READ RESEARCH BETTER

Evidence is a ladder

05Regulatory review and ongoing surveillance
04Large controlled human trials
03Early human studies
02Animal studies
01Cells, models and mechanisms

Promising is not the same as proven.

A biological mechanism may be interesting without demonstrating that a product is safe or effective in people. Ask where the evidence sits, how large the study was, what it measured, and whether the product studied matches the product being sold.

  • Is the compound FDA approved for this exact use?
  • Is the evidence human, animal, or laboratory based?
  • Was there a comparison group and meaningful follow-up?
  • Are the source, identity, sterility, and purity independently verifiable?

For educational purposes only. Not medical advice, diagnosis, treatment, or prescribing guidance. Research and regulatory information can change over time.

WHERE NUMBERS COME FROM

Not every dose
has the same backing

Protocol charts often mix FDA-labeled doses with numbers repeated online, and present both the same way. Before trusting a number, ask where it came from.

01

FDA label

Tested in controlled human trials and reviewed by the FDA for a specific product, use, and population. Example: the semaglutide and tirzepatide escalation schedules.

02

Published human trial

Doses studied in a registered, peer-reviewed trial, but not approved. Example: retatrutide 1–12 mg and eloralintide 1–9 mg weekly in Phase 2.

03

Approved outside the U.S.

Labeled in another country, such as thymalfasin (thymosin alpha-1) or Semax. Useful context, but not a U.S. approval and not always based on large trials.

04

Community convention

Ranges repeated on forums, protocol cards, and seller pages without a controlled human trial behind them. Repetition is not evidence. Pep Guide does not publish these.

How to spot a convention number

  • No citation, or phrases like “commonly used” or “typical protocol”
  • Rigid cycles (such as “5 on / 2 off”) with no trial behind them
  • The same range repeated across many seller sites
  • Doses built around vial sizes instead of studies

Two numbers from one trial

Large trials often report two results. The treatment-policy number counts everyone, including people who stopped. The efficacy number counts only people who stayed on treatment, so it is higher. CagriSema’s 20.4% and 22.7% come from the same trial.

Averages hide ranges

“20% average weight loss” means some people lost much more and some much less. Trial results describe groups of screened participants under monitoring, not what will happen to any one person.

Next: reading labels