—Methodology
How Luneri builds what it shows
Every result on Luneri comes from published human studies. Here is how they are found, read, graded and shown, and where the method has limits.
01Where the studies come from
For each supplement and practice, a research pipeline searches PubMed and OpenAlex, follows the citations of the main reviews and trials, and checks trial registries for studies that were never published.
It keeps human studies that actually gave the supplement or practice. A study that used it only as a comparator, or for something else such as bowel preparation, is not counted as evidence for it.
02How results are read
A model pipeline reads each study and records what it found for each marker: which way the marker moved, by how much when the study says, in which people, and with what kind of study. Every result links to the studies behind it.
Pages are built by code from that data. Nothing is written by a model when you open a page.
03The four evidence states
Supported: human studies found a change. No change: studies looked and found none, which is a real result. Conflicting: studies disagree, or the direction depends on who was studied. No result yet: no study in the dossier reported it. Where we searched, the page says why (no trial found, only in combination, not yet reviewed). It is a gap, not a negative result.
04Grades
Grades run from A to D and describe the studies behind a result: meta-analyses and large randomized trials at the top, small or uncontrolled studies at the bottom. A grade describes the strength of the evidence, not the size of the effect.
05What is left out, and why
Treatment instructions found in the source material are never shown. Graded results that link no study are marked unconfirmed. Studies whose own text says they did not test the supplement are kept out of the strongest-papers list.
06How claim check decides
Claim check finds the supplement and the effect in your claim, looks up every study-backed result for the matching markers, and compares the direction the studies measured with the direction you claimed.
07Limits
Extraction can be wrong, so every result links its studies for you to check. Studies differ in dose, duration and population, so a result describes what happened in the people studied, not what will happen to you. Luneri is research, not medical advice.