Digital motor endpoint options for movement disorder trials, compared honestly
There are five ways to measure motor function in an essential tremor, Parkinson disease, ataxia, or dystonia trial. Each measures something different, asks something different of sites, and carries a different amount of regulatory precedent. We make one of them, so read the last row with that in mind; the other four are described from public information as of 2026 and we have tried to be fair to each.
| Option | What it measures | Hardware and site burden | Strengths and limits | Regulatory status | Best fit |
|---|---|---|---|---|---|
Clinician-rated scales TETRAS, MDS-UPDRS Part III, SARA, BFMDRS, AIMS | Ordinal severity ratings across many items, assigned by a trained rater at a visit | None Rater time, training and certification, central calibration for multi-site programs | Regulatory precedent, clinical interpretability, decades of trial use Limits: Inter-rater ICC around 0.74 to 0.85 for TETRAS; ordinal compression; composite dilution; scoring drift | Accepted primary endpoints | Keep as the anchor. Add precision beside it. |
Wearable accelerometry Wrist or finger sensors (e.g., Kinesia family), actigraphy platforms | Tremor, bradykinesia, and dyskinesia from limb motion; continuous or task-based | Body-worn sensors per patient Device logistics, charging, compliance, returns; task-based versions need supervision | FDA-cleared devices exist; continuous monitoring captures fluctuation; large publication history Limits: Captures motion only; cannot see pressure, pen tilt, geometry, or motor planning; hardware dependence | Cleared as devices; used as exploratory and secondary endpoints; not DDT-qualified | Continuous monitoring of fluctuations in PD; less suited to fine-motor mechanism endpoints. |
Smartphone active tests Pharma-built apps (e.g., Roche navify and Floodlight, Konectom) | Batteries of short tasks: tapping, drawing, gait, speech, cognition | Patient smartphone or provisioned phone App onboarding, adherence management, device heterogeneity | Frequent sampling; used as a primary endpoint in at least one Phase 2 PD study Limits: Built around internal pipelines; limited external licensing; consumer device variability | Sponsor-internal endpoints; qualification pending | Large-sponsor programs with internal digital teams. |
Device-agnostic platforms Multi-sensor integration and analytics vendors (e.g., Koneksa, Medidata Sensor Cloud) | Whatever the attached sensors capture: activity, gait, sleep, vital signs | Varies by study Depends on sensor choice; enterprise procurement | Broad therapeutic coverage; pharma relationships; data management infrastructure Limits: Generalist; motor-disorder-specific measurement depends on the sensor selected; enterprise pricing | No qualified endpoints; extensive exploratory use | Multi-domain programs needing one data pipeline for many sensors. |
Digitized spiral drawing (NeuroQuantix) NeuroQuantix by Cerebral Innovations | One kinematic endpoint specified a priori for the mechanism (tremor frequency or amplitude, velocity, pressure, pen tilt, geometry, or smoothness) from a brief spiral drawing task | iPad Pro or Wacom digitizer with stylus; no wearables Under 30 minutes site training; a brief assessment at existing visits | Evidence database of 200+ published studies, methodology standardized from a scoping review; test-retest ICC up to 0.97 in ET for the implemented metric; captures dimensions accelerometers cannot; built to 21 CFR Part 11 and EU GMP Annex 11 Limits: Upper-limb fine motor only (not gait) | Fit-for-purpose evidence assembled for the FDA COA framework; exploratory and secondary endpoint use within an IND | Fine-motor and tremor mechanisms in ET, PD, ataxia, dystonia; small populations where enrollment math is decisive. |
Product names are the property of their owners. Descriptions of other vendors are drawn from their public materials and the published literature; corrections are welcome at david@cerebral-innovations.com.
How to choose
Start from the mechanism
A frequency-specific tremor drug needs a frequency metric. A bradykinesia drug needs velocity. A gait drug needs a gait measure. Choose the instrument that measures the thing your compound changes.
Weigh reliability over novelty
Enrollment is set by measurement error. Ask each vendor for test-retest ICC in your population and put it in the sample size calculator.
Do not swap the anchor
Keep the rated scale for precedent and add one objective measure as an exploratory endpoint. Promote it with within-program evidence.
Frequently asked
What are the main options for measuring motor function in a movement disorder trial?
Five: clinician-rated scales (TETRAS, MDS-UPDRS, SARA), wearable accelerometry, smartphone active-test batteries, device-agnostic sensor platforms, and digitized spiral drawing analysis. They differ in what they measure, what hardware and site effort they require, and how much regulatory precedent they carry. Most well-designed programs keep a rated scale as the anchor and add one objective measure matched to the mechanism.
How does digitized spiral drawing differ from a wearable sensor?
A wearable measures limb motion; a digitized spiral measures a goal-directed fine-motor task including pressure, pen tilt, drawing geometry, smoothness, and tremor spectra, none of which an accelerometer can see. A wearable can monitor continuously at home; a spiral is a brief, standardized assessment at a visit. They answer different questions.
Which option has regulatory acceptance?
Clinician-rated scales are accepted primary endpoints. Several wearable devices are FDA-cleared as devices, which is a device decision rather than an endpoint decision. Digital endpoints enter programs as exploratory or secondary endpoints within an IND with fit-for-purpose evidence for the context of use, per the FDA clinical outcome assessment framework; EMA has qualified stride velocity in Duchenne as a primary endpoint, which shows where the path leads.
Which option is right for a small or rare-disease population?
The one with the highest reliability for the mechanism you are treating, because reliability drives the enrollment math. For upper-limb tremor and fine-motor mechanisms, published spiral metrics reach ICC 0.97 and support as few as 9 subjects per arm at 95% power in the published example. For gait or axial mechanisms, a wearable or smartphone gait measure is the better fit.