Smartwatch data could widen cardiovascular research beyond biobanks
Cardiovascular disease remains the leading cause of death, and the slowdown in mortality gains has sharpened interest in data sources that can capture risk earlier and at scale. Smartwatches sit at the center of that idea. They continuously collect heart rate, heart rate variation, sleep metrics, and other physiologic signals in daily life, offering a stream of longitudinal data that traditional clinic visits rarely provide.
A single device on the wrist cannot replace a diagnosis, but it can document patterns that precede disease. In practice, that means a rising burden of arrhythmia, hypertension, heart failure, or ischemic events may become visible before symptoms drive a patient into care. The editorial argues that this type of data could complement existing resources such as UK Biobank, the All of Us Research Program, and the China Kadoorie Biobank, especially because those cohorts often draw from single countries and are not representative of the broader population.
The scale of the opportunity is hard to miss. Nearly a quarter of the world’s population owns a smartwatch, and adoption keeps growing. For cardiovascular research, that creates a potential global sensor network with far more day-to-day granularity than conventional epidemiology can offer.
Why wearable physiology matters for early cardiovascular signal detection
The appeal of smartwatch data is rooted in physiology. Photoplethysmography, the optical method used by most modern devices, detects blood volume changes in the skin’s microvasculature and converts them into heart rate estimates and related metrics. When those measurements are repeated over months or years, they can capture changes in autonomic tone, exertional response, sleep behavior, and resting hemodynamics that may reflect evolving cardiovascular stress.
Clinical plausibility is already visible in published reports. Smartwatch data have contributed to diagnoses of cardiomyopathy and atrial fibrillation, and a head-to-head comparison found high concordance between smartwatch-derived heart rate variation and high-resolution electrocardiogram-derived data in patients with cardiovascular disease. Those findings do not turn consumer devices into gold-standard diagnostics, but they do show that consumer-grade sensors can sometimes detect clinically relevant signal with enough fidelity to matter for research and, in some cases, care.
One sentence captures the promise and the problem: the device can see a lot, but only if the data can be trusted, shared, and interpreted correctly.
Standardization and privacy remain the largest barriers
The editorial makes clear that the biggest obstacle is not data scarcity but data governance. Smartwatch companies collect information under commercial terms, while research institutions operate under stricter ethical rules. Bringing those systems together requires more than a handshake agreement. It demands protocols for confidentiality, consent, data storage, access control, and cross-border compliance that can survive regulatory scrutiny.
Interoperability is already a bottleneck. In the Apple Heart & Movement Study, participants could link health records through a smartphone app, yet only 10% were able to do so because of interoperability problems between health records and smartwatch apps. That small figure reveals how fragile data linkage can be when platforms do not speak the same technical language. Without standard formats, broad population research becomes difficult to scale, and comparisons across manufacturers become even less reliable.
Device variation adds another layer of complexity. Different manufacturers use proprietary sensors and algorithms to derive metrics such as resting heart rate. Even when the underlying sensor is similar, the outputs may not be directly comparable. For researchers, that means longitudinal analyses can be distorted unless the devices are calibrated, validated, and standardized against shared reference methods.
Skin tone, cost, and age can distort who gets counted
Equity sits at the center of this editorial. Wearable research can only improve cardiovascular science if the people generating the data reflect the populations that bear the disease burden. At present, that is far from guaranteed.
Photoplethysmography may perform less effectively in darker skin tones, raising the risk that smartwatch data will be systematically less accurate in some groups than in others. If an algorithm reads one population more reliably than another, the resulting database embeds bias at the measurement stage. Correction algorithms and additional standardization are not optional technical refinements; they are necessary safeguards if wearable datasets are expected to inform equitable cardiovascular research.
Cost and usability also shape participation. More advanced smartwatches are more expensive, which can exclude lower-income households and skew datasets toward wealthier users. In one 2019 US snapshot, 31% of households buying a smartwatch had incomes above $75,000, while only 12% of households earning below $30,000 owned one. Age can be another barrier. Older adults, who face high cardiovascular risk, may be less willing or able to adopt newer and more complex devices, especially if the interface becomes less intuitive as device functionality expands.
These are not abstract concerns. In a British Heart Foundation public survey on smartwatch data for cardiovascular research, only 6% of 194 respondents identified as non-white. Even if that sample does not mirror the full owner base, it signals a real risk: research questions, algorithms, and interventions may be shaped by a narrow slice of the population while leaving others underrepresented.
Data ownership and consumer trust will determine scale
Ownership of wearable data is legally and ethically complicated. Users may be told they own their information and can delete it, yet company terms often reserve the right to control and share those same data. When research partners, manufacturers, hospitals, and public health agencies all want access, the circle of custodians widens quickly. Each added stakeholder increases the number of privacy questions that must be answered.
Consumer trust will depend on clarity. People need to understand how their data may be used, whether it could be tied to health records, and what protections exist against involuntary surveillance or unwanted medicalization. Dense terms and conditions are not enough. The editorial calls for concise, accessible communication so participants know what they are agreeing to before they share physiology that is deeply personal and continuously collected.
A practical route forward already exists in smaller form. The British Heart Foundation has explored linking smartphone and wearable information with National Health Service records, and Singapore’s Health Insights Singapore initiative uses smartwatches to assess health behaviors and lifestyles in residents. These programs show that collaboration is possible, but they also reveal how much infrastructure is needed to move from isolated projects to globally usable cardiovascular databases.
The next step is not more devices but better governance
Smartwatches could help researchers trace the early physiologic changes that precede cardiovascular disease and connect those changes to outcomes over time. That would strengthen prediction models, improve disease detection, and potentially identify windows for earlier intervention. The upside is substantial, especially in a field where mortality progress has stalled.
Still, the path from wrist sensor to research-grade cardiovascular database is narrow. Standardization across manufacturers, protection of privacy, equitable access in low- and middle-income countries, and accurate measurement across skin tones all have to be addressed before wearable data can be treated as broadly representative. Without those fixes, the technology will keep generating data, but much of it will remain difficult to trust, difficult to compare, and difficult to translate into fair cardiovascular research.