What Are Digitally Derived Measures in Clinical Trials? DDMs, Digital Endpoints and DHTs Explained

Digitally derived measures, or DDMs, are measures derived from data collected using digital health technologies. They can be used in clinical research to assess aspects of health or function and, depending on their intended use, may serve as clinical outcome assessments, biomarkers or components of multicomponent endpoints.

Digital health technologies can include hardware and software using computing platforms, connectivity and sensors. Examples can include smartphones, wearables and other connected technologies used to capture data within or outside traditional clinical settings.

The important distinction is that collecting more digital data does not automatically create a useful clinical measure.

The technology, the measure being generated and the way that measure is used within a clinical investigation all need to be considered separately.

That distinction has become particularly relevant following new activity from the US Food and Drug Administration in August 2026.

Why are digitally derived measures receiving more attention now?

On 20 August 2026, the FDA published Key Considerations for the Development and Use of Digitally Derived Measures for Clinical Investigations.

The paper brings together considerations from existing FDA guidance to support the development and use of DDMs as outcomes in clinical investigations.

FDA highlights the potential for digital health technologies, including technologies enabled by artificial intelligence, to capture information continuously, in real time and outside conventional healthcare settings. It also stresses that measures should be clinically relevant and meaningful to patients.

A week later, on 27 August 2026, the FDA held a public workshop focused specifically on digital health technologies and statistical considerations for digitally derived endpoints in clinical trials.

The workshop addressed areas including data standards, analytical approaches and the broader use of clinically meaningful digitally derived endpoints in drug and biological product trials.

The message for sponsors, CROs and clinical technology providers is increasingly clear:

The question is no longer simply whether clinical trials can collect more data digitally. It is whether that data can support measurements that are meaningful, reliable and appropriate for the intended study.

What is a digital health technology?

The FDA describes a digital health technology, or DHT, as a system using computing platforms, connectivity, software and/or sensors for healthcare and related uses.

Within clinical investigations, a DHT may include hardware, software or a combination of both.

That means the category can cover a wide range of technologies.

Depending on the study, these might include:

• smartphones
• wearable sensors
• connected monitoring devices
• software applications
• sensor-enabled devices
• technologies capable of collecting information remotely

The DHT is the technology used to collect or generate the underlying data.

The DDM is the measure derived from that data.

Understanding that distinction is important.

What is the difference between a DHT and a DDM?

A simple way to think about it is:

DHT = the technology

Data = what the technology captures

DDM = the measure derived from that data

For example, a sensor-enabled technology might capture repeated movement data.

Those raw signals are data.

A defined measure calculated from those signals could potentially become a digitally derived measure if it has been appropriately developed and supported for its intended use.

FDA gives examples of areas in which DDMs could potentially provide insight, including mobility, sleep, cardiovascular function and cognition.

But simply measuring movement, heart rate or another signal does not by itself establish that the resulting measure is clinically meaningful.

That requires evidence.

Is a digitally derived measure the same as a digital endpoint?

No.

A digitally derived measure is not automatically a clinical trial endpoint.

A DDM describes a measure derived from data collected using digital technology.

An endpoint describes how a clinical outcome or other measure is defined and evaluated within the study.

FDA notes that DDMs may be used as clinical outcome assessments, biomarkers or as components of multicomponent endpoints derived from multimodal data.

For an endpoint based on digitally derived data, the study still needs to define what is being assessed, when it is assessed, which technology is used and how measurements are interpreted or combined.

FDA's guidance materials emphasise that sponsors should be able to justify that the desired clinical outcome can be appropriately measured by the selected technology.

So:

Not every digital measure is an endpoint, and not every piece of digital data is a measure.

Are eCOA and ePRO the same as digitally derived measures?

Not exactly.

eCOA, or electronic clinical outcome assessment, refers to clinical outcome assessment conducted electronically.

ePRO, or electronic patient-reported outcome, is a form of electronic data collection in which participants report information about their own health, symptoms or experience.

DDM refers to a measure derived from data collected using a digital health technology.

There can therefore be overlap, but the terms should not be used interchangeably.

For example, a participant may complete an electronic questionnaire on a smartphone as part of an ePRO workflow.

That is different from a smartphone or connected sensor continuously capturing data from which a measure of activity or another characteristic is subsequently derived.

The distinction becomes increasingly important as clinical research expands from digital data entry towards digital measurement.

What can digitally derived measures assess?

The potential scope is broad, but the intended use must always be defined and supported.

FDA has discussed potential digitally derived information relating to areas such as:

• mobility
• sleep
• cardiovascular function
• cognition
• activity
• physical performance
• other aspects of health or daily functioning

One attraction of DHT-based measurement is the ability to collect information outside a traditional clinic visit.

A conventional assessment might provide a snapshot of a participant at a particular point in time.

Appropriately designed digital measurement may enable researchers to understand certain characteristics across more frequent observations or during normal daily activity.

That creates significant research opportunities.

It also increases the importance of measurement science, validation, usability and the technology environment through which the data is collected.

What does "fit for purpose" mean for digitally derived measures?

This is one of the central issues in the FDA's August 2026 paper.

FDA states that, in clinical investigations, a DHT used to generate a DDM should be verified and validated to be considered fit for purpose.

The agency also highlights the need to identify potential sources of error and factors that could negatively affect the validity of the measure.

Fit for purpose therefore does not mean that a particular wearable, sensor, smartphone or software platform is universally suitable for clinical research.

Suitability depends on the intended use.

A technology may be appropriate for one measurement in one participant population but inappropriate for another.

Sponsors need to consider questions such as:

• What exactly are we trying to measure?
• Is that measure clinically relevant?
• Is it meaningful to patients?
• Can the technology measure the relevant characteristic accurately and precisely?
• Can the intended participant population use the technology correctly?
• What factors could introduce error?
• How will the resulting measure be interpreted?
• How does the measure relate to the study endpoint?

The technology should follow the measurement objective, not the other way around.

Why does patient relevance matter?

A technically sophisticated measure is not automatically a meaningful measure.

The FDA's August paper specifically highlights involvement from patients, caregivers and clinicians when determining the meaningfulness and clinical relevance of a DDM.

This matters because digital technologies can potentially generate enormous quantities of data.

But more data does not necessarily mean better evidence.

A useful digitally derived measure needs to address something relevant to the clinical question being investigated.

That creates an important principle for digital clinical research:

Start with what matters to the study and the participant. Then determine whether digital technology provides an appropriate way to measure it.

Why does usability matter for digitally derived measures?

Because measurement quality can depend on how the technology is actually used.

FDA notes that evidence for validation should demonstrate that users understand and can follow the instructions for use.

The agency's broader DHT guidance also discusses practical factors such as physical design, ease of use and battery life.

These may seem like device-level considerations, but they can have research consequences.

If a participant cannot use the technology correctly, repeatedly forgets to charge it or finds the workflow confusing, the issue is no longer simply about user experience.

It can affect data availability and the overall research workflow.

For sponsors, that means usability, accessibility and operational practicality belong in the same conversation as sensor performance and analytical validity.

Can smartphones be used to support digitally derived measures?

Potentially, yes.

Smartphones contain computing capability, connectivity and, depending on the device and workflow, multiple sensors.

They can also act as an interface for other connected technologies.

But the presence of sensors does not automatically make a smartphone suitable for generating a particular clinical measure.

The intended workflow, hardware capabilities, software environment, measurement characteristics and validation requirements still need to be considered.

Where smartphones are used as part of a larger digital measurement system, questions may include:

• hardware consistency
• operating-system configuration
• connectivity
• application control
• battery performance
• sensor characteristics
• device lifecycle
• participant usability
• replacement strategy
• international deployment

The more important the device becomes to the measurement workflow, the more important the underlying device infrastructure becomes.

What role could AI play in digitally derived measures?

AI may become increasingly relevant, particularly when large volumes of sensor, voice, image or other multimodal data need to be processed.

The FDA's August paper explicitly recognises digital health technologies enabled by artificial intelligence within the wider DHT landscape. It also refers to software and AI verification and validation as part of the broader evidence considerations around DDM development.

Potential approaches might involve identifying patterns within complex signals, processing information closer to the point of collection or combining multiple forms of digital information.

But AI does not remove the fundamental requirements.

The resulting measure still needs to be relevant, interpretable and appropriately supported for its intended use.

AI can change how data is processed. It does not remove the need to demonstrate why the resulting measure matters.

Could on-device processing become important in clinical research?

Potentially.

As patient-facing technologies become more capable, some processing may be performed directly on the device rather than requiring every piece of raw information to be processed remotely.

This is often described as edge or on-device processing.

Depending on the study and technical architecture, this could be relevant to areas such as latency, connectivity requirements, data handling and participant-facing workflows.

However, the suitability of any on-device processing approach depends on the specific research application, validation requirements and applicable regulatory and information-security considerations.

For clinical research teams, the important discussion should therefore be less about whether AI is 'on device' or 'in the cloud' and more about:

• What is being processed, why is it being processed, and how does that support the intended research measure?

Why does device infrastructure matter for digital measurement?

Digital measurement does not happen in isolation.

Behind a DDM may sit:

• a physical device
• sensors
• software
• connectivity
• configuration
• data transmission
• participant instructions
• device management
• technical support
• replacement processes

Each component can influence the consistency of the data-capture environment.

This becomes particularly important in global, decentralised and hybrid studies where participants may be collecting data away from research sites for extended periods.

A clinical trial might have a well-designed digital endpoint, but it still needs technology capable of supporting the required workflow throughout the programme.

That is where device infrastructure becomes part of the digital measurement conversation.

Where does STK Life fit?

STK Life focuses on the patient-facing hardware and device infrastructure layer supporting modern clinical research.

STK has designed and manufactured mobile devices since 1993.

STK Helix is a purpose-built 5G AI-native smartphone designed to support clinical research workflows including decentralised and hybrid trials, eCOA, ePRO, remote participation and provisioned-device programmes.

The distinction is important:

Helix is the hardware foundation. Cortex is the AI capability layer.

STK Cortex™ is being developed as an evolving AI capability layer for STK Helix.

The Cortex roadmap includes planned capabilities around areas such as on-device AI, passive signal capture, voice workflows and standardised image capture for clinical research. These are roadmap capabilities and should not be interpreted as currently available functions unless separately confirmed.

STK Helix and STK Cortex are designed to support clinical research infrastructure and data-capture workflows. They are not positioned as diagnosing, treating or preventing disease, or as replacing clinical judgement.

What should sponsors consider before using digitally derived measures?

There is no universal checklist for every DDM, but study teams can start by asking:

• What meaningful aspect of health are we trying to measure?
• Why is a digital approach appropriate?
• Which digital health technology will generate the underlying data?
• Has the technology been appropriately verified and validated for the intended use?
• Can the intended participant population use it correctly?
• What sources of measurement error need to be considered?
• How will the raw data become a defined measure?
• How will that measure contribute to a clinical outcome, biomarker or endpoint?
• What happens when connectivity is unavailable?
• How will device, software and operating-system changes be managed during the study?
• How will lost, damaged or unavailable devices be replaced?
• How will the technology environment be supported for the duration of the trial?

The FDA's recent work reinforces an important principle:

Digital innovation needs to be connected to measurement relevance, evidence and operational reality.

Frequently asked questions about digitally derived measures

What does DDM mean in clinical trials?

DDM stands for digitally derived measure. The FDA uses the term for measures derived from data collected using digital health technologies.

What is an example of a digitally derived measure?

A DDM might quantify an aspect of mobility, sleep, cardiovascular function, cognition or another relevant characteristic using data collected by a digital health technology. The exact measure and its validity depend on its intended context of use.

Is a wearable device a digitally derived measure?

No.

A wearable is the technology.

The data it captures may be used to derive a measure.

The device, the data and the resulting measure are separate concepts.

Is a smartphone a digital health technology?

A smartphone can form part of a digital health technology system when its computing, software, connectivity or sensing capabilities are used for healthcare or related purposes, including defined clinical research workflows.

Are digitally derived measures the same as digital biomarkers?

No.

A digitally derived measure describes how the measure is generated.

Depending on its intended use, a DDM may be used as a biomarker, but DDM and digital biomarker are not interchangeable terms.

Are all digital endpoints DDMs?

A digitally derived endpoint may use one or more measures generated from DHT data, but the endpoint itself also defines how the measure is used and analysed within the study.

The terms should therefore not be treated as synonyms.

Is ePRO a digitally derived measure?

ePRO and DDM overlap but describe different concepts.

ePRO refers to participant-reported information collected electronically. DDM refers to a measure derived from data captured using digital health technology.

Does FDA require DHTs to be fit for purpose?

FDA's August 2026 paper states that a DHT used to generate a DDM in a clinical investigation should be verified and validated to be considered fit for purpose.

Can AI generate digitally derived measures?

AI-enabled digital health technologies may contribute to processing or deriving measures from digital data, but the resulting measure still needs appropriate evidence supporting its relevance, validity and intended use.

Why are DDMs important for decentralised clinical trials?

DDMs may enable some information to be collected remotely and potentially more frequently than would be practical during conventional site visits.

This can create new opportunities for decentralised and hybrid research, but it also increases the importance of usability, connectivity, technology validation and reliable patient-facing infrastructure.

The takeaway

Digitally derived measures represent an important evolution in clinical-trial technology.

The shift is not simply from paper to electronic data collection.

It is from asking:

"Can we collect this digitally?"

to asking:

"Can digital technology help us measure something meaningful, reliably and appropriately for this study?"

That is a much more important question.

As DHTs, sensors, AI and patient-facing technologies become more capable, clinical research teams will have access to increasingly rich forms of data.

The challenge will be deciding which data should become measures, which measures are meaningful and which technology environments can support them consistently.

For sponsors, CROs and clinical technology providers, digital measurement therefore requires more than innovative software.

It requires appropriate measurement science, validated technology, participant usability and reliable infrastructure working together.

Purpose-built mobile device infrastructure for global clinical trials.

Sources

• U.S. Food and Drug Administration, Key Considerations for the Development and Use of Digitally Derived Measures for Clinical Investigations, August 2026.
• U.S. Food and Drug Administration, FDA Virtual Workshop on Digital Health Technologies and Statistical Considerations for Digitally-Derived Endpoints in Clinical Trials, 27 August 2026.
• U.S. Food and Drug Administration, Digital Health Technologies for Remote Data Acquisition in Clinical Investigations, Final Guidance, December 2023.