ICR WHITE PAPER 024
THE ICR RESEARCH AND VALIDATION STANDARD
Rules for Moving From Conceptual Claims to Reproducible Evidence
David FischerInstitute for Coherence and Regulation (ICR)Knightdale, North Carolina, USASeptember 2026 | Version 1.0
Recommended citationFischer, D. (2026). The ICR Research and Validation Standard: Rules for Moving From Conceptual Claims to Reproducible Evidence. ICR White Paper 024 (Version 1.0). Institute for Coherence and Regulation.
DOI: 10.5281/zenodo.22712999
Abstract
The Institute for Coherence and Regulation (ICR) requires a formal standard governing how conceptual propositions, wellness observations, measurement instruments, mechanisms, and intervention claims may advance in evidentiary status. WP-024 establishes that standard. Its governing principle is: ICR will claim no more than the strongest reproducible evidence directly supports. The standard separates exploratory from confirmatory research, requires prespecified questions and outcomes when testing hypotheses, distinguishes reporting guidance from study-quality assessment, defines appropriate use of observational and randomized designs, requires transparent protocol changes, adverse-event reporting, conflict-of-interest disclosure, null-result reporting, data provenance, and independent replication, and creates explicit rules for evidence upgrading and downgrading. It incorporates established external standards rather than replacing them. CONSORT 2025 and SPIRIT 2025 provide current guidance for randomized-trial reports and protocols; STROBE guides reporting of major observational designs; CARE informs case reports; PRISMA supports systematic reviews; and COSMIN informs measurement-property research. Preregistration is treated as a time-stamped record of planned research rather than proof of study quality. The standard also creates an ICR Claim Registry and a modality firewall preventing evidence for the CRF from being used automatically as evidence for any specific wellness modality. The purpose is institutional restraint: favorable findings may strengthen a claim only through defined evidence steps, while contradictory, null, or methodologically stronger evidence can weaken it.
Keywords: research integrity; validation; preregistration; replication; reporting standards; evidence hierarchy; conflicts of interest; null results; Coherence & Regulation Framework
1. Governing Principle
ICR will claim no more than the strongest reproducible evidence directly supports.
This principle applies to books, courses, practitioner materials, websites, presentations, research reports, marketing, certification materials, and public statements.
The Institute's credibility depends as much on what it refuses to claim as on what it investigates.
2. Purpose of the Standard
The CRF is a conceptual framework containing testable constructs. The purpose of research is not to prove the framework correct; it is to determine which parts are useful, which require revision, and which should be rejected.
WP-024 establishes the rules by which evidence may change the status of an ICR claim.
The standard applies both to CRF research and to evaluations of ICR programs or modalities.
3. External Standards Take Priority
ICR will use established reporting and methodological standards whenever they fit the study design.
The Institute will not rename standard research practices merely to make them proprietary.
Examples include SPIRIT for randomized-trial protocols, CONSORT for randomized-trial reports, STROBE for observational research, CARE for case reports, PRISMA for systematic reviews, and COSMIN for measurement-property research.
A reporting checklist improves transparency; it does not by itself establish that a study was well designed or unbiased.
4. Evidence Levels
Level
Designation
What it can support
0
Conceptual
Definition, rationale, model, proposed mechanism, hypotheses
1
Feasibility / Descriptive
Can the protocol be delivered and measured? What was observed?
2
Associational
Variables covary under defined conditions
3
Comparative
Groups, conditions, or trajectories differ with an appropriate comparator
4
Mechanistic / Measurement Validation
A proposed mechanism or instrument receives direct validity evidence
5
Independent Replication
Key findings reproduced by investigators not dependent on ICR
6
Synthesis / Convergence
Multiple high-quality independent studies support a bounded conclusion
5. Evidence Levels Are Not Automatic
A study does not reach a higher level merely because it uses more participants, more sensors, advanced statistics, or a randomized design.
The evidence level depends on the specific claim, the validity of the measures, the design's ability to address alternatives, risk of bias, precision, consistency, and replication.
One study can support different claims at different levels.
6. Claim Types
Claim type
Example form
Minimum evidence direction
Conceptual
CRF proposes that recovery dynamics may reflect adaptive capacity
Level 0
Descriptive
Participants reported lower perceived stress after a program
Level 1
Associational
Recovery time was associated with subsequent performance
Level 2
Comparative
The intervention group improved more than comparator
Level 3
Mechanistic
Change occurred through a specified measured pathway
Level 4+
Measurement
Instrument validly measures defined construct
Level 4+ with measurement-property evidence
General effectiveness
Intervention reliably improves outcome in target population
Replication and convergence generally required
7. Exploratory Research
Exploratory research is legitimate and necessary. It searches for patterns, estimates feasibility, develops hypotheses, and identifies candidate measures.
Exploratory analyses should be labeled exploratory even when results are statistically significant.
Hypotheses generated after seeing the data should not be rewritten as if they were specified beforehand.
8. Confirmatory Research
Confirmatory research tests prespecified hypotheses using a protocol and analysis plan defined before the relevant outcomes are examined.
Primary outcome, primary comparison, analysis population, exclusions, and key analytic decisions should be specified in advance where feasible.
Departures from the plan may be justified but must be disclosed.
9. Preregistration
Preregistration creates a time-stamped record of research intentions before data collection or analysis. It improves transparency about what was planned.
Preregistration is not a badge of truth. A preregistered study can still have weak measures, poor execution, bias, inadequate power, or inappropriate analysis.
ICR should preregister confirmatory studies in an appropriate registry or repository and preserve amendments with dates and reasons.
10. Registered Reports
Where feasible, ICR should consider Registered Reports for important confirmatory studies.
In this publication model, the research question and methods can undergo peer review before results are known, reducing incentives for publication decisions based on outcome direction.
Availability depends on the target journal and study type.
11. Protocol Before Outcome
A formal protocol should exist before a confirmatory study begins.
Research question and rationale.
Study design.
Population and eligibility.
Intervention/exposure and comparator.
Primary and secondary outcomes.
Measurement schedule.
Safety monitoring.
Sample-size rationale.
Statistical analysis plan.
Missing-data plan.
Stopping or modification rules when applicable.
Data-management and privacy plan.
Conflict-of-interest statement.
12. Protocol Amendments
Research changes. A protocol can be amended when new information, feasibility problems, safety concerns, or errors require it.
Changes should be dated, justified, and distinguished from the original plan.
ICR must never silently rewrite a protocol after seeing results.
13. Study Design Must Match the Claim
Descriptive questions can be answered descriptively. Associations require appropriate observational analysis. Causal intervention claims require designs capable of addressing counterfactual alternatives.
Randomization is powerful for many intervention questions but does not correct invalid measurement, poor adherence, selective reporting, or inappropriate analysis.
Mechanism claims require measurement of the proposed mechanism.
14. Observational Research
Observational designs are appropriate for natural exposures, longitudinal trajectories, feasibility, prognostic research, associations, and questions where randomization is impractical or unethical.
STROBE provides reporting recommendations for cohort, case-control, and cross-sectional studies.
ICR should report confounding, selection, missing data, measurement limitations, and alternative explanations explicitly.
15. Randomized Trials
Randomized trials should follow current SPIRIT guidance for protocols and CONSORT guidance for reporting.
CONSORT 2025 is the current major update for reporting randomized trials, accompanied by SPIRIT 2025 for protocols.
Random allocation supports causal inference by reducing systematic baseline differences, but implementation quality and analysis remain critical.
16. Comparator Selection
A comparator should answer the scientific question rather than merely make the intervention look favorable.
Possible comparators include usual activity, wait-list, attention control, credible sham where appropriate, alternative intervention, dose comparison, or within-person counterbalanced condition.
The comparator's limitations must be stated.
17. Randomization
Randomization methods should be reproducible, allocation sequence generation should be documented, and allocation concealment used where relevant.
Predictable assignment is not adequate randomization.
Small studies should not assume randomization guarantees balance on every important variable.
18. Blinding
Blinding can reduce bias when participants, practitioners, outcome assessors, analysts, or adjudicators can reasonably be blinded.
Some wellness interventions cannot blind practitioners or participants convincingly.
When blinding is impossible, ICR should use objective or independently assessed outcomes where feasible and report the limitation.
19. Expectancy and Attention
Wellness research is especially vulnerable to expectancy, attention, ritual, interpersonal contact, and demand-characteristic effects.
These are not reasons to dismiss participant experience; they are alternative causal explanations that should be measured or controlled when the research question requires it.
Claims about a specific modality require stronger controls than claims about the experience of receiving a program.
20. Sample Size
Sample size should be justified by the primary objective.
Feasibility studies can use precision or process targets rather than conventional power calculations. Confirmatory studies should use an appropriate power or precision rationale based on the primary analysis.
Post hoc power calculations should not be used to rescue inconclusive results.
21. Primary Outcomes
Confirmatory studies should identify a limited number of primary outcomes before analysis.
A study measuring many outcomes increases the probability of apparently favorable findings by chance.
Secondary and exploratory outcomes remain valuable when labeled correctly.
22. Statistical Significance
A p-value is not the magnitude, importance, probability of the hypothesis, or evidence of mechanism.
ICR should emphasize effect estimates, uncertainty intervals, clinical or practical relevance where applicable, and the full pattern of results.
Binary significant/non-significant language should not substitute for interpretation.
23. Effect Size and Precision
Reports should present effect sizes and confidence or credible intervals appropriate to the analysis.
Wide intervals indicate uncertainty even when the point estimate appears large.
Small studies should be described with proportionate caution.
24. Multiple Comparisons
When many outcomes, time points, subgroups, models, or biomarkers are examined, false-positive risk rises.
Primary hypotheses should be prespecified and multiplicity addressed where appropriate.
Exploratory findings can be reported without pretending they were confirmatory.
25. Subgroup Analysis
Subgroup findings should generally be treated cautiously unless prespecified, biologically or theoretically justified, sufficiently powered, and supported by interaction testing.
Finding significance in one subgroup and not another does not itself prove the subgroups differ.
Post hoc subgroup discoveries require replication.
26. Missing Data
Missingness can bias results when it relates to outcome, burden, symptoms, adherence, or treatment response.
ICR should report how much data are missing, why when known, and how missingness was handled.
Sensitivity analyses should be used when assumptions materially affect conclusions.
27. Attrition
Participant withdrawal and loss to follow-up should be documented by group and reason when available.
A program cannot be judged solely from participants who completed it successfully.
High or differential attrition can materially weaken conclusions.
28. Adherence and Fidelity
ICR should distinguish whether an intervention was assigned, received, completed, and delivered according to protocol.
Fidelity data help determine whether a null result reflects an ineffective intervention, inadequate delivery, poor adherence, or another problem.
Per-protocol analyses should not silently replace intention-to-treat conclusions in randomized trials.
29. Safety and Adverse Events
Wellness does not mean risk-free. Studies should prospectively define relevant adverse events, serious adverse events where applicable, expected reactions, contraindications, and referral procedures.
All groups should be monitored when the design makes this relevant.
Absence of reported adverse events is not evidence of safety if adverse events were not actively collected.
30. Human-Subject Oversight
Research involving human participants should receive the ethical review, informed consent, privacy protection, and institutional oversight required by applicable law, regulation, sponsor, institution, and publication venue.
ICR should not label a project 'program evaluation' merely to avoid research oversight when the actual activity is designed to produce generalizable knowledge.
When uncertain, qualified IRB or research-compliance guidance should be obtained.
31. Vulnerable Participants
Additional protections may be needed for children, cognitively impaired participants, economically vulnerable participants, patients with serious illness, or others whose circumstances can affect voluntariness or risk.
Recruitment language should not overstate benefit or imply that participation is medical treatment when it is not.
Compensation should reimburse time and burden without becoming undue influence.
32. Data Integrity
Raw data should be preserved where ethically and legally appropriate, and transformations should be reproducible.
Fabrication, falsification, selective deletion, undocumented recoding, or changing outcomes to favor a conclusion are incompatible with ICR research.
Corrections should be documented rather than concealed.
33. Data Provenance
Each reported value should be traceable through source data, instrument version, processing steps, exclusions, and analysis.
WP-023 established this requirement for CRF measurement.
Data dictionaries and processing logs should accompany serious studies.
34. Reproducible Analysis
Analysis code or a sufficiently detailed analysis record should be preserved.
Where privacy, licensing, or proprietary constraints prevent public release, the limitation should be stated and an appropriate controlled-access path considered.
Reproducibility should not depend on undocumented manual steps.
35. Open Materials and Data
ICR should share protocols, de-identified data, code, and materials when ethically, legally, and contractually possible.
Participant privacy, consent restrictions, copyrighted instruments, and legitimate confidentiality obligations can limit openness.
Open science is a transparency practice, not permission to violate privacy.
36. Null Results
A well-conducted study that does not support the hypothesis is scientifically informative.
ICR will not suppress null results merely because they conflict with Institute expectations or commercial interests.
Null findings should trigger interpretation, replication, revision, or retirement according to evidentiary strength.
37. Negative and Contradictory Evidence
Evidence that contradicts an ICR proposition receives the same methodological scrutiny as favorable evidence.
Stronger contradictory evidence can downgrade a claim.
The framework must remain revisable.
38. Publication Bias
Publishing only favorable findings creates a distorted evidence base.
Preregistration, prospective study registration, results reporting, and public archiving of completed studies can reduce selective visibility.
ICR should maintain an internal/public study register identifying completed, ongoing, discontinued, and unpublished projects as appropriate.
39. Selective Outcome Reporting
Primary outcomes should not disappear from a report because they were unfavorable.
New outcomes discovered during analysis can be reported as exploratory.
Comparing the final report with the protocol or preregistration is a basic transparency check.
40. Case Reports and Case Series
Case reports can document unusual observations and generate hypotheses but cannot establish general efficacy.
CARE guidance should inform transparent reporting when a case report is appropriate.
Testimonials are not substitutes for case reports because they usually lack standardized measurement, chronology, alternative explanations, and systematic follow-up.
41. Systematic Reviews
When a sufficient literature exists, systematic reviews should use transparent eligibility criteria, reproducible searches, risk-of-bias assessment, and appropriate synthesis.
PRISMA provides reporting guidance for systematic reviews.
A narrative list of favorable studies is not a systematic review.
42. Measurement Validation
A new ICR instrument must be validated for its intended use rather than declared valid by internal preference.
WP-023 and COSMIN principles govern content validity, reliability, measurement error, structural validity where relevant, construct validity, responsiveness, and interpretability.
A measure can be useful for exploration before validation if clearly labeled.
43. Mechanism Claims
A mechanism claim requires direct measurement of the proposed pathway and a design capable of distinguishing it from alternatives.
Improvement after an intervention does not prove why the improvement occurred.
ICR should not use terms such as nervous-system regulation, endocrine balancing, cellular repair, immune modulation, energy-field change, or synchronization as mechanisms unless the relevant variables are directly and appropriately measured.
44. The Modality Firewall
Evidence supporting the CRF does not automatically support Reiki, PEMF, frequency-based wellness, red-light exposure, Structured Rest, or any other modality.
Likewise, evidence that a modality changes an outcome does not validate the entire CRF.
Each modality, mechanism, program, and framework proposition requires its own evidence chain.
45. Program Evaluation Versus Modality Efficacy
A multimodal program can be evaluated as a package. If participants improve, the result applies to the evaluated package under the study conditions.
The study cannot identify which component caused the change unless the design isolates components.
ICR should distinguish program-level effectiveness from modality-specific efficacy.
46. Conflicts of Interest
ICR studies should disclose intellectual, professional, and commercial interests relevant to the work.
David Fischer's authorship of the CRF and ownership or stewardship interests connected to ICR create a legitimate conflict that should be disclosed, not hidden.
Conflict disclosure does not invalidate research; it allows readers to evaluate incentives and safeguards.
47. Independence
As the evidence program matures, important confirmatory studies should include investigators, statisticians, outcome assessors, or institutions with meaningful independence from ICR.
Independent replication is especially important before broad healthcare-facing claims.
Replication by the originating organization alone remains useful but is not independent confirmation.
48. Replication
Replication should test the same core claim under sufficiently similar conditions while allowing transparent examination of generalizability.
Failure to reproduce a result should not be dismissed automatically as improper replication.
Differences in population, protocol, dose, measurement, and context should be examined prospectively.
49. Evidence Upgrading
A claim may be strengthened when better evidence becomes available.
Current status
Possible upgrade trigger
Conceptual
Reliable descriptive observation
Descriptive
Replicated association under defined conditions
Associational
Appropriate comparative evidence
Comparative
Direct mechanistic/measurement validation
Originator replication
Independent replication
Replicated finding
Convergent synthesis across studies/settings
Upgrading should apply to the specific claim tested, not to the entire framework.
50. Evidence Downgrading
A claim should be weakened when stronger evidence contradicts it, measurements prove unreliable, effects fail to replicate, bias is identified, estimates shrink substantially in better-controlled studies, or established explanations account for the result.
Downgrading is a normal part of scientific development.
Public-facing materials should be updated when claim status materially changes.
51. The ICR Claim Registry
ICR should maintain a controlled Claim Registry for important framework and modality statements.
Field
Purpose
Claim ID
Stable reference
Exact wording
Prevents claim drift
Construct/domain
Identifies what is being asserted
Evidence level
Current status
Supporting studies
Traceable evidence
Contradictory studies
Prevents selective citation
Permitted public wording
Controls overstatement
Prohibited wording
Defines boundary
Last review date
Ensures maintenance
Decision owner/reviewer
Accountability
52. Claim Language by Evidence Level
Evidence status
Preferred language
Conceptual
proposes; hypothesizes; may; candidate mechanism
Descriptive
participants reported; was observed; changed during
Associational
was associated with; correlated with; predicted
Comparative
differed from; greater change than comparator
Mechanistic
mediated by / caused through only when directly supported
Replicated
has been reproduced in independent studies
Convergent
evidence supports within the defined population/context
53. Prohibited Evidence Inflation
Do not convert association into causation.
Do not convert participant report into physiological mechanism.
Do not convert feasibility into effectiveness.
Do not convert one case into general efficacy.
Do not convert statistical significance into practical importance.
Do not convert a multimodal program result into proof of each component.
Do not convert CRF plausibility into modality validation.
Do not convert absence of observed harm into proven safety.
Do not convert originator replication into independent replication.
Do not convert publication into proof.
54. Publication and Peer Review
Peer review can improve a manuscript but does not guarantee truth, validity, or replication.
Repository white papers should be labeled as white papers or preprints unless they have undergone formal journal peer review.
ICR should preserve version history and update citations when a preprint later becomes peer-reviewed.
55. DOI and Version Control
A DOI provides persistent identification; it does not validate content.
Each permanent ICR white paper should carry a version number, publication date, recommended citation, and change history.
Material revisions should create a new version while preserving the historical record.
56. Corrections and Retractions
Errors should be corrected transparently. Major errors affecting conclusions may require withdrawal or retraction of a version.
The correction record should state what changed and why.
Institutional credibility is strengthened, not weakened, by visible correction of genuine mistakes.
57. Research Governance
ICR should separate research governance from marketing decisions.
Commercial pressure must not determine which outcomes are reported, how adverse events are classified, or whether contradictory evidence is published.
A research-review function should have authority to require more cautious wording.
58. Proposed Research Review Committee
As ICR research expands, a small Research Review Committee should review protocols, claim upgrades, major publications, conflicts, and safety plans.
At least one reviewer for consequential healthcare-facing claims should be meaningfully independent of the Institute.
The committee should document decisions and dissent.
59. Research Records
Protocols, consent forms, approvals, datasets, code, analysis outputs, deviations, adverse-event logs, conflict disclosures, manuscripts, peer-review correspondence, and claim decisions should be retained according to applicable requirements and institutional policy.
Record retention should be defined before formal research begins.
Credentialing record-retention rules should not automatically be assumed to govern research records.
60. Application to the Current Coherence Reset Evaluation
The current small Coherence Reset evaluation should be treated as exploratory program evaluation or a case series depending on its final protocol, oversight, and reporting purpose.
Its Institute-created 0-10 outcomes can describe participant trajectories but cannot validate the CRF, prove a mechanism, or establish general efficacy.
The evaluation's highest value is feasibility: recruitment, adherence, data completeness, acceptability, measurement behavior, participant experience, and generation of hypotheses for a stronger prospective study.
61. Path From Current Evaluation to a Stronger Study
Finalize a protocol and research question.
Select validated participant-reported outcomes where possible.
Add a defined functional outcome.
Standardize measurement timing.
Define adverse-event collection.
Prespecify primary outcome and analysis.
Use an appropriate comparator.
Seek required human-subject oversight.
Preregister before confirmatory data collection.
Report all prespecified outcomes.
Replicate before broad effectiveness claims.
62. Healthcare-Facing Evidence Threshold
Claims intended for hospitals, the VA, colleges, licensed clinicians, or healthcare research partners should meet a higher institutional threshold than general educational claims.
Conceptual plausibility can justify discussion and research collaboration, but not clinical efficacy claims.
Healthcare-facing materials should distinguish framework theory, preliminary ICR observations, established external science, and independently replicated evidence.
63. Ten Institutional Falsification Commitments
ICR will publish or archive important null findings rather than suppress them.
ICR will revise constructs that repeatedly fail reliability or validity testing.
ICR will downgrade claims after strong contradictory evidence.
ICR will not treat participant satisfaction as proof of mechanism.
ICR will not treat one physiological signal as whole-person coherence.
ICR will not claim modality efficacy from CRF theory.
ICR will not call an Institute-created measure validated without validation studies.
ICR will distinguish exploratory from confirmatory analyses.
ICR will seek independent replication before broad general claims.
ICR will retire a construct when established alternatives explain the same phenomenon without meaningful incremental value.
64. Minimum Standard for an ICR Research Report
Clear research question.
Design identified correctly.
Protocol/preregistration reference when applicable.
Eligibility and recruitment.
Intervention/exposure and comparator.
Measurement methods and timing.
Primary and secondary outcomes.
Sample-size rationale.
Statistical methods.
Missing data and attrition.
Protocol deviations.
Adverse events.
All prespecified results, including null findings.
Effect estimates and uncertainty.
Limitations and alternative explanations.
Conflicts of interest.
Ethics/oversight statement.
Data/material availability.
Evidence-level conclusion.
Permitted claim wording.
65. Institutional Decision Rule
Before strengthening any public claim, ICR should ask five questions:
Was the exact claim actually tested?
Was the measurement valid for that claim?
Could the design distinguish the claim from major alternatives?
Has the result been reproduced?
Would an independent reviewer use the same wording?
If the answer to any critical question is no, the wording should remain at the lower evidence level.
66. Relationship to WP-010
An earlier ICR paper established the initial evidence ladder and validation principles. WP-024 formalizes those principles as an Institute-wide governance standard and aligns them with the measurement architecture developed in WP-023.
Where wording differs, the more conservative requirement should govern until the series is harmonized in the final editorial pass.
The capstone paper will integrate the research standard without duplicating its operational detail.
67. Falsification and Revision of This Standard
The standard itself is revisable. Research-methods guidance changes, reporting standards are updated, laws and regulations change, and new evidence can identify better practices.
ICR should review the standard at least annually during active research development and whenever major external guidance changes.
Revision history should be public for permanent repository versions.
Harmonized Governance Role
WP-024 is the Institute-wide research governance standard. It does not replace WP-010, which governs translation from conceptual claim to testable study, or WP-023, which governs measurement architecture. WP-024 determines the rules under which ICR evidence may be designed, analyzed, reported, promoted, corrected, or used to support stronger claims.
Canonical Governing Rule
ICR will claim no more than the strongest reproducible evidence directly supports. Scientific language, public communication, educational material, program descriptions, and institutional recommendations should remain at or below the evidence level actually achieved.
Relationship to WP-010
WP-010 answers: How does a CRF concept become a falsifiable study? It governs the chain from conceptual claim through construct, operational definition, hypothesis, variable, study design, analysis, result, and claim language. WP-024 answers: What governance rules determine whether that evidence is sufficiently credible to be promoted, generalized, or used institutionally?
Relationship to WP-023
WP-023 answers: What exactly is being measured, by what instrument, with what measurement-property evidence and interpretation boundary? WP-024 requires those measurement decisions to be documented and audited before empirical findings are promoted.
Evidence Ladder 0-6
ICR uses a seven-level evidence ladder: Level 0 Conceptual; Level 1 Feasibility/Descriptive; Level 2 Associational; Level 3 Comparative; Level 4 Mechanistic or Measurement Validation; Level 5 Independent Replication; Level 6 Synthesis/Convergence. Levels describe the strongest support achieved for a specific claim, not the prestige of a paper, intervention, device, or researcher.
Evidence Levels Are Not Automatic
Completing a randomized study does not automatically create Level 3 or higher evidence, and publishing a mechanism paper does not automatically establish Level 4. Study quality, measurement validity, bias, precision, protocol adherence, alternatives, replication, and claim match determine the appropriate evidence level.
Claim-Specific Evidence
Evidence attaches to a defined claim. One study can support different claims at different levels. A program may have feasibility evidence, an outcome association, and no established mechanism simultaneously.
Exploratory Versus Confirmatory
Exploratory work may generate hypotheses, candidate measures, subgroups, mechanisms, and analysis approaches. Confirmatory work requires prespecified primary hypotheses, outcomes, analysis rules, exclusions, and interpretation criteria. Post hoc discoveries should not be rewritten as if they were prespecified.
Protocol Before Outcome
For confirmatory human research, the protocol and statistical analysis plan should be finalized before investigators have access to outcome patterns that could influence analytic choices. Deviations should be documented rather than silently incorporated.
Preregistration
Where an appropriate registry exists, confirmatory studies should be preregistered with sufficient detail to distinguish planned analyses from later exploration. Registration is a transparency tool, not proof of study quality.
Registered Reports
ICR should consider Registered Reports when a suitable journal offers them, particularly for high-stakes confirmatory tests. Peer review before results are known can reduce publication and analytic bias.
Reporting Standards
Study reports should use the current reporting guideline appropriate to the design. As of 2026, randomized trial protocols should use SPIRIT 2025 and randomized trial reports should use CONSORT 2025. Observational studies should use STROBE where applicable, systematic reviews PRISMA, case reports CARE, and other EQUATOR-listed standards appropriate to the design.
CONSORT and SPIRIT Are Reporting Standards
CONSORT and SPIRIT improve completeness and transparency of trial reporting and protocols; they are not substitutes for good design, ethics review, measurement validity, or appropriate analysis. ICR should avoid claiming that checklist compliance alone establishes evidentiary strength.
Protocol Amendments
Material protocol changes should be dated, justified, versioned, and disclosed. Amendments made after outcome information becomes available should be especially transparent because they can alter the risk of selective reporting.
Study Design Must Match the Claim
Descriptive claims require descriptive evidence; association claims require appropriate observational evidence; causal intervention claims generally require controlled designs capable of addressing confounding; mechanism claims require direct mechanistic measurements and causal logic; diagnostic claims require separate diagnostic-validation standards.
Randomization
When randomized allocation is used, the sequence-generation method, allocation concealment, implementation, deviations, and analysis population should be documented. Randomization reduces specific biases only when implemented correctly.
Blinding
Participant, practitioner, assessor, analyst, or outcome blinding should be used where feasible and relevant. When blinding is impossible, expectancy, attention, assessment, and analytic biases should be addressed through design and transparent limitation statements.
Comparator Selection
The comparator should test the actual claim. Wait-list, usual-care, attention-control, active-control, sham, dose-comparison, or no-intervention comparators answer different questions. A weak comparator should not be used to support a stronger causal or mechanism claim.
Expectancy and Attention
For interventions involving practitioner contact, ritual, touch, relaxation, environmental change, or participant expectations, expectancy and attention can influence outcomes. They should be measured or controlled when they are plausible alternative explanations.
Sample Size and Precision
Sample-size planning should be based on the primary research objective, anticipated effect or precision target, design, attrition, and analysis. Small exploratory studies can be useful but should not be presented as definitive because statistical significance was reached.
Primary Outcomes
Confirmatory studies should identify a limited set of primary outcomes in advance. Outcome switching or emphasizing favorable secondary outcomes after results are known is prohibited unless transparently labeled exploratory.
Effect Size and Uncertainty
ICR reports should emphasize effect estimates and uncertainty intervals in addition to p-values. Statistical significance does not establish clinical importance, practical importance, mechanism, or replication.
Multiplicity
Multiple outcomes, time points, subgroups, models, and comparisons increase false-positive risk. Confirmatory work should prespecify multiplicity handling or clearly limit claims.
Subgroup Analysis
Subgroup findings should generally be considered exploratory unless prespecified, adequately powered, based on interaction tests, and independently replicated. Separate significance in one subgroup and nonsignificance in another does not establish a subgroup difference.
Missing Data and Attrition
Reasons for missing data and dropout should be retained and reported. Analytic methods should reflect plausible missing-data mechanisms. Differential attrition can materially bias intervention comparisons.
Adherence and Fidelity
Intervention exposure, practitioner fidelity, protocol deviations, participant adherence, and co-interventions should be measured when they affect interpretation. A negative study with poor delivery and a positive study with uncontrolled co-interventions require different conclusions.
Safety and Adverse Events
Human studies should prospectively define relevant safety monitoring, adverse-event collection, stopping procedures, escalation pathways, and reporting responsibilities according to study risk. Wellness framing does not remove research safety obligations.
Human-Subjects Oversight
Research involving human participants should follow applicable institutional, legal, ethical, privacy, and informed-consent requirements. When IRB or equivalent review is required, ICR should obtain it before research begins. Program evaluation should not be relabeled to avoid oversight when the activity meets research criteria.
Vulnerable Participants
Studies involving children, cognitively impaired persons, economically or medically vulnerable populations, dependent relationships, or other heightened vulnerability require additional safeguards and appropriate oversight.
Data Integrity
Raw data, audit trails, data dictionaries, transformations, exclusions, derived variables, code, and analysis versions should be retained according to the protocol and applicable privacy requirements. Data should not be altered to improve a result.
Data Provenance
Every reported result should be traceable from source measurement through preprocessing, quality control, derivation, analysis, and figure or table. WP-023 governs the measurement-level provenance requirements.
Reproducible Analysis
Analysis code and computational environments should be preserved where feasible. A qualified independent analyst should be able to reproduce the reported numerical results from the governed dataset and documented workflow.
Open Materials and Data
ICR should share protocols, analysis plans, code, instruments it owns, and de-identified data where ethically, legally, contractually, and practically possible. Restrictions should be disclosed rather than implying openness that does not exist.
Null and Negative Results
Null, negative, contradictory, and adverse findings are part of the evidence base. ICR should not suppress them because they weaken a preferred framework, modality, program, or commercial offering.
Publication Bias
Selective publication of favorable results inflates apparent evidence. The Claim Registry and study registry should preserve the existence and status of initiated studies, including those that are discontinued or unpublished.
Case Reports and Case Series
Case reports and series can document feasibility, unexpected observations, safety signals, and hypothesis-generating patterns. They cannot establish efficacy, comparative effectiveness, prevalence, or a universal mechanism. CARE should be used where applicable.
Program Evaluation
Program evaluation can assess implementation, participant experience, feasibility, and preliminary outcomes. Unless designed as controlled research, it should not be presented as proof that the program caused the observed change.
Systematic Reviews
Systematic reviews should use transparent eligibility criteria, reproducible searches, risk-of-bias assessment, appropriate synthesis, and PRISMA reporting. A review cannot repair poor underlying studies.
Measurement Validation
Measurement-validation claims are governed jointly by WP-023 and established measurement-science standards. Reliability alone is not validity; correlation alone is not validation; and a device validated for one purpose does not validate a CRF interpretation.
Mechanism Claims
A mechanism claim should identify the proposed causal pathway, directly measure relevant intermediates, establish appropriate temporal ordering, address alternative mechanisms, and use perturbation or mediation evidence where feasible. Terminology such as energy, coherence, regulation, quantum, cellular, vagal, inflammatory, or nervous-system does not establish mechanism by itself.
The Modality Firewall
Evidence for CRF does not establish the efficacy of Reiki, PEMF, frequency-based approaches, scalar approaches, red-light exposure, structured rest, or another modality. Evidence for a modality does not validate the entire CRF. Each claim requires its own evidence chain.
Commercial Firewall
Commercial availability, client demand, testimonials, practitioner belief, course enrollment, product sales, or business success do not increase scientific evidence level. Research conclusions should be insulated from commercial incentives as far as practical.
Conflicts of Interest
Financial, intellectual, professional, and organizational conflicts should be disclosed. Institute ownership of a framework, program, credential, device relationship, course, book, or service relevant to a study is a conflict that should be transparently managed.
Research Independence
Where feasible, confirmatory testing should involve investigators, statisticians, assessors, or sites with meaningful independence from ICR and from commercial interests tied to the tested claim.
Independent Replication
Independent replication requires more than repeating a study internally. The replicating team should have meaningful independence in investigators, analysis, setting, or data and should test the same core claim using a sufficiently comparable protocol.
Evidence Upgrading
A claim may move upward only when stronger evidence directly addresses the claim and survives appropriate quality review. Quantity of weak studies does not automatically equal one strong study.
Evidence Downgrading
Claims should be downgraded when serious bias, imprecision, inconsistency, measurement failure, undisclosed deviations, failed replication, retraction, or credible contradictory evidence materially weakens confidence.
ICR Claim Registry
ICR should maintain a Claim Registry containing the canonical claim, construct, current evidence level, supporting studies, contradictory studies, replication status, permitted language, prohibited language, date reviewed, and responsible reviewer. This prevents marketing or educational language from drifting ahead of evidence.
ICR Study Registry
In addition to external registration where appropriate, ICR should maintain an internal Study Registry listing proposed, active, completed, discontinued, published, and unpublished studies. The registry should link protocols, versions, registrations, outcomes, adverse events, and publications.
ICR Measurement Registry
WP-023 establishes the Measurement Registry. WP-024 requires researchers to use the current approved measurement status and interpretation boundaries from that registry when planning and reporting studies.
Claim Language by Evidence Level
Level 0 language should use proposes, conceptualizes, or hypothesizes. Level 1 may use feasible, observed, or descriptively associated within the evaluated setting. Level 2 may use associated with or correlated with. Level 3 may use comparative language matched to the design. Level 4 may support validated measurement or mechanism language only for the specific validated construct or pathway. Levels 5-6 may support stronger generalization when replication and convergence justify it.
Prohibited Evidence Inflation
ICR publications and public materials should not convert 'associated with' into 'causes,' 'participants improved' into 'the program works,' 'device value changed' into 'the nervous system reset,' 'case series' into 'clinical proof,' or 'plausible mechanism' into 'proven mechanism.'
Publication and Peer Review
Peer review can improve work but does not certify truth. Publication status should be reported accurately, and preprints, conference abstracts, peer-reviewed articles, corrections, and retractions should be distinguishable.
DOI and Version Control
DOIs should be assigned only after the document has passed the intended publication audit. Material revisions should receive clear version identifiers and, where repository practice permits, version-linked records. A DOI makes a document persistent and citable; it does not validate its claims.
Corrections
Material factual, analytical, or interpretive errors should be corrected promptly and transparently. The correction should identify what changed, why, and whether conclusions were affected.
Retractions and Withdrawal
ICR should withdraw or retract Institute publications when findings are unreliable because of major error, fabrication, falsification, unethical research, invalid data, or another condition that makes continued reliance inappropriate.
Research Review Committee
ICR should establish a Research Review Committee as research volume grows. At minimum, the committee should review evidence-level assignments, protocol-governance compliance, conflicts, measurement status, claim language, major corrections, and requests to upgrade institutional claims.
Committee Independence
Where the Institute founder or another investigator has a direct intellectual or commercial interest in a claim, at least one reviewer without that direct interest should participate in evidence-level or claim-upgrade decisions whenever feasible.
Research Records
Core research records should be retained for a period appropriate to legal, ethical, contractual, journal, and funder requirements. ICR should adopt a documented retention schedule rather than assuming one universal period fits every study type.
Current Coherence Reset Evaluation
The current Coherence Reset participant evaluation should remain classified as preliminary program evaluation/case-series evidence unless its design, oversight, comparator structure, measures, and prospective protocol justify a stronger classification. Its Institute-created 0-10 ratings are exploratory under WP-023 unless separately validated.
Healthcare-Facing Threshold
Claims intended for hospitals, VA settings, colleges, licensed clinicians, or healthcare integration should face a higher institutional threshold than wellness education. Healthcare-facing claims should generally require validated measures, prospective design, appropriate controls, independent replication, safety evidence, and external peer review matched to the claim.
Institutional Falsification Commitments
ICR commits to publish or preserve negative findings; revise constructs that repeatedly fail; retire measures that do not validate; distinguish failed intervention tests from framework tests; not use testimonials as efficacy evidence; not infer cellular mechanisms from wellness observations; not treat one device as a whole-person measure; not suppress contradictory evidence; not upgrade claims without review; and permit external evidence to overturn Institute-originated ideas.
Minimum ICR Research Report
Every formal ICR research report should identify the claim tested, evidence level sought, protocol/version, registration status, population, inclusion/exclusion criteria, ethics/consent status, design, comparator where applicable, intervention/exposure, primary and secondary outcomes, measurement status, sample-size rationale, analysis plan, deviations, missing data, effect estimates with uncertainty, adverse events, limitations, conflicts, data/material availability, evidence-level conclusion, and permitted claim language.
Institutional Decision Rule
When evidence is ambiguous, ICR should choose the narrower claim. When measurement is uncertain, ICR should report the uncertainty. When results conflict, ICR should preserve the conflict. When a commercial interest favors a stronger interpretation, governance should favor the interpretation the evidence can actually carry.
Canonical Public Definition
The ICR Research and Validation Standard is the Institute's rulebook for deciding how research is planned, measured, analyzed, reported, replicated, and translated into public claims. Its governing principle is simple: claim no more than the strongest reproducible evidence directly supports.
68. Conclusion
The Institute for Coherence and Regulation should be judged not by the number of claims it can make, but by the discipline with which it determines what the evidence permits.
A conceptual framework earns credibility through measurement, transparent testing, correction, replication, and willingness to accept unfavorable results.
The governing rule remains: ICR will claim no more than the strongest reproducible evidence directly supports.
Declarations
Author and originator: David Fischer. Institutional affiliation: Institute for Coherence and Regulation (ICR), Knightdale, North Carolina, USA.
Competing interests: The author has intellectual and commercial interests in CRF, ICR educational programs, certifications, publications, and wellness services. These interests should be disclosed in future empirical studies.
Ethics: This conceptual governance white paper reports no human-subject research. Data availability: No dataset was generated.
Canonical designation: ICR-WP-024, Version 1.0, September 2026.
Harmonization note: Version 2.0 formally separates WP-010 methodology, WP-023 measurement architecture, and WP-024 governance; retains the 0-6 evidence ladder; updates randomized-trial protocol/reporting references to SPIRIT 2025 and CONSORT 2025; adds explicit commercial and device firewalls, internal Study/Claim/Measurement Registries, healthcare-facing evidence thresholds, correction/retraction governance, and a minimum ICR research-report standard.
Chan, A.-W., et al. (2025). SPIRIT 2025 statement: updated guideline for protocols of randomised trials. BMJ, 389, e081477. https://doi.org/10.1136/bmj-2024-081477
Hopewell, S., et al. (2025). CONSORT 2025 statement: updated guideline for reporting randomised trials. BMJ, 389, e081123. https://doi.org/10.1136/bmj-2024-081123
EQUATOR Network. Reporting guidelines for health research, including CONSORT, SPIRIT, STROBE, PRISMA, and CARE. https://www.equator-network.org/
References
Hopewell, S., Chan, A.-W., Collins, G. S., et al. (2025). CONSORT 2025 statement: updated guideline for reporting randomised trials. BMJ, 389, e081123. https://doi.org/10.1136/bmj-2024-081123
Chan, A.-W., Boutron, I., Hopewell, S., et al. (2025). SPIRIT 2025 statement: updated guideline for protocols of randomised trials. BMJ, 389, e081477. https://doi.org/10.1136/bmj-2024-081477
von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., & Vandenbroucke, J. P.; STROBE Initiative. (2007). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Medicine, 4(10), e296. https://doi.org/10.1371/journal.pmed.0040296
Gagnier, J. J., Kienle, G., Altman, D. G., Moher, D., Sox, H., Riley, D., & CARE Group. (2014). The CARE guidelines: consensus-based clinical case report guideline development. Journal of Clinical Epidemiology, 67(1), 46-51. https://doi.org/10.1016/j.jclinepi.2013.08.003
Page, M. J., McKenzie, J. E., Bossuyt, P. M., et al. (2021). The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71
Mokkink, L. B., Terwee, C. B., Patrick, D. L., et al. (2010). The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study. Quality of Life Research, 19, 539-549. https://doi.org/10.1007/s11136-010-9606-8
Nosek, B. A., Ebersole, C. R., DeHaven, A. C., & Mellor, D. T. (2018). The preregistration revolution. Proceedings of the National Academy of Sciences, 115(11), 2600-2606. https://doi.org/10.1073/pnas.1708274114
Appendix A - ICR Claim Review Form
Claim ID:
Exact proposed wording:
Public audience:
Claim type:
Current evidence level:
Direct supporting evidence:
Contradictory/null evidence:
Measurement validity:
Major alternative explanations:
Independent replication status:
Conflict-of-interest considerations:
Permitted wording:
Prohibited stronger wording:
Decision:
Review date:
Reviewer(s):
Appendix B - Canonical Public Statement
The ICR Research and Validation Standard requires Institute claims to remain proportional to the strongest reproducible evidence directly supporting them. Conceptual proposals, exploratory observations, associations, comparative effects, mechanisms, measurement validation, and replicated findings are treated as distinct evidence states. ICR research should use established external reporting and methodological standards, disclose conflicts, report null and adverse findings, preserve protocol transparency, and seek independent replication before broad claims.