What we are trying to do
Detox Database translates claims into questions that can be checked. The goal is not to reflexively promote or dismiss a practice. The goal is to show what is known, how it is known, what remains uncertain, and what could matter for safety.
Source priority
When available, entries prioritize:
- Government and public-health materials.
- Systematic reviews and evidence-based clinical guidance.
- Peer-reviewed human research.
- Foundational or mechanistic research, clearly labeled.
- Manufacturer materials for composition, instructions, and the manufacturer’s own claims—not as independent proof of effectiveness.
Evidence labels
- Supported: multiple relevant sources converge and important limitations do not overturn the practical conclusion.
- Promising but limited: early or incomplete evidence suggests a possibility but does not justify certainty.
- Mixed: credible sources or results disagree, or effects vary materially by context.
- Indirect: the source addresses a mechanism, population, dose, or outcome different from the claim being evaluated.
- Insufficient: available information cannot answer the practical question reliably.
- Not established: the claimed outcome has not been demonstrated by the evidence reviewed.
Editorial separation
An entry should visibly separate the claim, the evidence, the limits, and the safety context. Sources are linked where possible. Material corrections should be dated rather than silently rewritten.
Review cycle
Each entry should show its last substantive review date. High-interest or safety-sensitive entries should be revisited first when credible new evidence, regulatory action, or correction requests arrive.