Wellness Radar Subscribe
The Evidence Radar

Every claim, graded.

Most wellness writing dodges the question of evidence. We don't. Every claim on Wellness Radar receives one of five grades based on the published research behind it. The rubric is below. It applies to drugs, supplements, devices, protocols, and the framings we use to talk about them.

STRONG

Replicated in well-designed human trials

The claim is supported by either two or more high-quality randomized controlled trials with adequate sample size (n > 50 each), a meta-analysis of RCTs with consistent effect, or a Phase 3 trial that led to regulatory approval. The effect has been reproduced across independent labs and funding sources. The mechanism is understood.

  • Two or more independent RCTs with n > 50, OR
  • Meta-analysis of RCTs with consistent direction and magnitude, OR
  • Phase 3 + regulatory approval for the specific indication
  • Replicated across independent funders

Examples: Semaglutide for weight loss · Statins for LDL reduction · CPAP for moderate-to-severe sleep apnea

MODERATE

One good trial, or multiple consistent cohorts

The claim has real human evidence but isn't yet definitive. Typically one well-designed RCT plus supporting observational data, or multiple consistent cohort studies pointing the same direction with a plausible mechanism. For drugs, a Phase 2 RCT that hasn't yet received FDA approval for that indication.

  • One well-designed RCT + supporting observational data, OR
  • Multiple cohort studies (n > 1,000 combined) with consistent effect, OR
  • Phase 2 RCT showing meaningful effect, awaiting Phase 3
  • Some replication, but not yet across independent groups

Examples: Magnesium glycinate for sleep onset · Berberine for fasting glucose · Creatine for cognitive performance under stress

EMERGING

Early human signal, mechanism plausible

The claim has preliminary human evidence and a defensible biological rationale, but the trials are small (n < 50) or single-site, the data are unreplicated, or active Phase 1/2 work is still underway. Promising but unproven.

  • Pilot human trials with small sample sizes (n < 50), OR
  • Strong animal/mechanistic data plus at least one small human study, OR
  • Active Phase 1/2 trials with preliminary readouts
  • No independent replication yet

Examples: Urolithin A for muscle endurance · Rapamycin for human longevity · BPC-157 for tendon healing

WEAK

Animal or mechanism only, or flawed human data

The claim rests on in-vitro or animal data, theoretical mechanism, or human studies with serious design problems — open-label trials, tiny samples, no control group, or industry-funded studies that haven't been independently reproduced. The signal might be real; we don't know yet.

  • Only in-vitro or animal evidence, OR
  • Only theoretical mechanism without human data, OR
  • Human studies are open-label, n < 10, no control, or unreplicated industry-funded
  • No replication

Examples: Most nootropic stacks · Most peptide bioregulator claims · Most "anti-aging" supplements

HYPE

Marketing claim with no real evidence

The claim is asserted without peer-reviewed support, references "studies" without naming them, originates from influencer or sales material, or treats anecdote and survey data as evidence. This grade is not an accusation that the product doesn't work — only that the public claim isn't backed by what the source implies.

  • No peer-reviewed support, OR
  • "Studies show" without identifying the studies, OR
  • Influencer / sales claim with no underlying paper, OR
  • Anecdote, testimonial, or marketing survey presented as evidence

Examples: Most celebrity-endorsed supplements · "Boosts immunity" claims · "Clinically proven" without a citation

What the grade is not.

A grade is not a recommendation. STRONG means the effect exists; it doesn't mean the intervention is right for you. HYPE means the public claim isn't supported; it doesn't always mean the product is useless — just that the marketing has run ahead of the data.

Grades apply to specific claims, not whole products or articles. A single peptide might carry a STRONG grade for one effect, EMERGING for another, and HYPE for a third. We grade each claim on its own evidence.

How we assign grades.

Editorial reads the cited studies behind every claim before publication. We weight by study design (RCT > cohort > case-control > animal > in-vitro), sample size, replication across independent groups, and conflict of interest. Where the evidence is mixed, we err toward the lower grade.

Every Evidence Card on the site links back to this page so the rubric is auditable. If we get something wrong, you can compare our grade to our published criteria.

What changes a grade.

Grades update when the underlying evidence changes — a new RCT replicates an EMERGING effect, a Phase 3 trial fails, a paper gets retracted. When a grade moves we note it in the article changelog. Grades are not promotional content; they move down as readily as they move up.

Conflict-of-interest disclosure: Wellness Radar does not currently carry paid sponsorships, affiliate links to graded products, or vendor partnerships. If that changes, it will be disclosed on every affected article and on this page.

Limitations we own.

This rubric is a tool, not a verdict. Evidence in human biology is messy, populations differ, individual response varies, and good studies sometimes fail to replicate. We grade what's been published — we cannot grade what hasn't been tested. Many useful interventions sit at WEAK or EMERGING for years before reaching MODERATE or STRONG. Absence of evidence is not evidence of absence.

We are not a medical authority. We are a publication that reads the studies behind health claims and tells you what they actually show. Talk to a clinician before changing anything that matters.