Wearables proved people would wear them. Now they must prove health outcomes.

Twenty years ago, the pitch for wearable health technology was simple: put a sensor on the body, show people their data, and better health would follow. The initial promise of self monitoring and data for personal decision-making got the hardware on people's bodies and the data flowing. Today over one billion consumer wearable devices are in use worldwide. Hospitals are connecting ring and wristable data directly to electronic health records. A survey by Rock Health found that 54% of respondents track at least one health metric digitally. It’s safe to say that wearable health trackers have become ingrained in how many people live their lives and manage their health and wellness.

However, there are still many open questions around who these devices work best for, when the data is meaningful, under which circumstances, and how they influence long term health and reduce costs. We, as an industry, are still investigating if healthy behavior quantified by wearables is linked to long-term prevention of morbidity or mortality. We are also in the early stages of identifying the right use cases for wearable data to manage chronic health conditions or detect acute physiological changes for earlier clinical interventions to reduce medical spend and improve patient health.

We have reached a tipping point. It’s time to move from studying the acceptability of wearables (market proliferation has answered that question) and start hypothesizing and learning about how the devices and their associated data can drive meaningful health outcomes.

Wearables evidence: What we know

Let's begin by briefly reviewing what we currently know about digital health devices and health outcomes. The largest body of evidence to date centers on remote patient monitoring and activity tracking, while research on chronic condition management and infection identification is still emerging.

Remote Patient Monitoring

A 2025 scoping review in PLOS Digital Health examined 80 studies of wearables used for remote health monitoring outside the hospital. Six (8%) studies were randomized controlled trials (RCTs.) Of those six, which focused on diabetes, respiratory disorders, and neurodegenerative disorders, four reported a positive clinical effect, including evidence of improved patient outcomes associated with (largely medical grade) device use. These findings represent multiple scientific opportunities, from moving beyond medical or research grade devices to less expensive and more ubiquitous consumer health devices, as well as using personal baseline data collected prior to hospital stays to inform personalized recovery plans.

Activity Tracking

Multiple studies show that activity trackers consistently get people moving more, and we know that hitting 7,000+ steps per day is associated with lower risk for all-cause mortality, cardiovascular disease, type 2 diabetes, and dementia when compared to sedentary adults. A 2025 systematic review and meta-analysis of 23 randomized trials in older adults found that activity-tracker interventions produced a real, measurable increase in physical activity time and daily step counts immediately after the intervention. In adults with type 2 diabetes, wearable-based activity interventions increased step counts by around 1,583 steps per day, yet produced no significant change in HbA1c or fasting blood glucose. Improvements were more likely in people who started with poorly controlled diabetes (HbA1c above 8.5%) and in programs lasting at least 10–13 weeks. The takeaway: wearables are a solid tool for building activity habits, but we still have a lot to learn about how to combine activity monitoring with personalized counseling, diet changes, and clinical care to reach real clinical improvements like reduction in HbA1c in populations with diabetes.

Mental Health

Emerging evidence suggests that wearable technology can improve mental health. A 2026 systematic review and meta-analysis in the Journal of Affective Disorders examined 32 randomized controlled trials involving 3,061 adults to ask a straightforward question: do wearable device-based interventions actually improve depression, anxiety, and quality of life? The answer was yes on all three counts. Wearable-based interventions produced significant improvements in depression, anxiety, and quality of life. Notably, the effect on depression was even stronger among participants who also had a metabolic condition like diabetes or obesity, hinting that wearables may work best when they're doing double duty, both tracking a physical health issue while nudging mental health in the right direction at the same time.

Chronic condition management

Two studies out of Mount Sinai led by Robert Hirten found that wearable data including heart rate, heart rate variability, and activity, can predict inflammatory flares four to seven weeks before symptoms appear, in both rheumatoid arthritis (preprint) and inflammatory bowel disease patients. These findings suggest the potential use of wearable devices for disease monitoring, flare prediction, and better patient medical management to reduce the impact that these flares have on daily life.

Outside of academia, the CMS ACCESS program aims to test an outcome-aligned payment approach in Original Medicare to expand access to new technology-supported care options that help people improve their health and prevent and manage chronic disease. US Government-level attention further underscores that the time is now to double down on the use of wearable health technology to improve population health at scale.

Acute infection signals

The Covid-19 pandemic led to multiple research studies demonstrating the value of wearable devices in identifying signs of infection. The TemPredict Study collected daily questionnaire data and physiological data from Oura Ring to train an algorithm for illness detection. The resulting algorithm identified COVID-19 an average of 2.75 days before participants sought diagnostic testing. A later systematic review, Wearable Devices to Diagnose and Monitor the Progression of COVID-19 Through Heart Rate Variability Measurement: Systematic Review and Meta-Analysis (2023) concluded that wearable devices can help identify COVID-19 during the presymptomatic period as well as track its worsening. These studies have indicated that changes in heart rate, heart rate variability, and temperature can signal signs of acute infection and features like Symptom Radar, can help individuals recognize when they may be getting sick.

Where the Evidence Needs to Go Next

Taken together, this body of research tells a consistent story: wearables reliably measure activity and biometrics, and there is evidence that they can increase step counts. I believe that the future of wearables research should focus on measuring outcomes: both improvement of health and medical cost reduction. I want the data to be used by the right people, at the right time, to make the right decisions. The next generation of evidence requires a few concrete shifts:

• The field needs to test wearables as part of a stack rather than as a stand-alone intervention. The diabetes findings in particular suggest that step-tracking alone won't move HbA1c, but step-tracking paired with counseling, dietary support, or clinical follow-up might, and we need trials designed explicitly to test that combination rather than the device in isolation. We should also imagine how to scale these types of interventions with AI and telehealth so they can reach more people.
• Research needs to shift from medical-grade to consumer-grade devices. I think it’s safe to say that the majority of people using a wearable device to track their health are using a consumer device, so this is what we should measure. Additionally, understanding individual biometrics prior to a health event enables us to personalize care in a way that a device provided as part of a trial at the start of an intervention never can.
• Longer and more diverse trials are needed. Many of the studies above ran 10-13 weeks at best, which may be too short to detect changes in slower-moving outcomes like HbA1c or cardiovascular risk, and most study populations remain narrow in age, geography, and health status, limiting how confidently we can generalize the results. National or global cohorts using the same device can broaden our understanding of human physiology, improve algorithm development, and enhance generalizability. The All of Us Research Study is a good example of generating a large, diverse data set, but there are still limitations with the device participants can bring to the study.
• Finally, as programs like CMS ACCESS start testing outcome-aligned payment models, the research community has a real opportunity to embed rigorous evaluation directly into real-world deployment by turning large-scale rollout into large-scale evidence generation, rather than treating research and reimbursement as separate tracks. Oura is proud to be participating in this program with our partner, Counsel Health.

Oura’s approach to health outcomes research

At Oura, we are committed to an integrated research strategy, starting with device and algorithm validation through to health and financial outcomes research.

An example of the validation to outcomes study pipeline is our work in ovulation detection. In 2025, our science team, together with outside researchers, published a study in the Journal of Medical Internet Research testing the algorithm behind our Fertile Window feature against the traditional calendar method of ovulation prediction, using data from 1,155 ovulatory cycles across 964 de-identified members, including members with irregular cycles. The physiology-based method detected 96.4% of ovulations with an average error of ±1.26 days, compared to ±3.44 days for the calendar method. For people with irregular cycles specifically, 82% of our estimates landed within two days of the reference date, versus 32.5% for the calendar method.

This year we launched an IRB-approved prospective study with Maven Clinic to examine whether continuous wearable biometrics including temperature, sleep, and cycle data can meaningfully support people who are trying to conceive. Fertility research has historically relied on sparse, periodic data points: a basal body temperature reading here, a cycle-length estimate there. We think continuous Oura Ring data can lay a better foundation to help people plan for and build the families they want. The ovulation study is one example of the model we are building: device and algorithm validation followed by studying clinical use cases and validity.

The wearables industry spent two decades proving people would wear the devices. The next decade will be defined by whether we can prove the data makes people healthier, reduces costs, and improves efficiencies in our healthcare system. I don't want Oura's data to just describe people's lives; I want it to keep them healthier, longer, in a way that matters to each person individually. That's the work ahead.