{
  "slug": "hygiene-hypothesis",
  "canonical": "hsef_v53_final_nine_2026-10-12\\hygiene-hypothesis\\hygiene-hypothesis.review.md",
  "authoring_model": "Sol 6.1",
  "adversarial_reviewer": "Claude Opus 5.5 (Anthropic)",
  "date": "2026-10-14",
  "level_before": {
    "hsef_level": "Γ",
    "confidence": "Moderate"
  },
  "level_after": {
    "hsef_level": "Γ",
    "confidence": "Moderate"
  },
  "level_changed": false,
  "pmids_checked": 63,
  "pmids_failed": 0,
  "retractions_or_eocs": 0,
  "claims_spot_checked": 12,
  "claims_contradicted": 0,
  "validator": "PASS (0 fail, 0 warn, 31 pass)",
  "counts": {
    "Critical": 0,
    "Major": 0,
    "Minor": 3,
    "Editorial": 1
  },
  "findings": [
    {
      "id": "D1",
      "severity": "Minor",
      "location": "Final score, causal direction and M/S/D; scorecard extended_scoring_MSD.divergence",
      "original": "Divergence: Contested. No manufactured field consensus or arithmetic gap.",
      "corrected": "Contested verdict retained with its required artifact naming the two framings",
      "evidence": "hsef-framework.md Divergence required artifact"
    },
    {
      "id": "D2",
      "severity": "Minor",
      "location": "Consensus and original-source reconciliation",
      "original": "hazard ratio1.16, confidence interval0.82-1.64; immune-marker changes over14days",
      "corrected": "hazard ratio 1.16, confidence interval 0.82-1.64; immune-marker changes over 14 days",
      "evidence": "Round-2 brief, known upstream defect 4; values re-verified against PMID 42705251 (HR 1.16, 95% CI 0.82-1.64, log-rank p 0.35)"
    },
    {
      "id": "D3",
      "severity": "Editorial",
      "location": "REFERENCES item 35",
      "original": "DOI: .",
      "corrected": "DOI: none indexed.",
      "evidence": "PubMed record for PMID 9179425 carries no DOI"
    },
    {
      "id": "D4",
      "severity": "Minor",
      "location": "score card final_score.confidence_rationale",
      "original": "density, heterogeneous controls and incomplete access set the confidence, not correctness.",
      "corrected": "access depth dropped as a determinant; High-band re-test recorded; band holds at Moderate",
      "evidence": "framework Confidence band text"
    }
  ],
  "rejected_findings": [
    "Gamma for SC3 despite two positive randomized eczema trials - those trials supply a defined product rather than perturbing the registered general-deprivation exposure; cap precedence assigns them to SC2 at Epsilon",
    "Aggregation - primary-subclaim on SC3 declared before scoring; weakest-link Beta and scope-matched Gamma reported; SC2's Epsilon is scope-restricted and cannot set the headline",
    "Epsilon rather than Delta for SC2 - the context boundary is mapped (strain divergence inside one randomization, an adverse sensitisation result, genotype-dependent reversal)",
    "Role of mechanism 'reinforcing / stabilizing factor' - derived to step 5 with direction stated as factor- and host-dependent",
    "Momentum 'Mechanistic only' - artifact names specific work and treats ORBEX as completed; the round-2 momentum defect is not present",
    "Stability 'Stable' - bounded search disclosed",
    "Timeline years - each milestone matches its record's publication year",
    "Treatment of LEAP and EAT as food-antigen tolerance rather than microbial restoration, with the EAT intention-to-treat null retained - checked and correct",
    "Confidence band re-tested against the High text after removing access depth: short trials diverging by strain and endpoint, with several reports sharing one parent randomization and overlapping cohorts, is the Moderate text; band holds",
    "Coordinator-flagged Consensus-splice artifacts do not apply: this case is not a 2026-10-09 revised edition; its Consensus reconciliation section is original to the review",
    "Best-supported rival renderer defect: models_evaluated[0] here is M2, is_nominal false, rank 1, with the nominal at rank 3; the prose names the rival correctly"
  ],
  "residual_limitations": "Six partial-body and 57 bounded abstract/status examinations. Methods- and Results-level details absent from the abstracts (the Welsh trial's recruited risk mix, the adult age of mice in the Amish dust experiment, the ineffective SCFA mixture and absent durable Clostridial engraftment, the intraperitoneal route in the NOD LPS experiment) were checked for internal consistency only. The autoimmune clinical branch rests on observational and model evidence, as the review states.",
  "files_changed": [
    "hygiene-hypothesis.review.md",
    "hygiene-hypothesis.scorecard.json",
    "SHA256.txt",
    "validator-adversarial-2026-10-14.txt (new)",
    "ADVERSARIAL_REVIEW.md (new)",
    "adversarial-findings.json (new)"
  ],
  "propagated_to": [
    "hsef_v53_remaining_priority_11_15_2026-10-11\\hygiene-hypothesis\\hygiene-hypothesis.review.md"
  ],
  "divergent_copies": [],
  "coordinator_round2_artifact_checks": {
    "false_approval_boilerplate": "not applicable (no spliced provenance block in this case)",
    "confidence_set_from_access_depth": "corrected; band re-tested against the High text and held at Moderate",
    "best_supported_rival_renderer_bug": "checked; models_evaluated[0] is a non-nominal rank-1 model, prose correct",
    "reviewer_notes_contradiction": "not present"
  }
}