# Adversarial review log — menopause-hormone-symptoms

Authoring model: **Sol 6.1**. Adversarial reviewer: **Claude Opus 5.5 (Anthropic)**. Date: **2026-10-14**.
Round 2 of the Sol 6.1 adversarial review programme; group 7 of `NEW_REVIEW_INVENTORY_2026-10-14.json`.

## Summary

| Item | Value |
|---|---|
| Level before | Ζ — Multiple independent causal lines converge / High |
| Level after | Ζ — Multiple independent causal lines converge / High (unchanged) |
| Findings: Critical | 0 |
| Findings: Major | 2 |
| Findings: Minor | 2 |
| Findings: Editorial | 1 |
| Rejected findings | 7 |
| PMIDs checked / failed | 68 / 0 |
| Load-bearing claims spot-checked | 15 |
| Validator | FAIL (1 fail, 0 warn, 32 pass) — `v5.lay_summary` only; identical before and after. The shipped validator is 2.4 and enforces a v5.4 requirement against a v5.3 package. |
| Approval | **Approved as revised.** |

## Findings

| ID | Severity | Location | Original → corrected | Basis |
|---|---|---|---|---|
| AR-1 | Major | AT A GLANCE table; "Final score, causal direction and M/S/D"; score card `extended_scoring_MSD.momentum` | Momentum "Decisive work underway" → "Mechanistic only"; artifact rewritten | The named artifact, OASIS-3 (PMID 40920404), is a completed phase 3 trial of a nonhormonal intervention bearing on the rank-2 rival model, not an experiment that could cross SC2's next threshold. The review's own Stability entry says the raising triggers "are proposed rather than falsely declared ongoing named triggers", contradicting the Momentum value. Momentum modifies no letter. |
| AR-2 | Minor | Evidence cards [22] and [26]; the same two score-card ledger rows | "ErratumIn: PMIDnot supplied." → "ErratumIn: PMID not supplied." | Reference-builder rendering defect (missing space). PubMed genuinely supplies no erratum PMID for PMID 16932240 or PMID 16735935, so only the rendering was wrong. |
| AR-3 | Editorial | PROPOSED DECISIVE EXPERIMENTS readout/controls line | Four doubled periods → single periods | Field-concatenation rendering defect. |
| AR-4 | Major | AT A GLANCE, "Best-supported rival" line | Nominal ("Sex-steroid replacement…", rank 1) → "Nonhormonal thermoregulatory pathway modulation (rank 2)" | Renderer defect: the "Best-supported rival" line is generated from `competing_models.models_evaluated[0]`, which here carries `is_nominal: true` and rank 1, so the prose named the nominal as its own rival. Prose-only; the tournament, the level and Rule 9 are untouched. |
| AR-5 | Minor | Score card `final_score.confidence_rationale` | Access depth removed as a confidence determinant | The framework ties confidence to evidence density and reproducibility; the boilerplate rationale named "incomplete access" as a determinant. Rewritten to rest on density, replication and heterogeneity. Band re-tested against both band texts and unchanged. High stands: placebo-controlled MHT symptom trials span many independent groups, routes, countries and decades. |

## Rejected findings (checked, no change)

1. SC2's Ζ independence basis — the two lines share systemic-estrogen pharmacology, which argues for Ε; rejected because the framework's own independence list credits "a prediction derived from the mechanism … tested by an instrument unrelated to the one that generated the mechanism", which the sweating-threshold/skin-conductance work (PMIDs 11872200, 10411797, 15772562) satisfies against the symptom diary. The framework text does not force a change. Most contestable judgement in the review.
2. Citation years one later than PubMed e-pub year in references [4], [28], [29], [32], [33], [34] — print/e-pub convention.
3. Fused PMID labels in prose ("PMID24861828") — consistent house style, not the "randomize3808" defect class.
4. Collective-author leading-comma defect, empty `DOI: .` fields, retrieval handles — screened, none present; all 64 ledger rows carry PMID and DOI.
5. Aggregation — primary-subclaim pre-registered, SC2 named, both alternatives reported (weakest-link Β, scope-matched Ζ); no Ε subclaim sets the headline.
6. Rubric adaptation (metabolism/aging rubric for a menopause treatment claim) — declared prospectively, literal criteria quoted unchanged, substitution stated.
7. Lead-reported artifacts screened: no false-approval boilerplate (not a revised edition), no reviewer-notes/provenance contradiction. The confidence band was re-tested against the framework's Moderate and High texts and High stands.

## Benefit-and-risk check (treatment-relevant review)

The review carries no numeric effect size in prose, so absolute-versus-relative conflation is impossible. Directions, populations and endpoints all matched source: WHIMS dementia increase in women *enrolled at 65 or older* (12771112); APHRODITE benefit at 300 µg/day but not 150 µg/day with four breast-cancer diagnoses unresolved (18987368); PMID 16735935's satisfying-event endpoint not significant; MsFLASH's imprecise active-active difference (24861828); ELITE confined to CIMT with no coronary-CT difference (27028912); REPLENISH dose-specific with a bounded endometrial-safety population (29889748); testosterone muscle gains in the 25 mg supraphysiologic arm only (24281237).

## Residual limitations

Validator FAIL is a framework-version mismatch (v5.4 lay summary vs a v5.3 package); no lay summary was fabricated. Access depth unchanged (8 partial-body, 56 bounded abstract/status), so full-text and supplement numbers remain unaudited.

## Strings for the lead to propagate to package-level files

| Old | New |
|---|---|
| `Decisive work underway` (menopause-hormone-symptoms row, Momentum column) | `Mechanistic only` |
| `Best-supported rival: Sex-steroid replacement reduces hormone-withdrawal symptoms` | `Best-supported rival: Nonhormonal thermoregulatory pathway modulation (rank 2)` |

No level, confidence, role, source count or claim count changed, so OVERVIEW.md / READER.html / EDITORIAL_HANDOFF.md / BATCH_COMPLETION.json need no other edit for this case.

**Lead policy recorded (confidence bands, 2026-10-14).** The lead adjudicated the question raised in this pass and set a two-step test for the corpus. Step 1: strike access depth as a determinant everywhere — how much of a paper the reviewer could read is neither evidence density nor reproducibility, and it never justifies a band. Step 2: re-derive the band from what remains; if access depth was the only thing holding the band down and the cited evidence meets the High text, raise it, but if the review documents substantive evidence-quality or reproducibility concerns (indirect effects, incomplete washout, residual confounding of the mechanism, reporter rather than endogenous readouts) those are valid determinants and the band stands. A reviewer's uncertainty about the evidence is a legitimate reason for Moderate; a reviewer's uncertainty about their own reading access is not. This pass's rationale rewrite is step 1; the band decision below is step 2. Step 1 applied (rationale rewritten under AR-5). Step 2: band stands at High, which it already was; placebo-controlled hormone trials for vasomotor symptoms span many independent groups, routes, countries and decades, meeting the High text.

**Approval:** Following adversarial review and the corrections listed above, the revised document is approved. — Claude Opus 5.5 (Anthropic), 2026-10-14.