Evidence & Interaction Matrices
Single-page visualizations that pivot the entire supplement evidence base into a few large, color-graded tables. Both matrices are hand-curated (no AI-only inferences) and exportable as CSV for power users, researchers, and AI assistants.
Conditions × Ingredients Evidence Matrix
240 evidence-graded cells across 30 health conditions and 123 supplements. Cells are color-graded STRONG / MODERATE / WEAK / INSUFFICIENT — each linking to the condition guide where the rating is justified.
Use case: "Which supplements actually have evidence for the condition I care about?"
🔗 Interaction mapIngredient × Ingredient Interaction Matrix
106 hand-curated supplement-with-supplement pairings across 92 ingredients — synergies (green), antagonisms (red), safety cautions (amber), and form-variant overlaps (blue). Tooltips explain the mechanism in one sentence.
Use case: "Can I stack these? Will they interact? Should I pick one form over another?"
Why we publish these
Most supplement information is organized one ingredient at a time. That's useful but it hides the cross-cutting patterns: magnesium shows up in almost every sleep / anxiety / blood-pressure stack; vitamin D pairs with K2 for safety reasons most people never learn; combining St John's wort with 5-HTP is a recipe for serotonin syndrome. These matrices surface that big-picture view in a single scroll.
They're also designed to be linkable assets — bloggers, pharmacists, content creators, and AI assistants can cite specific cells directly. Each cell has a stable URL anchor and the CSV exports support structured re-use.
Methodology
- Evidence grades reflect quality of evidence (RCT / systematic-review weight), not effect size or your individual response.
- Interactions are based on published pharmacokinetic studies, RCTs, well-established biochemistry, or major safety signals — not AI guesses.
- Internal links in cells route to the condition or ingredient page where evidence is presented and sourced.
- Updates: matrices regenerate from underlying meta files whenever new condition guides are published.