How we research
We build every health page from a clear source hierarchy: official guidelines and government health bodies first, then peer-reviewed trials and meta-analyses with visible DOIs and PMIDs, then reputable clinical explainers. We cite real sources you can check, distinguish established physiology from promising-but-early findings, and update pages as the evidence changes.
The source hierarchy
Not all sources are equal, so we rank them. Official guidelines and government health bodies — the NIH and NIDDK, the FDA, national guideline groups — come first. Next come peer-reviewed randomized trials and meta-analyses, cited with a DOI and a PMID so you can open the original. Reputable clinical explainers come last and are used for plain-language framing, never as the basis for a claim.
How we cite
Every citation points to something real and resolvable. We list the study, the journal, the year, and the identifiers. We do not fabricate PMIDs, invent statistics, or dress up a press release as a clinical trial. If we can’t verify a source, the claim gets softened or cut.
How we grade evidence
We keep three buckets distinct:
| Strength | What it means | Example |
|---|---|---|
| Established | Textbook physiology | Carbs raise glucose; insulin drives storage |
| Well supported | Backed by multiple trials | Low-carb periods shift the body toward fat-burning |
| Promising | Single study or early signal | Carb restriction can mimic fasting’s metabolic effects |
When the honest label is “promising,” we use it. Carb fasting as a named protocol has less head-to-head trial evidence than standard low-carb diets or intermittent fasting, so some of what we say is reasoned from adjacent research — and we tell you when that’s the case.
We update
The evidence moves, so the pages move with it. Each page shows when it was last reviewed. See our full editorial policy for review, authorship, and disclosure.
Frequently asked questions
- What counts as a good source here?
- In order: official guidelines and government health bodies (NIH/NIDDK, FDA, and national guideline groups); peer-reviewed randomized trials and meta-analyses with a DOI and PMID; then reputable clinical explainers. We lead claims with the first two and use the third only for readability.
- How do you decide if the evidence is strong?
- We separate three things: textbook physiology (well established), findings supported by multiple trials (well supported), and single-study or early results (promising). We label which is which on the page, rather than flattening everything into 'studies show.'
- Do you ever say 'we don't know'?
- Yes. Carb fasting as a branded protocol has less direct trial evidence than standard low-carb or intermittent fasting, so some claims are extrapolated from adjacent research. We say so instead of implying a bespoke protocol has been proven on its own.