Literature cutoff: 9 October 2026. Reconciliation completed 11 October 2026. Consensus Pro browser output is a discovery/synthesis aid; original source sections determine attribution. Provider meters are not evidence probabilities or examined-primary counts.
Original DIVAS abstract reports beneficial lipid changes but a null primary vascular-function comparison. A projected mortality reduction is a risk calculation, not observed trial deaths. Original primary source: PMID 26016869. DOI: 10.3945/ajcn.114.097089. Access: abstract_metadata.
Original primary abstract describes a nonrandomized sequential dietary design. Average LDL/apoB reduction is informative but does not become a randomized event-endpoint trial. Original primary source: PMID 39111551. DOI: 10.1016/j.ajcnut.2024.07.032. Access: abstract_metadata.
Original RISSCI-2 abstract uses participants selected on prior response and does not establish absorption as the explanation. It is not an independent population replication of RISSCI-1. Original primary source: PMID 42664559. DOI: 10.1016/j.atherosclerosis.2026.121884. Access: abstract_metadata.
Original Minnesota reanalysis body separates cholesterol lowering from its clinical mortality result and incomplete historical follow-up. The linked 2024 correction repairs a thesis link; it does not reverse the results. Original primary source: PMID 27071971. DOI: 10.1136/bmj.i1246. Access: full_text_partial.
Original Sydney body reports unmeasured trans-fat exposure and other dietary uncertainties. They cannot be discarded as proven trans-fat contamination. Original primary source: PMID 23386268. DOI: 10.1136/bmj.e8707. Access: full_text_partial.
Consensus was used to identify literature and cross-check specific claims about study design, endpoints, scope and research gaps. Those claims were checked against original primary abstracts/status records and, for selected studies, original Methods or Results passages. Consensus output was not treated as original full text. Source-specific access depth and receipts are retained. Scite supplied one five-anchor citation-context/tally query; returned contexts were outgoing references, not an audited incoming contradiction set. SciSpace supplied one discovery pass; it added no primary-count credit.