Panel · Use case · Innovation

Endpoint Selection.

LiveConsensus Delphi

Which primary endpoint should the pivotal study of our digital therapeutic use?

SetCandidate primary endpoints for a pivotal study

SeatsDigital therapeutics, regulatory, HEOR, operations

ReaderDigital health and development leads

Not askedA study protocol

DepthStandard, four seats

Endpoint Selection is a ready-made use case for Innovation. It ships under the Consensus Delphi template.

For innovation leads. You get each seat's stance on every statement, where consensus holds, the residual disagreement and the consensus statement with its limits. Claims are cited or marked as gaps.

Add it to your library to reuse from New → From template. Edit the saved copy in your library on the service form.

The decision

Which primary endpoint should the pivotal study of our digital therapeutic use?

Who rates

Different systems ask for different kinds of benefit from software. The seats rate each candidate endpoint against regulators and payers.

Seat	What it rates from	Why it is at the table
Head of Digital Therapeutics	Whether each endpoint captures the benefit the software delivers.	Trial evidence.
VP Global Regulatory Strategy	Whether each endpoint will be accepted for a clearance or approval.^{1International Council for Harmonisation. ICH E9(R1): Addendum on estimands and sensitivity analysis in clinical trials.}	ICH E9(R1) and agency practice.
HEOR Evidence Lead	Whether each endpoint answers the payer question, including the German standard of patient-relevant benefit.^{2Federal Institute for Drugs and Medical Devices (BfArM). Digital Health Applications (DiGA): Interesting facts. DiGA Guide version 3.6.,3National Institute for Health and Care Excellence. Evidence standards framework (ESF) for digital health technologies.}	The BfArM and NICE frameworks.
Clinical Operations Director	Whether each endpoint can be measured within a feasible trial.	Trial practice.

What the run does

Regulators and payers may want different endpoints. The run reports where seats split between an endpoint a regulator accepts and one a payer values. That split is often the finding.^{2Federal Institute for Drugs and Medical Devices (BfArM). Digital Health Applications (DiGA): Interesting facts. DiGA Guide version 3.6.,3National Institute for Health and Care Excellence. Evidence standards framework (ESF) for digital health technologies.}

Estimand before endpoint. Each candidate is stated with the question it answers, using the estimand framework.^{1International Council for Harmonisation. ICH E9(R1): Addendum on estimands and sensitivity analysis in clinical trials.}

What a real Delphi would add. Clinicians, patients and regulators who can say which endpoint they would act on.

Recommended length

Standard

Four disciplines each rate every statement. Standard depth gives one round with seat counts and the residual disagreement in about four pages.

You can pick another length on the form. The other two stay available.

Assumptions the run makes

The run states the assumptions behind its statements and its rule, and says what each gap limits. Examples for this set:

The candidate endpoints cover the benefit the software delivers.
A single primary endpoint can serve both regulators and payers.
The trial can measure the endpoint without unblinding.

Standards, guidance and sources

Frame

ICH E9(R1), addendum on estimands and sensitivity analysis.^{1International Council for Harmonisation. ICH E9(R1): Addendum on estimands and sensitivity analysis in clinical trials.}
BfArM, DiGA directory and fast-track process.^{2Federal Institute for Drugs and Medical Devices (BfArM). Digital Health Applications (DiGA): Interesting facts. DiGA Guide version 3.6.}
NICE, Evidence Standards Framework for digital health technologies.^{3National Institute for Health and Care Excellence. Evidence standards framework (ESF) for digital health technologies.}

The decision practice behind the memo itself is on the template page.

What gets flagged

Any endpoint accepted by one seat only, any endpoint with no stated estimand, and any split between regulatory and payer seats.

What arrives

The statements and the consensus rule
Verdict and status
Independent first pass, by seat
Agreement by statement
Movement after feedback
Residual disagreement and cruxes
What a real Delphi would add
The consensus statement and its limits

What it will not do

The agreement is a simulation by AI specialists and not a Delphi with real experts. The consensus rule is fixed before any result, and agreement is not correctness.

It does not design the pivotal study.

It is a simulation by AI specialists and not a Delphi with real experts. A small panel of three to five seats is below every panel size found in the literature, and the report says so.

It does not treat agreement as truth. Consensus is not correctness.

It does not change the rule after seeing the result.

References

[1] International Council for Harmonisation. ICH E9(R1): Addendum on estimands and sensitivity analysis in clinical trials. Not read.
[2] Federal Institute for Drugs and Medical Devices (BfArM). Digital Health Applications (DiGA): Interesting facts. DiGA Guide version 3.6. Read on 9 October 2026.
[3] National Institute for Health and Care Excellence. Evidence standards framework (ESF) for digital health technologies. Not read in full.

Links checked 9 October 2026

These are the sources the template drew on. None of their authors or issuers has reviewed it.

About the Panel service

Each panelist answers in their own voice, then the discussion is drawn together. You get the question and who was in the room, every position in full, what they agreed on, where they split, and what follows. Disagreements are written down rather than smoothed into a consensus nobody actually holds, because the split is usually the part you needed.

To edit a saved template, open New → Panel → From template, or the library rail on the service form.