What are our claim rules?
- Match the source to the sentence. A citation must directly support the nearby claim.
- Preserve the studied scope. Category-level tea research is not relabeled as proof for Sencha, Earl Grey, or another style.
- Separate beverages and extracts. Concentrated green-tea supplements are not treated as ordinary brewed tea.
- Avoid false precision. We do not calculate exact caffeine, oxalate, EGCG, pH, or personal medical risk without valid measurements and a defensible model.
- No treatment language. We avoid “cures,” “detoxes,” “stops,” “safest,” and guaranteed outcomes.
What evidence do we prioritize?
We prefer current government health guidance, systematic reviews and meta-analyses, then well-designed controlled human studies. Observational research, laboratory work, animal studies, and traditional use may provide context but are labeled and cannot establish clinical efficacy.
How do we handle safety?
Safety notes are general and deliberately conservative. Pregnancy, childhood, liver or kidney disease, allergies, surgery, and medicine use can change what is appropriate. Individual advice belongs with a clinician or pharmacist who knows the person and product.
What is the calculator policy?
Every calculator must perform a useful, reproducible task and explain its inputs, units, assumptions, limitations, related tools, and relevant equipment. A tool is removed when its output cannot be defended.
How are corrections handled?
When a claim or link is found to be wrong, we remove or narrow the claim, replace the citation only when a directly relevant source exists, and update the review date. Sanity stores field-level review dates and supports a future public correction log.
How is commercial content separated?
Affiliate links are labeled. Product availability, rating, and price are not represented as live unless a compliant data feed supplies them. Affiliate relationships do not increase a health claim’s evidence grade.