Peptide evidence is easy to flatten into a verdict: “works,” “does not work,” or “approved.” The atlas is structured to resist that shortcut. Its grades belong to specific claims, its status records belong to jurisdictions and products, and its dates tell you when a fact was last checked—not how long it will remain true.
This guide explains how to read those layers together. It is about evidence literacy, not personal treatment decisions. The underlying rules are documented in Evidence grading and Scope and selection.
Start with the claim, not the molecule
The atlas grades an indication-specific claim rather than assigning one permanent score to a peptide. That distinction matters because the same molecule can have strong evidence for one use and weak evidence for another. A current authorization also covers a defined product, population, indication, and jurisdiction; it is not a universal certificate for every claim made about the active substance.
The semaglutide monograph shows the pattern. Its table assigns grade A to several authorized uses supported by controlled trials and regulatory review, while a use granted through an accelerated pathway is graded B because confirmation is still required. The molecule did not change between rows. The claim, endpoint, and regulatory context did.
Before reading a grade, identify the exact row: What condition or outcome is claimed? Which population was studied? Is the endpoint a patient benefit, a symptom score, a laboratory measure, or an imaging surrogate? Evidence for a biomarker should not be silently promoted into evidence that people feel better, function better, or live longer.
What the A–E grades mean
The grading method defines five main confidence levels:
A — established for a specific labeled use. Current authorization is paired with adequate controlled trials and post-market context. The label boundary remains part of the grade.
B — moderate human evidence. Multiple controlled studies or a strong pivotal study support the claim, but the claim does not have current authorization or still carries a material confirmation gap.
C — preliminary human evidence. Human observations exist, but they may be small, uncontrolled, early phase, or focused on surrogate outcomes.
D — preclinical only. Laboratory or animal findings exist without adequate human efficacy evidence for the claim.
E — anecdotal or marketing claim. Testimonials, extrapolation, or commercial assertions are not backed by adequate scientific support.
The method also uses X when adequate evidence contradicts a claim, a development program failed, or the claim does not match the studied material. X is not “even weaker than E.” It signals affirmative conflict rather than mere absence of trustworthy support.
A lower grade is not proof that an effect is impossible. It says the available evidence cannot justify a stronger conclusion. Likewise, an A does not mean risk-free, appropriate for everyone, or established outside the labeled context.
Why “research market” does not mean approved
The atlas includes approved active substances, clinical-stage candidates, endogenous or laboratory peptides, and products visible in the commercial research ecosystem. As the scope policy states, inclusion reflects scientific, clinical, regulatory, or public visibility—not endorsement.
“Research use only” is a marketing designation, not a regulator’s finding that a product is safe, effective, authentic, pure, or sterile. A listing can establish that a name is being marketed. It cannot establish that the material matches a clinical-trial product or that any human-use claim has passed regulatory review.
The BPC-157 monograph makes these layers visible. Its regulatory table records no identified FDA-approved or EMA-authorized product, while its claim table separates preliminary human reports from claims supported only by preclinical work. Broad online visibility therefore does not raise the evidence grade or change authorization status.
Identity is another reason for caution. Marketed names can refer to fragments, salts, complexes, mixtures, or poorly standardized material. Even a credible study of a well-characterized investigational product does not automatically validate material sold elsewhere under the same name.
How to read an Evidence-by-claim table
Read each row from left to right as a compact argument.
Claim or indication: This is the proposition being graded. Do not substitute a broader claim.
Stage: This describes development or authorization status, which is related to—but distinct from—confidence.
Grade: This is the atlas’s conclusion about the support for that exact claim.
Best human evidence: Check the design, participant count, comparator, and whether the cited record has results.
Main result: Identify the measured endpoint and the size and direction of the finding.
Important limitations: Look for short follow-up, missing controls, selective populations, surrogate endpoints, attrition, sponsor dependence, or unpublished results.
Then test whether the pieces agree. A large trial registry entry with no posted results cannot support an efficacy conclusion. Statistical significance does not guarantee clinical importance. An uncontrolled report can generate a hypothesis without resolving placebo effects, regression to the mean, selection bias, or natural recovery.
The retatrutide monograph offers a useful comparison. Its rows can carry grade B human evidence while the status table still says no marketing authorization was identified. Strong development-stage findings and regulatory approval are different facts. Read both tables before summarizing the entry.
What verification dates do—and do not—mean
Approval, trial, legal, and sport-status claims change. The atlas attaches an as_of, last_verified, or equivalent date to time-sensitive records. That date means the named sources or databases were checked through that point. It is a timestamp on the review, not a promise that nothing changed the next day.
Dates also do not make every source equally strong. A recently checked commercial assertion remains a commercial assertion. An older pivotal trial may remain decisive if its publication is valid and the endpoint is relevant. Recency and evidentiary weight answer different questions.
When a decision depends on current status, follow the linked primary record and recheck it. Look for label revisions, newly posted trial results, safety communications, withdrawals, registry changes, and jurisdiction-specific rules. The atlas’s refresh policy explicitly treats these claims as time-sensitive.
A compact reading routine
Use four questions whenever a peptide claim looks persuasive:
What exact claim is being made?
What human evidence directly tests that claim?
What is the current product- and jurisdiction-specific status?
When were those facts last verified, and what primary record supports them?
If any answer is missing, preserve the uncertainty. The goal is not to turn every entry into a yes-or-no verdict. It is to keep identity, evidence, regulation, safety, and time aligned so that a claim never becomes more certain in retelling than it was in the source record.