Methodology

This page exists because "trust me" isn’t a source. Here’s exactly how a claim gets from "someone said this on the internet" to a published verdict on this site.

How topics get chosen

Two triggers: either a claim keeps showing up in searches and conversations (protein timing, "does gut health affect mood," mushroom coffee), or a genuinely relevant new study or meta-analysis publishes and is worth a plain-language summary. I don’t cover a trend just because it’s viral — if there’s no research angle worth summarising, it doesn’t get a page here; it might get a short myth entry instead.

The source hierarchy

Not all evidence is equal, and pretending otherwise is how bad nutrition advice spreads. In order of how much weight a claim gets:

  1. Meta-analyses and systematic reviews — the closest thing to a settled answer.
  2. Multiple RCTs pointing the same direction — strong, if not yet synthesised.
  3. A single RCT — useful, but treated as preliminary, not settled.
  4. Observational data — good for spotting patterns, never enough alone to claim "X causes Y."
  5. Animal or in-vitro studies — always flagged as not yet confirmed in humans.
  6. Expert consensus or official guidelines (NHMRC, WHO) — practical, even without one definitive trial behind them.
  7. Insufficient evidence — a real, legitimate verdict. "We don’t know yet" is an answer, not a cop-out.

How verdicts get assigned

Every myth-vs-fact entry gets one of four verdicts: Fact,Myth, Mixed Evidence, or It Depends. Mixed Evidence and It Depends aren’t an escape hatch for claims I don’t want to call — they’re reserved for claims where the literature is genuinely split or the answer actually depends on a specific variable (population, dose, context), and I say plainly what that variable is. A claim that leans one way but isn’t airtight still gets a Myth or Fact verdict, just with an honestly lower evidence-quality tag attached.

Fact-checking is ongoing, not one-time

Every published piece carries a "Last fact-checked" date. Evergreen deep-dives get revisited roughly every 6-12 months to confirm citations still resolve, check whether a newer meta-analysis has superseded an older single study, and update the verdict if the evidence base has actually moved.

Found an error?

Tell me. Genuinely — a citation-driven site is only as good as its willingness to be corrected. Get in touch via the contact page.