Panel · Template

Consensus Delphi.

LiveConsensus Delphi

Test candidate statements against a rule stated before any result: each specialist rates every statement independently, and you get agreement as seat counts, what reached consensus, what did not, and the residual disagreement left visible.

You getA consensus report of about four pages (Standard)

DepthsLight (three seats), Standard (four) or Full (five, two rounds)

Written byAI specialists who rate each statement alone

SourcesEvery claim cited or marked as a gap

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

When to use it

You have candidate statements, such as outcomes, criteria or principles, and need to see which ones a panel of informed specialists would back, under a rule fixed in advance.

Who it is for. Leads who build shared standards: medical, evidence, access, regulatory, safety, quality and compliance leads preparing outcome sets, evidence standards, principles or readiness criteria.

Why consensus needs a rule set in advance

A review of 100 Delphi studies found that nearly all claimed to measure consensus, but only 72 defined it, and only 64 of those defined it in advance. A consensus that is discovered at the end can be made to fit. This template states the rule at the top of the report, has each seat rate every statement alone, reports agreement as counts, and labels the whole exercise as a simulation by AI specialists, because AI debate is known to drift toward agreement.^{4Diamond, I. R. et al. Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies. Journal of Clinical Epidemiology, 2014;67(4):401–409.,9Yao et al. Peacemaker or Troublemaker: How Sycophancy Shapes Multi-Agent Debate. arXiv:2509.23055, 27 September 2025.,10Kasprova, V. et al. Too Polite to Disagree: Understanding Sycophancy Propagation in Multi-Agent Systems. arXiv:2604.02668, 2026.}

What you get

The statements and the rule. Numbered candidate statements and the consensus rule stated before any result: the threshold, what happens if it is not reached and how many seats rated.^{3ACCORD (ACcurate COnsensus Reporting Document): a reporting guideline for consensus methods in biomedicine. PLOS Medicine, 2024. doi:10.1371/journal.pmed.1004326.,4Diamond, I. R. et al. Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies. Journal of Clinical Epidemiology, 2014;67(4):401–409.}

Verdict and status. How many statements reached consensus for, consensus against, no consensus, or are unstable, and how far you may rely on the result.^{3ACCORD (ACcurate COnsensus Reporting Document): a reporting guideline for consensus methods in biomedicine. PLOS Medicine, 2024. doi:10.1371/journal.pmed.1004326.}

The independent first pass. Each seat’s rating of every statement, made without reference to the other seats, with a reason. In the one head-to-head study found, Delphi groups were more reliable and nominal groups reached closer agreement, so independence is a trade-off and the report says which it chose.^{1Delphi method. Wikipedia.,2Hutchings, A., Raine, R., Sanderson, C. and Black, N. A comparison of formal consensus methods used for developing clinical guidelines. Journal of Health Services Research and Policy, 2006;11(4):218–224.,7Bolger, F. and Wright, G. Improving the Delphi process: lessons from social psychological research. Technological Forecasting and Social Change, 2011;78(9):1500–1513.}

Agreement by statement, and movement. The seat counts for each statement. At Full depth, a second round after feedback shows who moved and why, which is the nearest the platform comes to a second Delphi round.^{6Khodyakov, D. Generating evidence using the Delphi method. i2insights, 17 October 2023.}

Residual disagreement, what a real Delphi would add, and the limits. The cruxes where seats split, what a panel of real experts would add, and a consensus statement that says what it is not.^{5Jünger, S. et al. Guidance on Conducting and REporting DElphi Studies (CREDES) in palliative care. Palliative Medicine, 2017;31(8):684–706.,8Williamson, P. R. et al. The COMET Handbook: version 1.0. Trials, 2017. doi:10.1186/s13063-017-1978-4.}

What is in it, by length

Section	Brief	Standard	Comprehensive
The statements and the consensus ruleThe statements and the rule fixed in advance	●	●	●
Verdict and statusHow many statements reached consensus	●	●	●
Independent first pass, by seatEach seat’s rating of every statement	●	●	●
Agreement by statementThe seat counts for each statement	●	●	●
Movement after feedbackWho moved in round two, and why	–	–	●
Residual disagreement and cruxesWhere seats still split and why	–	●	●
What a real Delphi would addWhat real experts and more rounds would add	–	●	●
The consensus statement and its limitsThe statement, and what it is not	●	●	●

A Light run is a single-pass consensus check with three seats and about two pages. Standard has four seats and one round. Full has five seats and a second round after feedback.

Choosing a length

Choose Standard for most sets of statements. Choose Full when you want to see whether seats move after seeing each other’s reasons. A Light run is a check and not a Delphi, because with one round there is no feedback.

The standards and practice behind it

The structure and the checks come from published work on Delphi methods and consensus reporting. These are the references we drew on and what we took from each. None of their authors has reviewed the template, and the run is a simulation by AI specialists, not a Delphi with real experts.

What we took	From	Where it shows
Keep the four classic features: anonymity, iteration, controlled feedback and a group response	The Delphi method, as summarized on Wikipedia, which cites the RAND work of Dalkey and Helmer^{1Delphi method. Wikipedia.}	Independent ratings, then feedback at Full depth
Define consensus before the first round and say what happens if it is not reached	Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies (Diamond and colleagues, 2014)^{4Diamond, I. R. et al. Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies. Journal of Clinical Epidemiology, 2014;67(4):401–409.}	The rule stated at the top of the report
Report how agreement was reached, using a 35-item reporting guideline for consensus methods	ACCORD, the ACcurate COnsensus Reporting Document (PLOS Medicine, 2024)^{3ACCORD (ACcurate COnsensus Reporting Document): a reporting guideline for consensus methods in biomedicine. PLOS Medicine, 2024. doi:10.1371/journal.pmed.1004326.}	Threshold, rounds and respondents reported with the result
Standardize the reporting of Delphi studies, including what a round is	CREDES: Guidance on Conducting and REporting DElphi Studies (Jünger and colleagues, 2017)^{5Jünger, S. et al. Guidance on Conducting and REporting DElphi Studies (CREDES) in palliative care. Palliative Medicine, 2017;31(8):684–706.}	The label for each run as a single-pass check or a two-round run
Rate in two rounds with discussion between them, and keep the panel small enough for discussion to be meaningful	The RAND/UCLA Appropriateness Method, as described by Khodyakov^{6Khodyakov, D. Generating evidence using the Delphi method. i2insights, 17 October 2023.}	The Full depth with a second round
Feed back reasons and not only statistics, because conformity pressure survives anonymity	Improving the Delphi process: lessons from social psychological research (Bolger and Wright, 2011)^{7Bolger, F. and Wright, G. Improving the Delphi process: lessons from social psychological research. Technological Forecasting and Social Change, 2011;78(9):1500–1513.}	Reasons recorded with every rating
Treat Delphi as a route to core outcome sets and similar standards	The COMET Handbook: version 1.0 (Williamson and colleagues, 2017)^{8Williamson, P. R. et al. The COMET Handbook: version 1.0. Trials, 2017. doi:10.1186/s13063-017-1978-4.}	The outcome-set use case
Expect sycophancy and loss of stance diversity when AI agents deliberate	Peacemaker or Troublemaker (arXiv, 2025) and Too Polite to Disagree (arXiv, 2026)^{9Yao et al. Peacemaker or Troublemaker: How Sycophancy Shapes Multi-Agent Debate. arXiv:2509.23055, 27 September 2025.,10Kasprova, V. et al. Too Polite to Disagree: Understanding Sycophancy Propagation in Multi-Agent Systems. arXiv:2604.02668, 2026.}	Independent first-pass ratings and visible residual disagreement

What it will not do

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] Delphi method. Wikipedia. Read in the template research. Secondary source; the Dalkey and Helmer (1963) paper it cites was not retrieved.
[2] Hutchings, A., Raine, R., Sanderson, C. and Black, N. A comparison of formal consensus methods used for developing clinical guidelines. Journal of Health Services Research and Policy, 2006;11(4):218–224. Abstract read in the template research. One study in one clinical domain, comparing Delphi and nominal groups.
[3] ACCORD (ACcurate COnsensus Reporting Document): a reporting guideline for consensus methods in biomedicine. PLOS Medicine, 2024. doi:10.1371/journal.pmed.1004326. Read in the template research (first 100,000 characters).
[4] Diamond, I. R. et al. Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies. Journal of Clinical Epidemiology, 2014;67(4):401–409. Figures as given in a summary of the abstract; the full text was not retrieved.
[5] Jünger, S. et al. Guidance on Conducting and REporting DElphi Studies (CREDES) in palliative care. Palliative Medicine, 2017;31(8):684–706. Known from the EQUATOR listing; the checklist was not read.
[6] Khodyakov, D. Generating evidence using the Delphi method. i2insights, 17 October 2023. Read in the template research: the RAND/UCLA two-round design and the cap of about 18 panelists.
[7] Bolger, F. and Wright, G. Improving the Delphi process: lessons from social psychological research. Technological Forecasting and Social Change, 2011;78(9):1500–1513. Abstract read in the template research.
[8] Williamson, P. R. et al. The COMET Handbook: version 1.0. Trials, 2017. doi:10.1186/s13063-017-1978-4. Bibliographic record and summary only; its recommendations on thresholds and rounds were not read.
[9] Yao et al. Peacemaker or Troublemaker: How Sycophancy Shapes Multi-Agent Debate. arXiv:2509.23055, 27 September 2025. Abstract read on 8 October 2026.
[10] Kasprova, V. et al. Too Polite to Disagree: Understanding Sycophancy Propagation in Multi-Agent Systems. arXiv:2604.02668, 2026. Abstract read in the template research.

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.

Use cases for the Consensus Delphi

8 ready-to-run use cases. Each opens its own page with the question, the specialists, the standards and the references.

COLiveCore Outcome SetWhich outcomes should every trial in this rare-disease indication measure and report?Consensus DelphiClinical Operations

Open Open →

EDLiveEvidence-Standard DefinitionWhat should count as sufficient real-world evidence for a payer submission in our category?Consensus DelphiMarket Access

Open Open →

GSLiveGood-Practice StatementsWhich statements should govern how we answer unsolicited off-label questions from clinicians?Consensus DelphiMedical Affairs

Open Open →

ESLiveEndpoint SelectionWhich primary endpoint should the pivotal study of our digital therapeutic use?Consensus DelphiInnovation

Open Open →

DPLiveData-Sharing PrinciplesWhich principles should we commit to publicly on sharing trial and real-world data?Consensus DelphiPublic Affairs

Open Open →

RMLiveRisk-Minimization MeasuresWhich additional risk-minimization measures are proportionate for the identified risk?Consensus DelphiPharmacovigilance

Open Open →

LCLiveLaunch-Readiness CriteriaWhich criteria must all be met before we release the launch to market?Consensus DelphiBrand & Marketing

Open Open →

CPLiveCoalition Position StatementsRate candidate statements for a coalition position across public affairs, patient advocacy, compliance and access policy, under a consensus rule fixed first.Consensus DelphiPublic Affairs

Open Open →

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

When to use it

You have candidate statements, such as outcomes, criteria or principles, and need to see which ones a panel of informed specialists would back, under a rule fixed in advance.

Why consensus needs a rule set in advance

What you get

What is in it, by length

Section	Brief	Standard	Comprehensive
The statements and the consensus ruleThe statements and the rule fixed in advance	●	●	●
Verdict and statusHow many statements reached consensus	●	●	●
Independent first pass, by seatEach seat’s rating of every statement	●	●	●
Agreement by statementThe seat counts for each statement	●	●	●
Movement after feedbackWho moved in round two, and why	–	–	●
Residual disagreement and cruxesWhere seats still split and why	–	●	●
What a real Delphi would addWhat real experts and more rounds would add	–	●	●
The consensus statement and its limitsThe statement, and what it is not	●	●	●

A Light run is a single-pass consensus check with three seats and about two pages. Standard has four seats and one round. Full has five seats and a second round after feedback.

The standards and practice behind it

What we took	From	Where it shows
Keep the four classic features: anonymity, iteration, controlled feedback and a group response	The Delphi method, as summarized on Wikipedia, which cites the RAND work of Dalkey and Helmer^{1Delphi method. Wikipedia.}	Independent ratings, then feedback at Full depth
Define consensus before the first round and say what happens if it is not reached	Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies (Diamond and colleagues, 2014)^{4Diamond, I. R. et al. Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies. Journal of Clinical Epidemiology, 2014;67(4):401–409.}	The rule stated at the top of the report
Report how agreement was reached, using a 35-item reporting guideline for consensus methods	ACCORD, the ACcurate COnsensus Reporting Document (PLOS Medicine, 2024)^{3ACCORD (ACcurate COnsensus Reporting Document): a reporting guideline for consensus methods in biomedicine. PLOS Medicine, 2024. doi:10.1371/journal.pmed.1004326.}	Threshold, rounds and respondents reported with the result
Standardize the reporting of Delphi studies, including what a round is	CREDES: Guidance on Conducting and REporting DElphi Studies (Jünger and colleagues, 2017)^{5Jünger, S. et al. Guidance on Conducting and REporting DElphi Studies (CREDES) in palliative care. Palliative Medicine, 2017;31(8):684–706.}	The label for each run as a single-pass check or a two-round run
Rate in two rounds with discussion between them, and keep the panel small enough for discussion to be meaningful	The RAND/UCLA Appropriateness Method, as described by Khodyakov^{6Khodyakov, D. Generating evidence using the Delphi method. i2insights, 17 October 2023.}	The Full depth with a second round
Feed back reasons and not only statistics, because conformity pressure survives anonymity	Improving the Delphi process: lessons from social psychological research (Bolger and Wright, 2011)^{7Bolger, F. and Wright, G. Improving the Delphi process: lessons from social psychological research. Technological Forecasting and Social Change, 2011;78(9):1500–1513.}	Reasons recorded with every rating
Treat Delphi as a route to core outcome sets and similar standards	The COMET Handbook: version 1.0 (Williamson and colleagues, 2017)^{8Williamson, P. R. et al. The COMET Handbook: version 1.0. Trials, 2017. doi:10.1186/s13063-017-1978-4.}	The outcome-set use case
Expect sycophancy and loss of stance diversity when AI agents deliberate	Peacemaker or Troublemaker (arXiv, 2025) and Too Polite to Disagree (arXiv, 2026)^{9Yao et al. Peacemaker or Troublemaker: How Sycophancy Shapes Multi-Agent Debate. arXiv:2509.23055, 27 September 2025.,10Kasprova, V. et al. Too Polite to Disagree: Understanding Sycophancy Propagation in Multi-Agent Systems. arXiv:2604.02668, 2026.}	Independent first-pass ratings and visible residual disagreement

What it will not do

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.