Research

The evidence Kampa is built on.

Every measurement in Kampa traces to published work, and the papers below are the ones that actually changed what we built - what test to ship, what threshold counts as a real change, what not to build at all. Where the evidence doesn't support a claim, Kampa doesn't make it.

Measuring slowness of movement

The CloudUPDRS smartphone software in Parkinson's study: cross-validation against blinded human raters npj Parkinson's Disease, 2020

The only study that runs 16 smartphone tests against the same blinded raters - 60 subjects, 990 tests, 2,628 blinded video MDS-UPDRS III ratings. Hand rotation predicted its clinical sub-item 74.6% of the time; finger tapping managed 53.2%. The authors note explicitly that this contradicts the field's habit of building tapping-only apps.

What Kampa did with it: shipped hand rotation alongside finger tapping rather than tapping alone.

What we changed our mind about: the same table scores a tremor test at 97%, and we originally refused to build one on the grounds that the Apple Watch already measures tremor passively. That reasoning was wrong in two specific ways. A watch is worn on one wrist, so tremor in the other hand is invisible to it. And it returns a single number, which cannot separate the three kinds of tremor a clinician scores separately - at rest, holding a position, and during movement. Those are different symptoms. Kampa now measures all three directly.

Reliability and validity of the Roche PD Mobile Application for remote monitoring of early Parkinson's disease Scientific Reports, 2022

316 recently diagnosed, treatment-naive participants ran a short active-test battery at home every day. Adherence was 96.3%, the battery took a median of 5.3 minutes a day, and the pre-specified sensor features held up on retest - ICC at or above 0.75, median 0.9.

What Kampa did with it: this is the evidence that a short at-home check is something people will actually do, and that what comes out is stable enough to compare against itself over time. It is why a movement check exists at all. Kampa's version is seconds rather than a daily five-minute battery, and it is entirely optional.

What it does not show: ICC is test-retest reliability - the measure is consistent, not necessarily a measure of disease severity. Those are separate questions and we treat them separately.

Sensor verification and analytical validation of algorithms to measure gait and balance and pronation/supination Sensors, 2022

Turn counts from the phone's gyroscope agreed with human raters counting turns by eye at ICC 0.935, and held at 0.97-0.99 even under deliberate perturbation - arm raised and lowered, stopping and starting, rotating about several axes.

What Kampa did with it: adopted the protocol - phone held in the hand, arm out, screen palm-up, rotate palm-up to palm-down - and the gyroscope-plus-PCA approach that finds the true axis of rotation, which is what makes the measurement survive someone who doesn't hold a clean axis at home.

What it does not show: this validation was run in healthy volunteers. It establishes that the algorithm measures rotation correctly, not that the measurement tracks Parkinson's severity - that is what the two studies above are for.

Quantification of limb bradykinesia in patients with Parkinson's disease using a gyrosensor International Journal of Precision Engineering and Manufacturing, 2011

Rotation measured by gyroscope correlated at r = -0.78 with the hand-rotation bradykinesia sub-score, and correlated better with that sub-score than with others.

What Kampa did with it: treated rotation as a measure of the specific thing it claims to measure, rather than a proxy for general severity.

Smartphone-based estimation of item 3.8 of the MDS-UPDRS-III for assessing leg agility IEEE Open Journal of Engineering in Medicine and Biology, 2020

A phone strapped to the thigh scored r = 0.92 against clinicians - higher than the clinicians' agreement with each other, at 0.88. Held against the thigh by hand instead, accuracy drops to the 63-70% range.

What Kampa did with it: built the strapped version, and asks for a strap or a snug pocket at mid-thigh rather than pretending the hand-held variant is equivalent. Holding the phone puts your hand's own tremor into a leg measurement - a problem the literature does not address - so it is offered as a fallback and recorded separately, never mixed into the same trend line.

What it does not show: that r = 0.92 belongs to the study, not to Kampa. It was measured with a research sensor sampling twice as fast as a phone and scored over dozens of engineered features we do not compute. We report four plain quantities - how high the thigh lifted, how many steps, whether they stayed as high at the end, and the longest pause - and attach no correlation to them.

Measuring tremor on demand

Assessing Parkinson's rest tremor from the wrist in patients with deep brain stimulation Sensors, 2025

18 people with bilateral STN deep brain stimulation, 108 tremor episodes. Wrist acceleration matched the clinician's rest-tremor rating at r = 0.92 to 0.97, and the 3-7 Hz band was the strongest single feature. Turning stimulation off produced a three-point change on the clinical scale that the wrist could see.

What Kampa did with it: took the 3-7 Hz band as the window its tremor checks measure in, since it spans both rest and held-position tremor.

What it does not show: 18 people at one center, one kind of stimulation, and a research-grade wrist sensor rather than a consumer watch or a phone. It tells us which frequencies carry tremor, not that our numbers inherit its accuracy.

Standardized arm positions for measuring postural tremor with a smartphone Frontiers in Neuroscience, 2022

A phone held in the hand with the arm extended explained 45% of the variance in clinician tremor ratings in Parkinson's, and 59% in essential tremor, separating affected from unaffected at an area under the curve of 0.89 to 0.93.

What Kampa did with it: adopted the arm position, and the rule that nothing may support the arm while you hold it - a resting wrist damps tremor, so a supported arm produces a reading that cannot be compared with yesterday's.

The digitized Archimedes spiral drawing test: a scoping review Scoping review of 11 studies, 2025

Across 11 studies in Parkinson's and essential tremor, spiral measures correlated with the motor scale at r = 0.50 to 0.64, and agreed with human raters at a weighted kappa of 0.65 to 0.95.

What Kampa did with it: reversed an earlier decision. We had put spiral drawing on the no-build list citing a study where a generic "draw a shape" task barely tracked severity at all. That verdict belonged to the measure, not the task: the same hand and the same thirty seconds scored properly reaches 0.50 to 0.64. The lesson - never accept or reject a test by its name, ask which feature is extracted - now applies to everything we build.

What it does not show: the headline figure belongs to a composite score whose formula was never published. Kampa does not compute it and does not claim it.

Spiral drawing with an instrumented pen in Parkinson's disease Frontiers in Neurology, 2023

Spirals drawn by people with Parkinson's carried measurably more movement in the 4-7 Hz tremor band than those drawn by controls.

What Kampa did with it: this is the source of the one number the spiral check reports - the share of your tracing movement that sits in the tremor band. It is a described quantity, not a severity score.

Wearing-off and OFF time

Minimal clinically important difference in Parkinson's disease as assessed in pivotal trials of pramipexole extended release Parkinson's Disease, 2014

Across the pramipexole immediate- and extended-release pivotal trials, a clinically meaningful reduction in daily OFF time falls between 1.0 and 1.3 hours.

What Kampa did with it: took the stricter end - 60 minutes a day - as the floor for calling a change in your OFF time real rather than noise, and works in daily OFF minutes, the unit the threshold is defined on. It is the difference between an app that reports every wobble as a finding and one that only speaks when the change would matter clinically.

Walking and gait change

Defining the clinically meaningful difference in gait speed in persons with Parkinson disease Journal of Neurologic Physical Therapy, 2014

Establishes how much gait speed has to change in Parkinson's before the change means anything to the person walking.

What Kampa did with it: it supplies the floor Kampa uses to decide whether a shift in your walking is worth showing you - and, just as importantly, to say confidently that nothing has changed when nothing has.

Minimal clinically important differences of spatiotemporal gait variables in Parkinson disease Gait & Posture, 2024

Extends the same question past speed to stride length, cadence and the rest of the gait picture.

Gait progression over 6 years in Parkinson's disease: effects of age, medication, and pathology Frontiers in Aging Neuroscience, 2020

How gait actually changes over years rather than weeks, which is the timescale a tracking app has to be able to distinguish from a bad afternoon.

Progressive gait deficits in Parkinson's disease: a wearable-based biannual 5-year prospective study Frontiers in Aging Neuroscience, 2019

Five years of wearable-measured gait, and the evidence that a wearable can see progression at all.

What this page does not claim

Everything above is other people's research. This is the part about what it does and does not say about Kampa.

Citation is not endorsement

Linking a study means Kampa used its findings to make a design decision. It does not mean its authors, their institutions, or their employers have reviewed, tested, endorsed, or are affiliated with Kampa in any way.

Kampa is not a medical device

Kampa has no regulatory clearance and does not seek to replace clinical assessment. It is not intended for the diagnosis, treatment, cure, or prevention of any disease, including Parkinson's. It reports measurements against your own history; it never issues a score, a grade, a clinical verdict, or a dosing instruction. See the Terms of Use.

Reliability is not validity

A measurement that repeats consistently is not automatically a measurement of the thing you care about. Several tests in this literature are highly reliable and only moderately valid, and we treat the two claims separately rather than quoting whichever number is higher.

These studies were not run on Kampa

They validate protocols and thresholds that Kampa implements. They are not evidence about Kampa's own accuracy, and no result here should be read as a measurement of this app.