Ten research investigations

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Question.** Do high-exposure oral silymarin preparations meaningfully improve liver activity in a defined chronic hepatitis C population?

**Primary original study:** Fried et al., *JAMA* 2012, randomized, multicenter, double-blind placebo-controlled SyNCH trial; **PMID [22797645](https://pubmed.ncbi.nlm.nih.gov/22797645/)**; DOI [10.1001/jama.2012.8265](https://doi.org/10.1001/jama.2012.8265). **154 participants** with chronic HCV and elevated ALT who had not responded to interferon-based treatment; silymarin in two exposure groups or placebo for 24 weeks.

**Results.** Primary ALT response met by **two participants in each arm** (about 4% per group; P >= .99). No significant between-arm difference in mean ALT change (P=.75), HCV RNA (P=.54), or quality-of-life outcomes. Treatment-group adverse event profiles were comparable in this trial.

**Limits.** This is a specific preparation and a selected *historical interferon-nonresponder* cohort, not direct-acting antiviral-era care, not a trial of all hepatic diagnoses, and not evidence that food thistle, every branded extract, or all doses have identical effects. Clinical lab-marker null is not a liver-injury safety certification.

**Editorial gate.** P0 candidate only if existing site copy implies milk thistle treats viral infection or that increased liver lab activity proves antiviral benefit. No self-treatment substitute language.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary original study:** Schulman et al., *JAMA* 2006, VINTAGE MI single-center randomized double-blind trial; **PMID [16391217](https://pubmed.ncbi.nlm.nih.gov/16391217/)**; DOI [10.1001/jama.295.1.58](https://doi.org/10.1001/jama.295.1.58). **153 enrolled** after first STEMI, L-arginine added to standard post-MI care or placebo; six months.

**Results.** No statistically significant benefit to ventricular ejection fraction or vascular stiffness. During follow-up **six participants (8.6%) assigned L-arginine died vs zero assigned placebo** (P=.01); the safety-monitoring committee stopped enrollment. The authors concluded L-arginine should not be recommended following acute MI.

**Limits.** The RCT is not proof of the exact causal pathway, a population-wide adverse-event incidence estimate, or a finding that food-derived arginine or other clinical populations have the same mortality risk. The negative ejection-fraction result and severe event imbalance both merit reporting, with precise *post-infarction* scope.

**Editorial gate.** **P0 safety-review candidate** for L-arginine product/article pages if any claim encourages cardiovascular use without distinguishing recent MI.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary original study:** Omenn et al., *NEJM* 1996, Beta-Carotene and Retinol Efficacy Trial; **PMID [8602180](https://pubmed.ncbi.nlm.nih.gov/8602180/)**; DOI [10.1056/NEJM199605023341802](https://doi.org/10.1056/NEJM199605023341802). Randomized, double-blind placebo-controlled cancer-prevention intervention combining beta-carotene with retinyl palmitate, **18,314** participants with smoking or asbestos-related high risk.

**Original-trial signal.** Trial terminated early in 1996 amid **28% greater lung-cancer incidence** and **17% greater overall mortality** in the combination group relative to placebo, as reported in subsequent trial accounting. An increased cardiovascular mortality signal was also identified.

**Original follow-on analysis of the same trial:** Goodman et al., *J Natl Cancer Inst* 2004, **PMID [15572756](https://pubmed.ncbi.nlm.nih.gov/15572756/)**; DOI [10.1093/jnci/djh320](https://doi.org/10.1093/jnci/djh320). The later follow-up reports initial harm and trajectories after intervention stopped. **Do not count follow-up as a second randomized cohort.**

**Limits.** This regimen combines **beta-carotene and preformed vitamin A** at high supplement exposures, in a high-risk population. It cannot attribute the entire effect separately to each ingredient or imply dietary carotenoids in vegetables carry the same hazard. Do not borrow the 28% figure as a risk estimate for all vitamin A products or never-smokers. No dosing suggestion.

**Editorial gate.** **P0** false-reassurance and population/combination safety check on antioxidant, vitamin A, beta-carotene and smoker-focused material.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary original study:** Lonn et al., *JAMA* 2005 HOPE / HOPE-TOO continuation, **PMID [15769967](https://pubmed.ncbi.nlm.nih.gov/15769967/)**; DOI [10.1001/jama.293.11.1338](https://doi.org/10.1001/jama.293.11.1338). Randomized original HOPE cohort: patients age >=55 with vascular disease or diabetes; extended median follow-up **7 years**. Vitamin E versus placebo, with important attrition/consent changes in the extension.

**Results.** Major cardiovascular events **21.5% vitamin E vs 20.6% placebo** (RR 1.04, 95% CI .96–1.14; P=.34), primary null. Cancer incidence **11.6% vs 12.3%** (RR .94, CI .84–1.06; P=.30), also null. Heart failure **increased** (RR **1.13**, CI 1.01–1.26; P=.03), with hospitalization-for-heart-failure RR **1.21** (CI 1.00–1.47; P=.045).

**Limits.** Do not interpret a favorable numerical cancer point estimate as prevention; do not treat an RCT extended cohort as a fully independent second trial. High supplement exposure in vascular-risk adults is not dietary vitamin E. Secondary-event findings should be reported with the primary-outcome null and multiplicity/extension context.

**Editorial gate.** **P0** if vitamin E content claims cardiovascular or cancer prevention without the large-event null or heart-failure signal.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary original study:** Nair et al., *NEJM* 2006, two-year double-blind randomized placebo-controlled investigation; **PMID [17050889](https://pubmed.ncbi.nlm.nih.gov/17050889/)**; DOI [10.1056/NEJMoa054629](https://doi.org/10.1056/NEJMoa054629). **87 older men** (different DHEA/testosterone/placebo arms) and **57 older women** (DHEA/placebo) with selected low baseline hormone measurements.

**Results.** Supplemented DHEA increased measured DHEA sulfate concentrations; changes in surrogate hormone levels **did not translate to physiologically relevant improvements** in body composition, exercise performance, insulin sensitivity or quality of life versus placebo. Certain sex/treatment-specific bone mineral-density and male testosterone fat-free-mass changes occurred, which must not be presented as global antiaging efficacy.

**Limits.** Small sex-specific groups, an endocrine exposure rather than ordinary dietary nutrition, selected older adults, and two-year duration. “No major harms seen” in a small trial does not establish long-term safety or safety in hormone-sensitive diseases.

**Editorial gate.** P1 “antiaging,” “energy,” libido or resilience marketing check; don't extrapolate from serum values to health benefits.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary original study:** Berth-Jones and Graham-Brown, *Lancet* 1993, placebo-controlled, double-blind three-arm atopic dermatitis study; **PMID [8099640](https://pubmed.ncbi.nlm.nih.gov/8099640/)**; DOI [10.1016/0140-6736(93)90697-F](https://doi.org/10.1016/0140-6736(93)90697-f). Arms were evening primrose oil, evening primrose + fish oil, or placebo for 16 weeks; **123 recruited, 102 completed**.

**Results.** Trial reported **no clinically demonstrated improvement** with active fatty-acid treatment in investigator-recorded disease severity, topical steroid requirement or participant symptom assessment relative to placebo.

**Caveats.** Study has a linked **1993 erratum**; verify corrected source text before any numerical exposure claims. Trial is of *atopic dermatitis*, not PMS, mastalgia, rheumatoid arthritis or other claimed uses; it is not an allergy/adverse-event incidence study.

**Editorial gate.** P1 no-blanket-eczema-cure review; distinguish evening primrose alone from the fish-oil combination.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary original study:** Newton et al., *Ann Intern Med* 2006, HALT trial; **PMID [17179056](https://pubmed.ncbi.nlm.nih.gov/17179056/)**; DOI [10.7326/0003-4819-145-12-200612190-00003](https://doi.org/10.7326/0003-4819-145-12-200612190-00003). **351 women age 45–55** reporting at least two vasomotor symptoms/day; one-year randomization among black cohosh, multibotanical, multibotanical plus soy counseling, hormone-therapy comparator or placebo.

**Results.** Herbal interventions showed **no statistically significant improvement vs placebo** in daily hot flashes or symptom intensity at 3, 6 or 12 months, except **worse intensity** at 12 months in the multibotanical + soy arm (P=.016). Do not generalize that negative to the hormone-therapy active comparator.

**Limits.** Exact product identity/species matters: black cohosh `Actaea racemosa` and Asian Actaea adulterants cannot be treated as the same material. The soy-counseling arm is a combined intervention, not an isolated-soy trial. Outcomes measured were menopausal vasomotor symptoms, not cancer, liver injury, or hormone lab response.

**Editorial gate.** P1 hot-flash efficacy confidence and adulteration caveat check.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary original study:** Kleefstra et al., *Diabetes Care* 2007; **PMID [17303791](https://pubmed.ncbi.nlm.nih.gov/17303791/)**; DOI [10.2337/dc06-2192](https://doi.org/10.2337/dc06-2192). Six-month randomized double-blind placebo-controlled study in a Western population with type 2 diabetes taking oral hypoglycemic medication, testing chromium **yeast**.

**Results.** **No significant between-group difference in change in HbA1c** (primary endpoint), nor measured lipids, BMI, blood pressure, body fat or insulin resistance. Claims based on pre/post blood sugar without comparator superiority would overstate efficacy.

**Limits.** Chromium yeast is **not chromium picolinate**; don't import effect/harm magnitudes from another chemical form or another population. This is not a comparison with modern diabetes medications, nor evidence to substitute for clinical care.

**Editorial gate.** P1 formulation-specific HbA1c null and “essential mineral = treats diabetes” language check.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary original study:** Levis et al., *Arch Intern Med* 2011, randomized double-blind placebo-controlled two-year trial; **PMID [21824950](https://pubmed.ncbi.nlm.nih.gov/21824950/)**; DOI [10.1001/archinternmed.2011.330](https://doi.org/10.1001/archinternmed.2011.330). **248 women** (122 assigned soy and 126 placebo), age 45–60 near menopause; bone mineral density was the prespecified primary outcome.

**Results.** Mean spine BMD changes **-2.0% soy vs -2.3% placebo**; total hip **-1.2% vs -1.4%**, femoral neck **-2.2% vs -2.1%**; **no significant between-group difference**. Hot flashes and constipation were reported more commonly in the soy group. The trial also did not demonstrate overall menopausal symptom prevention.

**Limits.** Soy isoflavone **tablets** are not equivalent to tofu, overall dietary soy protein, isolated genistein or hormone therapy. Bone mineral density is not the same endpoint as fractures; the trial is not a direct fracture prevention study. Other soy trials have varied results for vasomotor symptoms; no class-wide absolute verdict.

**Editorial gate.** P1 prevention claims for osteoporosis and blanket “natural estrogen replacement” framing.

Original notes reproduced from the merged research dossier without converting preliminary interpretations into approved medical recommendations. Source links are provided above. Other limitations in the full text still apply.

**Primary RCT A:** Prasad et al., *Ann Intern Med* 2000; **PMID [10929163](https://pubmed.ncbi.nlm.nih.gov/10929163/)**; DOI [10.7326/0003-4819-133-4-200008150-00006](https://doi.org/10.7326/0003-4819-133-4-200008150-00006). Double-blind placebo trial of **50 adults recruited within 24h of cold symptoms**, **48 completers**. Zinc acetate lozenges or placebo were used repeatedly while symptoms persisted. **Mean cold duration 4.5 vs 8.1 days**, favoring zinc; several self-rated symptom domains improved. **Measured cytokine differences were not statistically significant**: mechanistic improvement cannot be claimed from this experiment.

**Primary RCT B:** Hemilä et al., *BMJ Open* 2020; **PMID [31980506](https://pubmed.ncbi.nlm.nih.gov/31980506/)**; DOI [10.1136/bmjopen-2019-031662](https://doi.org/10.1136/bmjopen-2019-031662). Double-blind commercial-lozenge trial in Finnish working adults; **253 initially randomized to packages**, but **87 colds** contributed to primary analysis. **Recovery rate ratio 0.68**, 95% CI **0.42–1.08**, P=.10; **primary null**. Self-reported adverse effects were more common in zinc group (37% reported none vs 69% placebo reported none).

**Contradiction analysis.** These are **two genuinely independent randomized cohorts**, unlike different papers from CARET. They differ in products, protocol and treatment duration, enrollment-to-illness denominator, taste/unblinding risks and cold-episode ascertainment. The positive small RCT is not invalidated simply by a later null, and the null commercial trial cannot be ignored. No automatic pooled effect can be computed from their means and recovery hazard ratio. Neither trial examines zinc delivered intranasally or proves population-wide prevention of infection.

**Editorial gate.** P1 formulation/primary-endpoint/side-effect comparison. **PMID 31980506 appears in existing enrichment reconciliation files**, so the source must be reused if canonical mapping matches; do not double count it.

## Evidence ledger integrity, duplicate controls and release acceptance

1. **Count:** 10 distinct targeted clinical questions. **12 original source PMIDs**: 22797645, 16391217, 8602180, 15572756, 15769967, 17050889, 8099640, 17179056, 17303791, 21824950, 10929163, 31980506. **11 independent randomized participant cohorts**, because the two CARET articles are original-randomization and its later follow-up and **must not** be treated as independent trials. Zinc 2000 and 2020 are independent.
2. **Round-level topic deduplication:** Compared dossier headings in Deep Enrichments 10–14 on the default branch; these ten target questions are distinct. However, an ingredient appearing in earlier rounds for a **different indication** is not a novel compound, and a topic-level distinction is not proof of canonical identifier novelty. E.g. earlier DE11 beta-carotene/AREDS2 addresses retinal disease and differs from the CARET smoking/asbestos harm study.
3. **GitHub code-search screening:** PMID 31980506 returned existing research-index/reconciliation matches. Other selected exact PMIDs did not surface in the limited GitHub default-branch code-search results; **absence of a search match is not evidence of novelty**. Full primary-ledger source-key joins have not been executed, so **net-new canonical citations = unestablished**, never “12 new sources.”
4. **Null outcomes preserved:** Milk thistle ALT/HCV/QoL; L-arginine ventricular function; vitamin E cardiovascular/cancer prevention; DHEA physiological performance; evening primrose eczema; black cohosh hot flashes; chromium HbA1c; soy bone density; later zinc recovery rate. Non-significance is not proof of zero effect, and positive within-arm changes are not placebo-adjusted benefits.
5. **Safety signal proportionality:** VINTAGE MI mortality comparison is concerning and trial-specific; CARET hazards are for a **combined intervention** in a high-risk cohort; HOPE-TOO heart failure is an event outcome in a defined vascular-risk group. Do not convert these into universal clinical event rates for retail dietary products.
6. **No review inflation:** CARET follow-up is same trial; multibotanical/soy arms do not show effects of each ingredient individually; DHEA male testosterone arm is a different hormone/intervention. Reuse matched entities/formulations, exclude duplicates from any unique-study totals, and keep both positive and null studies in contradiction records.
7. **Errata, retractions, and freshness:** Evening primrose record has a 1993 linked erratum. This pass reviewed original abstracts and publisher originals but did **not** certify exhaustive errata, retraction, full-text risk-of-bias, or new-trial surveillance across every publication. Independent reviewer to run those checks.
8. **Downstream action queue:** P0 reviews of (i) recent MI / L-arginine, (ii) smoker/asbestos beta-carotene+retinyl-palmitate, (iii) vitamin E in vascular-risk users, and (iv) HCV treatment claims for milk thistle. P1 reviews of ingredient-specific efficacy/identity and contradictory zinc findings. **These are potential editorial priorities, not confirmed defects on live pages.**
9. **No release permission:** This file alone changes no canonical source ledger, clinical safety content, medical recommendation, source tier, consent UI, product/affiliate placement, SEO or site deployment. Require article-level existing-claim audit, independent evidence signoff, exact canonical PMID+DOI join, CI verification of exact head and explicit authorized publication before any future promotion.
10. **ROI status:** research-only asset; no measured traffic, AI citation, conversion, revenue or SEO improvement attributable to this pass.