Xu, C., Ye, C., IsHak, W. W., & Gao, W. (2020). Investigation of cortisol dynamics in human sweat using a graphene-based wireless mhealth system. Matter, 2(4), 921–937. https://doi.org/10.1016/j.matt.2020.01.021
U.S. Department of Health and Human Services. (n.d.). Cushing’s syndrome - NIDDK. National Institute of Diabetes and Digestive and Kidney Diseases. https://www.niddk.nih.gov/health-information/endocrine-diseases/cushings-syndrome
Comments