What can wearables measure in Parkinson's?
Most at-home research uses inertial sensors (accelerometers and gyroscopes) and splits into two groups: gait and motor symptoms. A 2021 systematic review found 24 papers that captured data in free-living conditions; 14 covered gait and 10 covered symptoms. For symptom assessment, the review reports that researchers achieved "accuracies of over 90% in a free-living environment." A 2025 review of real-world gait outcomes found that walking speed was the most-used measure, and that associations with motor severity "were inconsistent."
| Area | What studies measured | What the evidence says |
|---|---|---|
| Motor symptoms | Bradykinesia, tremor, dyskinesia, and motor state fluctuations in the on/off phenomenon (10 of 24 papers) | Accuracies over 90% reported in free-living settings (Sica 2021) |
| Gait | Walking (8 papers), turning (3), falls (2), physical activity (1); kinematic parameters such as gait cycle duration, step length and velocity | 14 of 24 papers covered gait (Sica 2021) |
| Real-world gait outcomes | 21 digital mobility outcomes across 27 reports: walking speed (92% of reports), step length (59%), stride time (48%) | Separated PD from controls; associations with motor severity inconsistent; "standardised protocols are needed for comparability" (Kirk 2025) |
Sources: Sica et al., PLoS One 2021; Kirk et al., NPJ Digit Med 2025.
FDA-cleared tools
FDA classifies tremor-measuring devices under product code GYD ("Transducer, Tremor", 21 CFR 882.1950, Class II). The table lists a subset of GYD 510(k) clearances related to Parkinson's, oldest first. It is not a complete list, and we have not summarized each device's intended use here: read the 510(k) summary on FDA's database for what each device is cleared to measure. A 510(k) clearance is not FDA approval, and it does not make a measure a qualified drug-development endpoint.
| Device | 510(k) number | Applicant | Decision date |
|---|---|---|---|
| Kinesia | K063872 | Cleveland Medical Devices, Inc. | 2007-04-06 |
| Personal KinetiGraph (PKG) System | K140086 | GKC Manufacturing Pty, Ltd. | 2014-08-22 |
| PKG System | K161717 | GKC Manufacturing | 2016-09-20 |
| PKG System Gen 2 Plus | K211887 | GKC Manufacturing | 2022-03-11 |
| Rune Labs Tremor Transducer System | K213519 | Rune Labs, Inc. | 2022-06-10 |
| NeuroRPM | K221772 | New Touch Digital, Inc. | 2023-03-17 |
| Sine Monitoring Parkinsons Disease Tremor Dyskinesia Apple Watch | K253055 | Nordic Kinetics AB | 2026-06-17 |
| Empatica Health Monitoring Platform for Parkinson's Disease Monitoring | K260374 | Empatica S.r.l. | 2026-08-14 |
Sources: openFDA classification, product code GYD; openFDA 510(k) records; check each number on the FDA 510(k) database. openFDA describes its data as unvalidated.
Validating a Parkinson's digital endpoint
The reviews point to the same gaps: studies use different sensors, placements and definitions, and gait outcomes do not track motor severity consistently. Our reading (editorial, not from a source) for teams planning a study:
- Pick the measure first. Decide which symptom or gait outcome matters to patients and to your claim before choosing the device.
- Test in free-living conditions. Lab accuracy may not carry over to daily life, which is where the reviewed studies focused.
- Standardize the protocol. Fix sensor placement, wear time and outcome definitions so results compare across sites and studies.
- Separate the regulatory question. A cleared device and a validated trial endpoint are different things; plan each on its own.
Sources: Sica et al., PLoS One 2021; Kirk et al., NPJ Digit Med 2025. This page is a summary, not medical or regulatory advice.
Related resources
Frequently asked questions
Can a wearable measure Parkinson's symptoms at home?
Research says yes for several symptoms. A 2021 systematic review of at-home inertial-sensor studies found work on bradykinesia, tremor, dyskinesia and on/off fluctuations, with accuracies over 90% reported in free-living settings.
Which gait measure is used most in real-world Parkinson's studies?
Walking speed. A 2025 systematic review found it in 92% of reports, followed by step length (59%) and stride time (48%).
Are there FDA-cleared wearables for Parkinson's?
Check FDA's own databases rather than a vendor's claim. Search FDA's 510(k), De Novo and PMA databases by product code (for example GYD, tremor transducer) or by device or company name, then read the cleared or authorized indication in the decision summary. A clearance covers only that stated indication; it is not FDA approval and does not make a measure a qualified trial endpoint.
Scope it in 30 minutes
Book a scoping call and we'll name the smallest engagement that moves you forward. Most teams start with the fixed-scope Initial Audit.
Book a 30-min scoping callPrefer to write? Send a short note.