Editorial Standards

A record begins as research, not as a conclusion. An author gathers the relevant human trials, systematic reviews, and meta-analyses, then drafts the record around the questions readers actually ask — Does it work? Is it safe? What’s still unknown? — rather than around a predetermined verdict.

The draft then passes through an editorial workflow with distinct review stages: research is completed, then a scientific review checks the claims and calibration, then a medical review examines safety framing, and only then is a record cleared for publication. A record is never auto-published; each stage is an explicit, logged sign-off.

BioSignal evaluates the specific, popular claims people encounter — not vague impressions. Each claim receives a verdict drawn from a fixed set: Supported, Mixed, Not Established, or Insufficient — paired with its own confidence level. This keeps a strong claim and a weak claim about the same compound clearly distinct, so a single headline can’t stand in for the whole picture.

Not all evidence is equal. BioSignal weighs each finding by the strength of its design and the consistency of results across independent studies. Randomized human trials and rigorous systematic reviews carry the most weight; mechanistic reasoning and animal studies inform plausibility but cannot, on their own, establish a human effect. The full hierarchy is described in the Methodology.

Confidence reflects how strong and consistent the underlying evidence is — not how large or exciting the effect would be. A well-replicated finding earns High confidence even if the effect is modest; a dramatic claim resting on one small study does not. The four levels, and how to read them, are defined on the Confidence Levels page.

When studies disagree, BioSignal does not pick a winner to appear decisive. It reports the disagreement, weighs the studies by quality and design, and lowers confidence accordingly. A record can and does say that the evidence is genuinely mixed — that is an honest finding, not a failure to conclude.

Records are living documents. A record is revised when new trials or reviews shift the balance of evidence, when a verdict or confidence level changes, when safety information is added, or when an error is corrected. Every change is versioned and logged — see Version History.

These commitments are non-negotiable and apply to every record:

• No sponsored rankings. Records are never ordered or shaded to favor a sponsor.
• No affiliate influence. BioSignal earns nothing from what you choose to take.
• No paid placement. No company can pay to appear, rank higher, or read more favorably.
• Unknowns are published. Open questions are stated plainly, not hidden to seem authoritative.
• Negative evidence is included. Null and contradicting results are reported, not omitted.
• Limitations are disclosed. Where the evidence is thin, indirect, or narrow, we say so.