{"schema_version":1,"research_id":"eoa_inverse_innovation_exp05_external_evaluation_20260803","source_assessment_id":"layer_decay_and_expiration_management__psychology:P5:v0","cell_id":"layer_decay_and_expiration_management__psychology","search_queries":["site:hhs.gov OHRP protocol changes IRB approval except immediate hazards records research","site:fda.gov protocol deviations important protocol deviation guidance clinical investigations","behavioral research protocol deviations inconsistent administration reproducibility study procedures","laboratory protocol version control deviations software product electronic protocol deviation management","site:pmc.ncbi.nlm.nih.gov psychology experiment protocol deviations procedural fidelity experimenters study","site:pubmed.ncbi.nlm.nih.gov preregistration deviations psychology experiments adherence primary study","site:osf.io support registrations version control frozen registration amendments","ICH E6 R3 essential records protocol amendments version history deviations official","site:help.osf.io registrations edit approved registration amendment version history","site:help.osf.io wiki version history OSF files versions official","electronic lab notebook protocol version control audit trail deviations product official research","clinical trial protocol deviation management software workflow official product audit trail","clinical trial protocol amendment impact assessment software version effective date workflow official product","behavioral experiment protocol management software run sheet conditional instructions session official","site:openclinica.com protocol deviations audit trail version study official","site:castoredc.com protocol deviation tracker study versioning official","\"Raising the value of research studies in psychological science\" publication date","\"Comparing dream to reality\" 211037 publication date","site:help.osf.io/article/387-files \"Last updated\"","Prelude EDC \"Simplifying Mid-Study Updates\" date"],"sources":[{"source_id":"S1","title":"Comparing dream to reality: an assessment of adherence of the first generation of preregistered studies","publisher":"Royal Society Open Science","url":"https://pubmed.ncbi.nlm.nih.gov/34729209/","source_class":"PRIMARY_RESEARCH","publication_date":"2021-10-27","accessed_at":"2026-08-03","claims_supported":["In 27 preregistered Psychological Science studies, only two contained no deviations from the preregistration.","Nine studies disclosed none of their identified deviations; common discrepancies involved sample size, exclusion criteria, and analysis.","The study establishes that procedural-plan deviations and incomplete disclosure visibly occur in psychology, although it does not measure accumulated operational exceptions during session administration."]},{"source_id":"S2","title":"Raising the value of research studies in psychological science by increasing the credibility of research reports: the transparent Psi project","publisher":"Royal Society Open Science","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC9890107/","source_class":"PRIMARY_RESEARCH","publication_date":"2023-02-01","accessed_at":"2026-08-03","claims_supported":["A multi-laboratory psychology project used laboratory logs to document protocol deviations and contextual events affecting data.","The authors state that logs can reveal protocol-fidelity problems and identify data affected by mistakes or unexpected events.","This demonstrates a relevant laboratory workflow and expressed investigator need for traceable procedural-state information, but not demand for exception-expiration software."]},{"source_id":"S3","title":"Protocol Deviations for Clinical Investigations of Drugs, Biological Products, and Devices","publisher":"U.S. Food and Drug Administration","url":"https://www.fda.gov/regulatory-information/search-fda-guidance-documents/protocol-deviations-clinical-investigations-drugs-biological-products-and-devices","source_class":"GOVERNMENT_OR_REGULATOR","publication_date":"2024-12","accessed_at":"2026-08-03","claims_supported":["FDA says its regulations do not define protocol deviation or provide a deviation-classification system.","FDA identifies consistent classification, reporting, and documentation as important to interpretable and useful deviation information.","The guidance names sponsors, investigators, and IRBs as responsible stakeholders, while remaining draft, nonbinding, and primarily applicable to regulated clinical investigations rather than ordinary behavioral laboratories."]},{"source_id":"S4","title":"Institutional Review Board (IRB) Written Procedures: Guidance for Institutions and IRBs","publisher":"U.S. Department of Health and Human Services, Office for Human Research Protections","url":"https://www.hhs.gov/ohrp/regulations-and-policy/requests-for-comments/guidance-for-institutions-and-irbs/index.html","source_class":"OFFICIAL_GUIDANCE","publication_date":"2011","accessed_at":"2026-08-03","claims_supported":["Changes to approved human-subjects research generally may not be initiated without IRB review and approval, except to eliminate apparent immediate hazards.","IRB procedures should address proposed-change documents, review routes, timelines, consent-form consequences, written decisions, and audits of research records.","This identifies the institution, IRB, and investigator as authorizers and supports the proposal's prohibition on automatic expiration of ethics-controlled rules."]},{"source_id":"S5","title":"ICH E6(R3) Guideline for Good Clinical Practice","publisher":"International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use","url":"https://database.ich.org/sites/default/files/ICH_E6%28R3%29_Step4_FinalGuideline_2025_0106_ErrorCorrections_2025_1024.pdf","source_class":"STANDARD","publication_date":"2025-10-24","accessed_at":"2026-08-03","claims_supported":["Essential trial records should be identifiable and version controlled and should identify authors, reviewers, and approvers as appropriate.","Repositories should support identification, version history, search, retrieval, readable retention, and traceable alteration.","Investigators should document deviations and prevent accidental or premature destruction of required records; these controls establish adjacent practice but do not specify expiring, context-selected exception objects."]},{"source_id":"S6","title":"Files & Folders: File Versions on OSF","publisher":"Center for Open Science","url":"https://help.osf.io/article/387-files","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"2026-07-21","accessed_at":"2026-08-03","claims_supported":["OSF preserves replaced project files as downloadable revisions.","OSF exposes version details and leaves provenance information after irreversible deletion.","This is established whole-file versioning and provenance prior art, but the documentation does not describe authority-scoped leases, context compilation, or dependency-gated exception retirement."]},{"source_id":"S7","title":"Protocol Deviation Management Software – Vera Deviate","publisher":"Bionabu","url":"https://www.bionabu.com/vera-deviate","source_class":"COMMERCIAL_FIRST_PARTY","publication_date":"2026","accessed_at":"2026-08-03","claims_supported":["The product consolidates protocol deviations into governed records with status, owner, review roles, electronic signatures, escalation, closure, and an append-only audit trail.","It imports deviation information from EDC exports and site logs and maps records by subject.","This substantially overlaps the proposed inventory, ownership, lifecycle-state, and audit functions, though the public description does not show pre-session rule compilation, review-lease expiration, dependency-safe retirement, or historical run-sheet restoration."]},{"source_id":"S8","title":"Simplifying Mid-Study Updates with Prelude EDC’s Protocol Amendment Module","publisher":"Prelude","url":"https://preludeedc.com/resource/simplifying-mid-study-updates-with-prelude-edcs-protocol-amendment-module/","source_class":"COMMERCIAL_FIRST_PARTY","publication_date":"2024-03-12","accessed_at":"2026-08-03","claims_supported":["Prelude supports protocol versions, effective dates, testing before deployment, site-specific amendments, and automatic assignment of subjects to a protocol version.","The platform dynamically presents eCRFs according to the applicable protocol version and records the version in effect for new subjects.","This is the closest located analogue to context-sensitive current-state compilation, but operates on protocol/eCRF versions rather than separately governed conditional exception layers."]}],"problem_evidence":{"support":"STRONG","rationale":"Primary psychology evidence shows deviations from preregistered plans are common and sometimes undisclosed, while a multi-laboratory psychology project treated laboratory logs of procedural deviations as important for fidelity and data interpretation. FDA independently identifies inconsistent deviation classification and documentation as a problem. Evidence specifically quantifying stale, contradictory operational exceptions in behavioral laboratories was not found, so the exact proposed failure mode remains partly inferred.","source_ids":["S1","S2","S3"]},"stakeholder_evidence":{"support":"STRONG","rationale":"The identifiable adopter is a principal investigator or laboratory manager; the institution and IRB are authorizers for human-subjects changes. HHS expressly requires controlled review and communication of approved-study changes, and FDA/ICH assign deviation-documentation and record-control duties to investigators, sponsors, and IRBs. No behavioral-laboratory procurement commitment or budget-holder interview was found.","source_ids":["S2","S3","S4","S5"]},"prior_art":{"proximity":"SUBSTANTIAL_COLLISION","closest_analogues":[{"name":"Prelude Protocol Amendment EDC Module","similarity":"Supports versioned, effective-dated protocol changes, site-specific applicability, predeployment testing, and automated subject-to-version assignment—close to compiling the currently applicable procedural state.","remaining_difference":"Public documentation describes whole protocol/eCRF versions, not individually expiring conditional exceptions with independent authority, dependency, quarantine, and restoration states.","source_ids":["S8"]},{"name":"Bionabu Vera Deviate","similarity":"Provides structured deviation records, ownership, role-based review, lifecycle status, closure, signatures, and an append-only audit trail.","remaining_difference":"It appears oriented toward capturing and resolving deviations after occurrence; public materials do not show approved-exception compilation into future session instructions or lease-based sunset.","source_ids":["S7"]},{"name":"OSF project file versioning","similarity":"Preserves prior file versions, makes them retrievable, and retains deletion provenance.","remaining_difference":"It versions documents rather than resolving conditional rule objects, approvals, session contexts, incompatibilities, and live dependencies.","source_ids":["S6"]},{"name":"ICH E6(R3) essential-record management","similarity":"Establishes version control, approval attribution, traceable alteration, retention, search, retrieval, and protection against premature destruction as standard trial-record practice.","remaining_difference":"It specifies governance outcomes rather than a session-time exception compiler or review-lease mechanism.","source_ids":["S5"]}],"distinctive_claim_remaining":"Against effective-dated whole-protocol versioning plus conventional deviation tracking, representing each approved conditional exception as a separately authorized, expiring object and compiling only context-matching active objects before a session will reduce obsolete or unapproved rule inclusion and reviewer disagreement without increasing protected-exception omissions, while dependency gates and archived manifests preserve exact batch reconstruction. This is contrastive and falsifiable, but not yet empirically supported.","confidence":"HIGH"},"implementation_evidence":{"support":"MODERATE","rationale":"Existing products demonstrate version/effective-date assignment, subject-specific presentation, structured deviation workflows, audit trails, and recoverable versions; ICH specifies the necessary record controls. The remaining integration is technically plausible using a rule store, immutable identifiers, approval states, context predicates, manifests, and read-only compilation. Unverified issues include completeness of trigger metadata, detection of informal dependencies, restoration of executable environments, privacy of sensitive accommodation/deviation records, usability burden, and validation for regulated deployment.","source_ids":["S5","S6","S7","S8"]},"scores":{"meaningful_impact":{"score":4,"rationale":"Procedural inconsistency can affect participant treatment, data validity, and reconstruction; primary and official sources establish those concerns, though realized impact of this intervention is unmeasured.","source_ids":["S1","S2","S3","S4"]},"stakeholder_pull":{"score":4,"rationale":"Investigators, institutions, and IRBs have explicit governance duties and commercial deviation products indicate operational demand; direct purchasing intent from behavioral laboratories is absent.","source_ids":["S3","S4","S5","S7"]},"incremental_advantage":{"score":3,"rationale":"Object-level sunset and session-context compilation could improve on whole-file versions and post-hoc deviation logs, but Prelude already implements effective-dated, criterion-based protocol assignment.","source_ids":["S7","S8"]},"distinctiveness_plausibility":{"score":2,"rationale":"No exact public match was located, but the proposal assembles several established capabilities and substantially collides with amendment-management and deviation-workflow products.","source_ids":["S5","S6","S7","S8"]},"technical_implementability":{"score":4,"rationale":"Every major primitive except reliable semantic dependency discovery is demonstrated in existing record, deviation, or amendment systems; integration and validation remain nontrivial.","source_ids":["S5","S6","S7","S8"]},"adoption_authority_feasibility":{"score":3,"rationale":"PI and IRB authority boundaries are identifiable, but deployments spanning ordinary, safety, consent, and accommodation rules require institution-specific policy and validation.","source_ids":["S3","S4","S5"]},"evidence_readiness":{"score":3,"rationale":"A participant-free synthetic test is readily constructible with objective composition and reconstruction outcomes, but no comparative result yet exists.","source_ids":["S1","S2","S8"]},"safety_net_benefit":{"score":4,"rationale":"Human authorization, immutable manifests, quarantine, tombstones, and restoration are well aligned with official prohibitions on unauthorized change and premature record destruction; effectiveness is untested.","source_ids":["S4","S5","S6"]},"scalability":{"score":3,"rationale":"Rule compilation and structured workflows can scale computationally, but metadata capture, authority review, exception modeling, and institutional integration may scale poorly across heterogeneous laboratories.","source_ids":["S5","S7","S8"]}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"10K_TO_50K","scope":"Preregister, build, and run a participant-free synthetic comparison with approximately 40–80 trained reviewers, seeded exception histories, rollback, restoration, scoring, and analysis.","confidence":"MODERATE","assumptions":["Uses an internal low-code or lightweight web prototype rather than a validated regulated system.","Resource-equivalent estimate includes investigator, developer, reviewer, and analysis time.","No live participants, protected IRB records, or production integrations are used."],"source_ids":["S1","S2","S6","S8"]},"initial_deployment_startup":{"band_2026_usd":"50K_TO_250K","scope":"Single-laboratory MVP with exception schema, role controls, read-only run-sheet compiler, immutable manifests, quarantine, archive, and basic import from existing documents.","confidence":"LOW","assumptions":["One institution and one protocol family.","Manual review resolves uncertain context and dependency matches.","Excludes formal regulated-system validation and complex identity-management integration."],"source_ids":["S5","S7","S8"]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Institutional launch across multiple laboratories including security review, identity and document-system integration, migration, policy design, training, accessibility, validation, monitoring, and support.","confidence":"LOW","assumptions":["Approximately 5–20 laboratories at one institution.","Includes professional services and internal staff resource equivalents.","Does not include multinational regulated-trial deployment or replacement of an enterprise CTMS/EDC."],"source_ids":["S4","S5","S7","S8"]},"annual_recurring":{"band_2026_usd":"50K_TO_250K","scope":"Hosting, support, access administration, policy maintenance, exception review, archive/restore exercises, audits, training, and incremental integrations for one institution.","confidence":"LOW","assumptions":["Moderate study volume and storage.","Human review remains mandatory for protected exceptions.","No public vendor pricing was located; range is workload- and scope-based rather than procurement-verified."],"source_ids":["S4","S5","S7"]}},"verified_pipeline_gates":{"externally_supported_problem":{"status":"YES","reason":"Primary psychology studies document plan deviations, incomplete disclosure, and laboratory reliance on deviation logs; FDA independently identifies inconsistent deviation classification and documentation.","source_ids":["S1","S2","S3"]},"externally_credible_adopter_or_authorizer":{"status":"YES","reason":"The PI/laboratory manager is a credible adopter, while the institution and IRB have explicit authority over changes to approved human-subjects research.","source_ids":["S2","S3","S4","S5"]},"distinct_testable_incremental_claim":{"status":"YES","reason":"The proposal can be directly compared with document/version and deviation-log baselines on obsolete-rule inclusion, protected-rule omission, agreement, time, dependency detection, and historical reconstruction.","source_ids":["S6","S7","S8"]},"bounded_next_evidence_step":{"status":"YES","reason":"A synthetic, participant-free randomized evaluation can exercise current-state compilation, quarantine rollback, and archive restoration without changing a live study.","source_ids":["S1","S2","S8"]},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"The next step is read-only and synthetic; live use remains gated by PI/IRB authority, and ethics- or safety-linked rules are excluded from automatic expiry.","source_ids":["S4","S5"]},"credible_cost_scope_and_range":{"status":"UNCERTAIN","reason":"The four bands are explicitly scoped and consistent with the demonstrated breadth of comparable systems, but no public vendor pricing, labor quotation, institutional integration estimate, or regulated-validation estimate was found.","source_ids":["S5","S7","S8"]}},"next_evidence_step":"Preregister and run a participant-free randomized synthetic study. Seed one behavioral protocol with at least 24 exception objects spanning equipment substitution, remote administration, prompt clarification, participant accommodation, exclusion handling, consent protection, supersession, departed ownership, missing approval, and batch-specific deviation. Randomly assign 40–80 trained reviewers to (A) a representative baseline of protocol files, amendments, messages, and deviation log; (B) effective-dated whole-document versioning; or (C) the read-only Sunset Board. For at least 12 session contexts and 8 historical batches, measure exact run-sheet composition, obsolete/unapproved inclusion, protected-exception omission, reviewer agreement, completion time, dependency-detection precision/recall, and exact historical reconstruction. Exercise one erroneous quarantine and rollback and one archive restoration. Falsify the intervention if it fails to reduce obsolete/unapproved inclusion versus both comparators, omits any seeded consent/safety/accommodation protection, lowers reconstruction accuracy, increases median completion time by more than 25%, or produces an untraceable included rule.","blocking_evidence":["No field estimate establishes the prevalence, size, or consequences of unresolved operational-exception stacks specifically in behavioral laboratories.","No comparative test shows that object-level expiration and context compilation outperform effective-dated whole-protocol assignment or a structured deviation-management product.","No evidence establishes that explicit trigger metadata can safely represent tacit expertise, ambiguous contexts, or rare participant protections.","Informal dependencies in staff memory, analysis code, external replication packages, and local script copies may be invisible to the proposed deletion gate.","Privacy, records-retention, accessibility, and validation requirements have not been assessed for any particular institution or jurisdiction.","The four cost bands lack vendor pricing, implementation quotations, and measured reviewer workload."],"research_disposition":"PARTNERED_RESEARCH_PROGRAM","world_novelty_boundary":"The search supports novelty only relative to the eight opened public sources. World novelty, patentability, freedom to operate, market size, and realized impact were not measured. Unopened proprietary systems, patents, institutional tools, and non-public product capabilities may contain the same combination.","arm":"COMPLETE_PROPOSAL_PORTFOLIO","candidate_version":0,"controller_recommendation":{"action":"STOP_EMPIRICAL_RESEARCH_NEEDED","repairable":false,"material_progress_observed":true,"progress_targets":["Complete the preregistered three-arm synthetic comparison against both document/message practice and effective-dated whole-protocol versioning.","Demonstrate zero protected-exception omissions and full authority traceability across all seeded session contexts.","Show a predeclared reduction in obsolete or unapproved rule inclusion and improved reviewer agreement without more than a 25% median time penalty.","Achieve at least 95% precision and recall on seeded dependencies and 100% exact reconstruction for historical batches.","Successfully roll back an erroneous quarantine and restore/recompile every sampled archived run sheet.","Measure reviewer workload, metadata ambiguity, false context matches, and failures caused by tacit or informal dependencies.","Obtain institution-specific legal, privacy, records, accessibility, security, and authority review before any live-study pilot.","Replace assumption-based cost bands with a prototype labor log and at least one deployment or vendor quotation."],"reason":"Bounded web research establishes a real governance problem, credible authorities, implementable components, and substantial adjacent prior art, but it cannot determine whether the integrated exception-sunset compiler improves procedural accuracy without creating protected-rule omissions or excessive burden. That evidence requires a controlled empirical test. Under the required controller rule, this STOP recommendation is non-repairable."},"proposal_index":5}