Measuring What Matters in Alzheimer's and Parkinson's Clinical Trials
- Koneksa Health
- Jun 26
- 5 min read
Every June, Alzheimer's & Brain Awareness Month turns attention to the millions of people living with Alzheimer's and other neurodegenerative diseases, as well as to the families and caregivers who carry so much of the daily load. It is also a moment to be honest about the research meant to help them. Alzheimer's disease (AD) and related dementias were estimated to impose a global economic burden of roughly $2.8 trillion in 2019, projected to approach $17 trillion by 2050, and Alzheimer's now ranks among the top 10 causes of mortality. The scale is staggering, the urgency is real, and progress depends on a deceptively simple question: are we measuring what actually matters?
Why Neurodegenerative Diseases Are So Hard to Measure
Diseases like Alzheimer's and Parkinson's progress gradually and across many dimensions at once: cognition, movement, speech, sleep, and day-to-day function. Yet much of how clinical research tracks that progression still relies on episodic, clinic-based assessments: a clinician-rated scale or cognitive test administered every few weeks or months. These assessments are valuable, but they capture only a snapshot in time.
The problem is that disease-related change, medication effects, and treatment response often operate on different and overlapping timescales. When you sample sparsely, you are forced to assume a straight line between visits, and that assumption obscures the nonlinear, fluctuating reality of central nervous system (CNS) disease. Variability from rater judgment and visit timing cannot be averaged out, because the measurements are simply too infrequent. The consequence is high-stakes: promising therapies can be abandoned because a real effect was buried in noise, and ineffective ones can advance on the strength of a statistical mirage. Either way, patients wait longer.
For a deeper look at how digital and traditional measures compare, see our recent breakdown: Digital Biomarkers vs. Traditional Biomarkers: A Complete Comparison.
The Case for Digital Endpoints in CNS Clinical Trials
High-frequency, longitudinal digital measurement, aligned to disease biology, offers a way to
close that gap. By capturing objective, sensitive measures remotely and continuously from the comfort of a patient's home, researchers can separate true biological signal from noise, model disease trajectories rather than single visits, and improve sensitivity to treatment effects.
The efficiency gains can be dramatic. In one simulation cited in our CNS whitepaper, moving
from episodic measurement to continuously monitored data reduced the required per-arm
sample size from approximately 490 to 26 at the same effect size, because each participant
effectively becomes their own control over time. Smaller, faster, statistically powered trials lower per-program costs, accelerate go/no-go decisions, and free resources for broader pipelines without sacrificing regulatory rigor.
This is the work Koneksa is built around. Our digital measurement framework spans 30+
validated measures across motor, cognitive, speech, sleep, and functional domains; the
multidimensional picture neurodegenerative disease demands.
Go deeper:
Our whitepaper, Implementing Digital Endpoints in CNS Clinical Trials, lays out
the scientific and operational foundations, from biometric design and mixed-effects modeling to patient-centric measure development and regulatory considerations.
Parkinson's Disease: Capturing Motor Fluctuations in Daily Life
In Parkinson's disease (PD), the gold-standard MDS-UPDRS is administered during discrete
clinic visits, which cannot fully capture day-to-day or within-day variability in motor function,
including the On and Off states that define the patient experience. High-frequency, home-based digital assessment changes that. Continuous motor measurement enables within-subject trajectory modeling and helps separate short-term medication effects from longer-term disease progression.
Our Finger Tap Assessment is a good example: a fine-motor measure that pairs high-temporal-resolution capture with four complementary derived measures (tap count, speed, correctness, regularity), sustaining roughly 90% average compliance across repeated remote visits. It's one of the highest-yield measures in our library, with about 75% of its use focused in PD programs.
Explore how we approach motor measurement in Parkinson's:
Alzheimer's Disease: Detecting Subtle Decline Earlier
In Alzheimer's, episodic assessments like the CDR-SB can require substantial change before
progression is detectable, a real limitation when many disease-modifying therapies are intended to act early, in the transition from pre-symptomatic to clinically manifest disease. Continuous digital measures of speech and language, gait variability, and sleep patterns offer the potential to detect subtle cognitive and functional decline earlier than episodic clinical assessment alone.
It's worth underscoring a theme from our whitepaper: successful adoption of digital endpoints depends less on the availability of technology and more on how measurement strategies are constructed: aligning the construct of interest, the method of derivation, and the endpoint's intended role in a trial.
Biomarker Strategy Is Becoming Foundational, and Digital Measurement
Across the field, biomarker strategy is being reframed from a supporting tool into the foundation of neurodegenerative drug development shaping patient selection, enrichment, trial size, and even regulatory and commercial positioning. As a recent perspective from Worldwide Clinical Trials put it, the question is no longer whether biomarkers belong in development strategy, but how early and how comprehensively to integrate them.
We'd build on that in one important way. Much of the biomarker-strategy conversation centers on molecular and imaging markers such as genomics, proteomics, fluid assays, PET. Those are essential, but they describe biology, not daily function. A complete strategy also needs the functional layer: how patients actually move, speak, sleep, and think between visits. Molecular markers can tell you a therapy is engaging its target; high-frequency digital functional measures help tell you whether that engagement is changing the patient's life, and they extend the same efficiency logic (smaller, more homogeneous, faster trials) from molecular biomarkers to functional ones.
A Field Moving Forward Together
Regulators are moving with the science. The EMA's qualification of stride velocity 95th centile
as the first digitally derived primary endpoint, and the inclusion of digital health technologies in the finalized ICH E6(R3) Good Clinical Practice guideline, both signal growing acceptance of digital measures in clinical development.
Better measurement is necessary, but it isn't sufficient on its own. Advancing neurodegenerative research also takes deep therapeutic and operational expertise to design and run the trials, which is why we're glad to be part of a broader ecosystem of teams committed to the same goal, including P95 Julius Clinical, whose neuroscience team recently shared a thoughtful perspective on moving Alzheimer's and CNS research forward. As their team put it, every advancement begins with a shared commitment to improving lives. We'd add that it also depends on how precisely we can measure progress along the way.
Talk to Koneksa About Your Measurement Strategy
This Alzheimer's & Brain Awareness Month, the most meaningful thing we can do for patients is keep raising the bar on the science, including how we measure it. If you're thinking about
measurement strategy for an AD, PA, or broader CNS program, our team can help:
Frequently Asked Questions
What are digital endpoints in clinical trials?
Digital endpoints are clinical trial outcome measures derived from digital health technologies, such as wearables, sensors, and smartphone-based assessments, that capture objective, often continuous data on a patient's function and behavior outside the clinic. In CNS research they include measures of gait, tremor, speech, sleep, and cognition.
How do digital biomarkers improve Alzheimer's and Parkinson's research?
By measuring symptoms frequently and remotely, digital biomarkers reduce within-subject noise, capture day-to-day fluctuation that single clinic visits miss, and model disease as a trajectory rather than a snapshot, improving sensitivity to treatment effects and, in simulations, sharply reducing required sample sizes.
Why are episodic clinic visits insufficient for measuring CNS disease?
CNS diseases progress nonlinearly and fluctuate over hours and days. Infrequent visits force researchers to assume a straight line between measurements, conflating biological signal with noise from visit timing and rater judgment, which reduces statistical power.