How to read this wiki
The site deliberately separates what the supplied corpus supports from the organization added to make the corpus navigable.
Three provenance layers
Decision-oriented research questions
- What is the relevant reference class and base rate?
- Is the reported relationship predictive or causal?
- Is the evidence between people, within the same person over time, or both?
- How large is heterogeneity around the mean?
- Could selection, collider bias, reverse causation, or measurement differences generate the pattern?
- Does the result transport to this population, institution, age, ancestry, environment, or severity level?
- Does changing the proposed cause change the patient- or person-important outcome?
- Does efficacy survive implementation, adherence, and environmental friction?
Corpus boundary
The current source corpus consists of the project’s existing category/term clusters, the supplied external URLs embedded in those research notes, and two Apple Notes web archives containing the same research prompt. The archives contribute the research objective but no additional citations. Future ingestion should append evidence rather than silently rewriting the canonical research inputs.
Editorial maintenance
Stable term URLs and heading IDs are generated from canonical term names. New sources should be attached to existing terms where possible; genuinely distinct concepts should receive their own entries and reciprocal related-term links. Contested constructs should retain explicit status notes rather than being normalized into settled terminology.