¶ Evidence Grading & Editorial Policy

¶ What We Mean by “Evidence”

When we assess a claim (“X improves sleep” / “Y lowers LDL‑C”), we try to map it to:

• A population (who?)
• An endpoint (what is measured?)
• A dose/parameter (how much/how often?)
• A timeline (when should it change?)

¶ Evidence Hierarchy (Practical)

We prioritize evidence in roughly this order:

1. Human randomized trials and meta-analyses on the endpoint of interest
2. Human observational evidence (hypothesis-generating; confounded)
3. Human mechanistic studies (biomarkers, physiology)
4. Animal/in‑vitro (useful for mechanisms; weak for real-world effects)

For formal frameworks that describe evidence levels and certainty, see the Oxford Centre for Evidence‑Based Medicine levels and the GRADE approach.^[1][2]

¶ Our Page-Level “Evidence Signal”

Many pages include an evidence badge (e.g., “Promising”, “Validated”). It’s a simplified summary that combines:

• Quality of evidence (risk of bias, consistency, precision, directness)
• Magnitude and relevance of effect (clinically meaningful vs tiny)
• Safety and interaction risk (especially for self-experimentation)

Important: a high evidence signal does not mean “safe for everyone” or “appropriate for self-experimentation.” Safety depends on dose, context, and the person.

¶ Editorial Workflow (How Pages Are Built)

1. Define the claim and the measurable endpoints.
2. Collect best-available human evidence (prioritize systematic reviews/meta-analyses; then RCTs).
3. Summarize effects using a human outcomes table when possible.
4. Safety review:
   • contraindications and caution populations
   • supplement–drug and supplement–supplement interactions
   • “don’t self-experiment” triggers
5. Mechanism summary (plain language) — supportive, not primary.
6. Tracking plan: how to know if it’s working (metrics + cadence).
7. Update discipline: add a changelog entry when evidence or safety meaningfully changes.

¶ Recency and Updating

Biohacking content becomes outdated quickly due to:

• New RCTs/meta-analyses
• Re-analyses of older evidence
• Safety signals (e.g., interaction reports, side-effect patterns)

When you see “What’s new” updates, they’re logged here: Biohacking: What’s New.

¶ Epigenetic Clocks and “Biological Age” Claims

We treat biological age tests as tools with limitations, not definitive measures. When we reference clocks, we prefer methods with published validation and clear endpoints (e.g., pace-of-aging vs mortality prediction). Examples include DunedinPACE and GrimAge.^[3][4]