Science1 publisher2 min readPublished
Drug trials collecting wearable data climbed to 128 in 2024 from fewer than 30 a decade earlier
Researchers at Mount Sinai counted drug trials collecting wearable data on ClinicalTrials.gov, and ten years of growth turned out to be gradual, with the researchers quoted by Nature split on whether sensor capability or validation work is the holdup.
The Scientist · Science desk
What happened
- Researchers at the Icahn School of Medicine at Mount Sinai searched ClinicalTrials.gov and found drug trials collecting data from wearables rose from fewer than 30 in 2014 to 128 in 2024.
- Others in the same feature say technological capability is not the obstacle, pointing to measurement and validation, regulation, and concerns about privacy, data security and big tech's role in health care.
- Martin Cowie, a vice-president at AstraZeneca, said of wearables in trials that "it's about a lot more than collecting data".
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- decision A sponsor choosing where the wearables budget goes has to decide between better hardware and the work of proving a measure means something. On the validation account, it is the proving work that holds up the trial.
- constraint Teams working in the 38.0% of smartwatch use outside cardiology, neurology and oncology have fewer prior trials to point at, so they carry more of the measurement work themselves.
- exposure Participants' physiological data sits with device makers whose main business is not clinical research. Consent and data-security terms sit on the trial protocol alongside the endpoints.
- precedent A rise of a bit over fourfold in ten years is the base rate a planner now has for the next five. That is a long way below the fully remote, dose-personalizing study the feature imagines for 2036.
The Mount Sinai search is a count of studies. It found 128 drug trials collecting wearable data in 2024 and fewer than 30 in 2014 [1]. That is more than a fourfold rise over ten years [13]. Turning 128 into a percentage of drug development would need the registry's total for 2024, and the feature does not report it.
The 2025 review of smartwatches in trials shows where the devices went. Cardiology accounted for 28.7% of device use, neurology 21.8% and oncology 11.5% [2]. Those three add to 62.0%, leaving 38.0% spread across every other area [14]. Part of the reason is the order in which sensors reached consumers. The sensors a device carries define the medical areas where it gets used [12]. Cardiology led in part because heart-rate tracking through photoplethysmography was built into many early fitness bands and smartwatches [10].
Two accounts of the holdup sit in the same article. Some sensor-technology researchers expect their work to expand what wearables can be used for and so drive greater adoption [3]. Others say technological capability is not the obstacle. They point to measurement and validation, to regulation, and to ethical concerns over privacy, data security and the expanding role of big tech companies in health care [4]. Martin Cowie, a vice-president at AstraZeneca in Cambridge, UK, said "I'm an enthusiast for the technology" [5], and that "it's about a lot more than collecting data" [6].
Nature reports the disagreement and no data that would settle it. Prachi Desai is a clinical data management specialist at the contract research organization Quanticate. She said: "If we can empower patients to do trials in their own communities, we can get better data and reduced dropout rates, and achieve cost savings" [7]. Better data, lower dropout and cost savings are what the design is meant to deliver, and the article reports them as an expectation.
A registry search records that wearable data were collected [1]. Whether a regulator accepted the resulting measure as evidence is a different question. That question decides whether a sponsor spends next on hardware or on validating what the hardware produces. On the second account of the holdup, validation is where the spending has to go [4]. The surge of interest followed rapid growth in the wearables market [8]. Ten years on, Nature reports that use in trials has risen gradually and that wearables as a key clinical aid are still some way from reality [9].
What to watch
- An updated ClinicalTrials.gov count for 2025 and 2026 would show whether the rise continued past 128 or stalled.
- Published dropout and cost comparisons between wearable-enabled and site-based arms would test the case Desai makes for community trials.
- A wearable-derived measure accepted as a primary endpoint in a regulatory submission would give the validation argument a decided case to cite instead of an opinion.