Where the claims come from
Every factual claim on this site links to a source: a randomized trial, a meta-analysis, a cohort study, a position stand, a guideline, or, for what Rhonda Patrick and her guests have said, the podcast episode itself. The numbered links on each page go to a reference list at the bottom that says what the study found and what it can't show. All the sources are also on one page.
How every source gets checked
Each paper was checked against its official record before the page went up. A script reads its PubMed record (or its Crossref record when there's no PubMed entry) and confirms the paper exists and that the title, year, first author, and DOI match. It then looks for every number quoted from the paper in the paper's own abstract, and any number it can't find there gets checked by hand against the full text or rewritten. Podcasts and web pages are checked by opening them.
I research these pages with AI research assistants, and that's exactly why every citation goes through that mechanical check, and a second, independent read of every number against its source, before anything is published.
The four grades
- Strong
Several randomized trials or meta-analyses agree. - Moderate
One good randomized trial, or large cohort studies that line up. - Early
Small or short trials, pilot work, or mostly animal data. - Mixed
Good studies disagree.
Each finding gets its own grade, and each page gets an overall one. The overall grade describes the evidence for the main claims, not how much I like the protocol.
What the study types can and can't tell you
- A randomized trial assigns people to a treatment or a comparison by chance, so a difference between the groups is probably caused by the treatment. Small or short trials can still mislead.
- A meta-analysis pools many studies. It's only as good as the studies it pools.
- A cohort study follows people over time and compares those who did something with those who didn't. It shows an association, not a cause: people who use a sauna four times a week may differ in a dozen other ways from people who use one once a week.
- A mechanistic or animal study shows how something might work. It doesn't show that it does in people.
Money and conflicts
Where a trial was funded by the company that sells the product, or written by people who sell it, the page says so. This site takes no money from anyone: no ads, no sponsors, no affiliate links, and no free product.
Some pages say where I buy a product, on Amazon or from the maker. Those are plain links with no affiliate or referral code, checked in a real browser to make sure each one goes to the brand's own listing.
Keeping it current
The research frontier and the news list move fastest, and each entry carries its date. Approvals, prices, and trial results change, so every page shows the day it was last reviewed.
Rhonda Patrick
Rhonda Patrick, PhD, hosts FoundMyFitness and has covered sauna, cold, exercise, creatine, NAD+, omega-3s, and sulforaphane in depth, so her work is cited wherever it applies. Her views are summarized from her published papers and from the episodes and talks linked on each page. She hasn't reviewed or endorsed this site, and neither has anyone else cited here.
Corrections
If you find a mistake, a broken link, or a study I should have included, tell me through charlessieg.com. Pages carry the date they were last reviewed.