Sqwod Evidence / Methodology

How we grade the evidence

Every Sqwod Evidence briefing follows one rule: every claim tied to a named, linked study — and graded for strength. Where evidence is strong we say so; where it’s thin, we say that too.

Evidence grades

Strong

Backed by multiple meta-analyses or randomised controlled trials pointing the same way. We state it plainly.

Moderate

Real evidence with caveats — a single meta-analysis, mixed effect sizes, or moderate GRADE certainty. We flag the limits.

Early

Promising but unsettled — small studies, animal or preclinical work, or open questions in the trial registry. Framed as “being studied,” never as a promise.

Source tiers

Tier 1 — cite freely

Meta-analyses and systematic reviews of randomised controlled trials, and large RCTs. The strongest human evidence.

Tier 2 — cite with context

Individual RCTs, observational and cohort studies. Useful, but we name the study design and its limits.

Tier 3 — signal, not proof

Preprints, animal/in-vitro work, and the clinical trial registry (ClinicalTrials.gov). We use these to show what’s coming — clearly labelled as not yet proven.

A real name stands behind it. Every briefing is reviewed for accuracy before publishing and carries a byline. On health topics, a source nobody stands behind is worse than none.

Not medical advice. Sqwod Evidence is coaching information, not medical advice. Anything involving a medical condition, pregnancy, or medication belongs with a qualified professional.

We correct in public. Evidence moves. When new studies land, we update the briefing and log it in the changelog. Spotted an error? Corrections →