2.2 Context of use
With patient-informed endpoints giving your strategy a strong, empathetic core, the next step is to define exactly how those endpoints will operate within your trial—their specific role, scope, and relevance.
This is where the [Context of Use (COU)] comes in, serving as the blueprint that ensures your digital measures are precise, appropriate, and ready for regulatory scrutiny.
Think of COU as the “who, what, when, and why” that ensures your digital measure is fit-for-purpose.
OVERVIEW
Understanding context of use
Every credible digital endpoint strategy begins with a clearly defined COU – a concise statement outlining how your endpoint will be applied in the trial: the population it serves, the clinical setting, the specific aspect of health it measures, and how it contributes to decision-making.
Anchoring the COU to both the Concept of Interest and the Meaningful Aspect of Health (see Section 2.1: Patient-informed endpoints) ensures the endpoint reflects patient priorities and regulatory expectations. A well-scoped COU also positions the team to estimate a minimal clinically important difference (MCID), the threshold of change that is both detectable and meaningful in clinical terms.
Adopters, this aligns the endpoint with your trial’s objectives and regulatory pathway, ensuring it integrates smoothly into protocols and analysis plans.
Developers, detail how your technology supports that use case, from data accuracy in real-world conditions to compatibility with the intended population.
Getting the COU right early prevents downstream issues, like mismatched data or rejected submissions, and builds confidence that your measure advances meaningful clinical insights.
A note on terminology – Context of use versus indications for use
Intended use / Indications for use:
- Describes what the medical device will be cleared/approved to do
- Example: “Intended to measure heart rate in adults”
- Relevant when seeking device marketing authorization
Context of use (COU):
- Describes how an endpoint or measure will be used in a specific trial
- Example: “Daily step count in COPD patients as secondary endpoint in Phase 3 trial”
- Relevant when using sDHTs to generate clinical trial endpoints
Both CDER and CDRH use “context of use” when discussing endpoints. CDRH additionally uses “intended use” or “indications for use” when discussing device marketing authorization.
In this roadmap: When we discuss establishing fitness-for-purpose for endpoints captured by sDHTs, we’re referring to context of use. Device marketing authorization (intended use) is a separate regulatory question.
in practice
For digital endpoints in clinical trials, the context of use typically includes:
What is being measured (e.g., walking speed, sleep efficiency, arrhythmias)
The population being studied (e.g., adults with early-stage Parkinson’s disease)
The purpose of the measure (e.g., as a primary efficacy endpoint, exploratory biomarker)
How and where the data is collected (e.g., continuous passive monitoring using a wrist-worn accelerometer in a home setting)
Duration and frequency of measurement
Example: A COU might specify using a specific wearable to track daily step count in adults with multiple sclerosis to assess mobility improvements over six months in a Phase II study.
Stakeholder considerations
Adopter considerations
For adopters, defining COU is the foundation of trial credibility. Adopters should:
- Begin with the clinical question. Clarify the patient experience or disease process the endpoint is meant to capture. This keeps the measure tied to an unmet need, not the capabilities of any single technology. The work around patient-prioritized aspects of health and concepts of interest outlined in Section 2.1: Patient-informed endpoints are a critical orientation. If you already have a tool and want to understand which clinical questions it may be most appropriate for, this work will be essential.
- Classify the endpoint appropriately. Distinguish whether the measure is a clinical outcome assessment (COA) or a biomarker (discussed below), as this shapes evidentiary standards and validation expectations (see Section 4: Your validation strategy for more on regulatory implications).
- Engage patients and clinicians. Input from patient communities, advocacy groups, and clinical experts grounds the endpoint in outcomes that are both relevant and interpretable. This input should inform MCID estimation, ensuring the trial is designed to detect a change that is meaningful.
- Look to precedent. Reviewing existing validation evidence and published endpoints helps avoid duplication and strengthens the rationale for advancing a novel measure.
Developer contributions
Developers play an equally important role in bringing clarity and feasibility to this stage:
- Translate concepts into measures. Map the pathway from the intended endpoint to the measures their sDHT captures, showing how technical measures align with clinical intent.
- Clarify technical assumptions. Share what is known about operating conditions, and data pipelines, developers help adopters judge whether the measure is feasible in the proposed COU.
- Support MCID estimation. Provide evidence that anchors what degree of change is technically detectable, complementing clinician and patient insights about what degree of change is meaningful.
Due Diligence
Before finalizing your COU, pause for a landscape review.
Before initiating new evidence generation, it is essential to first survey the existing landscape. Early review of public data sources reveals precedent and maturity, and can identify gaps that justify novel digital endpoints. This ensures you are not duplicating efforts and helps assess whether there is an unmet measurement need.
This step helps you avoid duplicating existing efforts, strengthen endpoint rationale, and support regulatory conversations with evidence of unmet need or validation precedent.
Sources to consider as you review the existing landscape:
- Peer-reviewed scientific literature
- FDA databases (e.g., 510(k), PreMarket Approval), summary letters, and Voice of the Patient/PFDD meetings
- Core digital measures developed in precompetitive working groups like DATAcc by DiMe (see the Industry spotlight below)
FDA SPOTLIGHT
“FDA encourages collaboration among multiple stakeholders and the use of methods to combine and leverage existing data (e.g., national registry data, archival databases, published literature) to fit the specific needs of the research questions and study goals.”
– Section II.F.2 (Leveraging Existing Data), p. 19
what’s at stake?
Biomarkers vs. Clinical Outcome Assessments
Biomarker
clinical outcome
assessment
Biomarkers are objective indicators of biological processes, disease states, or responses to treatment. In the context of sDHTs, these may include physiologic or behavioral signals captured by sensors that are intended to reflect underlying biology rather than direct patient experience.
Clinical outcome assessments (COAs) measure how a patient feels, functions, or survives and are intended to directly reflect treatment benefit. When derived from sDHTs, COAs translate digital signals into measures that represent patient-relevant aspects of health within a defined context of use.
Evidentiary expectations
A biomarker requires analytical validation to demonstrate that the digital signal and processing pipeline accurately and reliably measure the biological process of interest, along with clinical validation appropriate to its intended role in decision-making. Additional justification is needed if the biomarker is proposed as a surrogate for a patient-meaningful outcome, consistent with FDA’s biomarker-related guidance and terminology.
Evidentiary expectations
A COA also requires clinical validation to demonstrate that the endpoint meaningfully reflects change in the concept of interest within the defined context of use. Patient input is essential to establish relevance, and when COAs are derived from sDHTs, analytical validation of signal acquisition and processing is necessary to ensure the measure is reliable and interpretable.
Diving deeper
Next steps
If your endpoint is classified as a biomarker:
Anchor your approach in the FDA–NIH BEST glossary’s biomarker definitions and categories.
For biomarkers, evidentiary expectations focus on two pillars: analytical validation and clinical validation (showing that the biomarker meaningfully reflects disease biology, progression, or treatment response). Some teams may choose to pursue FDA’s Biomarker Qualification Program (see below), but this is not the default path for most digital biomarkers. Even outside the formal program, these evidentiary domains remain essential to ensure regulators view the biomarker as fit-for-purpose in a clinical trial context.
If your endpoint is classified as a Clinical Outcome Assessment (COA):
Look to FDA’s Patient-Focused Drug Development Guidances 3 and 4 (see below), which describe expectations for selecting, developing, and incorporating COAs into regulatory decision-making.
For COAs, evidence must demonstrate a clear line of sight from patient input to the outcome assessed. This includes documenting how meaningful aspects of health (MAH) informed the endpoint, defining a minimal clinically important difference (MCID) that reflects meaningful change (discussed below!), and ensuring traceability throughout the process. These inputs should then feed into your V3+ plans, especially clinical validation and usability validation, to show that the measure is both scientifically rigorous and meaningful to patients.
Clinical meaningfulness
According to FDA’s PFDD Guidance 4, Draft, 2023 (FDA) Section II.A.1, sponsors should include “support for the importance of the endpoint to patients and/or caregivers from literature review and/or primary data collection,” and, per Section III.B.1, may consider using anchor-based methods to justify meaningful score differences and interpret the treatment benefit.
exampleS
MCID, sDHTS, and Alzheimer’s disease: Regulatory and implementation perspectives
FDA neurosciences reviewers outline practical ways sponsors can demonstrate that trial endpoints truly reflect benefits that matter to patients and caregivers for individuals with Alzheimer’s disease. They emphasize:
- Tying within-patient meaningful change thresholds to patient experience data.
- Selecting stage-appropriate COAs and combining cognition + function to show benefit.
- Leveraging innovative COAs and biomarker or surrogate endpoints when they are “reasonably likely” to predict benefit.
DiMe Digital Measures projects
Defining a Context of Use and identifying a Meaningful Aspect of Health is a rigorous process that requires months of stakeholder alignment. If your research involves Sleep, Physical Activity, Nocturnal Scratch, or Alzheimer’s and Related Dementias (ADRD), much of this foundational work has already been done for you.
The DiMe Digital Core Measures projects provide pre-competitive, expert-vetted blueprints that allow you to skip the “blank page” stage. These expert-led core digital measures projects (worked examples for MAH to COI to digital endpoint) include ontologies, conceptual models, best practices, and other resources.
KEY TAKEAWAYS
If there is lingering ambiguity about the COU or how an endpoint should be classified, scheduling a short early touchpoint with FDA can provide practical guidance on plausibility and evidentiary priorities. This step can be especially valuable before committing resources to study design or contracting (see Section 3: Engage regulators). Your endpoint’s COU will shape your validation needs across all four domains of V3+: verification, analytical, clinical, and usability (see Section 4: Your validation strategy).
Define a clear context of use for your particular use case.
Prepare an MCID justification memo with clinician input.
Draft a working conceptual model, which is a visual map that links patient-meaningful aspects of health to concepts of interest and clearly defined digital endpoints.
Draft the endpoint rationale for any candidate digital endpoints for inclusion in briefing documents, statistical analysis plan, etc.
2.2 Context of use
Library resources to guide you
The sDHT roadmap library gathers 200+ external resources to support the adoption of sensor-based digital health technologies. To help you apply the concepts in this section, we’ve curated specific spotlights that provide direct access to critical guidance and real-world examples, helping you move from strategy to implementation.
Features essential guidance, publications, and communications from regulatory bodies relevant to this section. Use these resources to inform your regulatory strategy and ensure compliance.
Gathers real-world examples, case studies, best practices, and lessons learned from peers and leaders in the field relevant to this section. Use these insights to accelerate your work and avoid common pitfalls.